US - Utah

[real cedar]

I want to get up that one first. Get a router up there but not all the way up.

I don't want it to be snowed out in winter or at least too inaccessible for that ones use case

Thurstons different

Im going to put that one high up

I can get up near the Adams canyon ridge

[leaden crow]

One of my friends regularly hikes out of bounds above 9990 lift at PCMR and snowboards the spine. I'm pretty certain I could talk him into maintenance on desolation if needed.

[real cedar]

You know, if he actually does do that, I might just join him because that sounds like something I would do.

Also, just had an interesting tidbit of knowledge given to me. It turns out, one of the people in the avenues that I've been trying to talk to, he's able to see my messages via a Python command line, but it's not getting transferred to the app, even though my messages are getting through from up north and his messages are getting through to me on my app. So something is going on between the app and us, but not on the actual network.

[leaden crow]

That sounds like the limitation of the API only accepting a single connection. So he has a bluetooth / mobile app and serial cli script both running? The message is likely only going to get through to one of them.
I have the same issue with the portduino devices. If both the web app and curses client are running, they contend for the same API connection and often only 1 of them gets the message.
I'm sure the devs could give you a better explanation. That is my understanding.

[real cedar]

No they were never running at the same time

We both switched over to test

Stuff always comes through 9n the Python and ncurses clients

The app is spotty

[leaden crow]

Not good. Android or ios app?

[real cedar]

Android only it seems it's limited to

New rak enclosures are nice

[vital hemlock]

@real cedar you have a busy radio. I can basically never trace to it, but it's always talking so it's at the top of my list.

Are you able to trace to mine, bp_rpi ?

[leaden crow]

lol, which one? I think he has more nodes than me now.

I keep getting crazy traceroutes from NSL to SL. They are hopping from C6C > C6H > Onaqui > NPR …

[vital hemlock]

His DC801C6 node 🙂

[leaden crow]

The hackerspace node doesn’t have LoS to anything right now. Since we put the E22 hat on it’s a little bit better. I got 2 of 17 traceroutes back from it (nsl node)

[vital hemlock]

I hear packets from DC801C6 more than once a minute, it's a busy busy node. Wondering why. It's not a problem though!

[real cedar]

People's nodes inside the building all talk through it

Cuz they cant get our of the concrete

Also it can see potm directly

And it moves almost all of the avenues and uni traffic rn

So when someone enters who's say running scripts on their node it all routes through it. I have plans for a mt Olympus router and rudys flat to fix this

Its not awful. I'll climb it after 2 more weeks of sunshine

Just need a bit more melting

So much to my own dismay @leaden crow @vital hemlock I cave to the needs of the valley first

So my node plan priority is simple
Mt Olympus (Router) > Rudys Flat (Router Late) (we should make POTM this too) > Mt Ogden Steed Mtn Benchmark (router)

Then if I have nodes and time to spare.. Leas MTN and deseret

In theory Olympus can help share the load of routing.. With NPR.. Then NPR can more or less (especially with 2.6.x) route traffic that needs to get to toele and up north more and I'm guessing city nodes will often prefer olympus

[vital hemlock]

It actually appears to be location updates. If it's stationary you might consider changing the broadcast interval 👍

[real cedar]

Its broadcast interval should be 3600.

I'll be at the space tonight to run tweaks

Gonna change its position to fixed. Broadcast once every 14400

[vital hemlock]

Yeah, it's location and neighbor info is coming every 30-45 seconds

[real cedar]

WTF you can't even set those that low

😂🤣

I'll go mess with it

Neighborinfo in firmware is literally limited to 3600 I believe

[vital hemlock]

I dont think it's always been doing this. I've only noticed it perpetually at the top of my list for the last few days

So, sometimes it makes it two minutes, but often less

They broadcast together, at exactly the same second, each time.

Four separate packets

[real cedar]

Uh oh

I wonder if someone unintentionally changed something

I'll do more to give it a good config and I'll put my public key into it for admin

[real cedar]

And pay more attention to it. I'll get it fixed. It should not be doing that. Also it should be a client

Not router

[vital hemlock]

Router is just the process I believe, it has the right client setting

No worries, just letting you know!

[real cedar]

Good. I was considering router late. It's not true that it can't see npr and potm in one hop. I often get only 2 hops back from it when I traceroute

[vital hemlock]

If yours ran into a bug the rest of us might as well

[real cedar]

Yeah I will dig into its debug logs

The thing it does have is LOS to all of the aves, the uni. Olympus and cottonwood heights

And those areas prior to it were deadzones for me

Even with LOS to NPR which suggests NPR has a loss of sensitivity

(Too much noise)

I'm still working on that LF51 filter. It will be cost Prohibitive for every node to have it but router filters it will be doable. Plus I think normal peoples nodes in the valley should just use a standard 20 to 26 mhz wide 902-928 only because I think clients should retain frequency selection rather than be perma tied to something

Routers obv aren't going to move or change for a long time so one expensive filter doesn't hurt

@vital hemlock you right its easily dealing with 15-21% traffic at times. Its probably killing the network. In the area. I'll fix that ASAP haha

[vital hemlock]

Thanks C6. I promise I wasnt trying to police the airwaves, just noticed and thought I'd ask.

[real cedar]

Nah dude it's stuff like that which I rely on to know if something happened while I remain an island up north

Its definitely fixed

Its got direct LOS to NPR and POTM. I've gotten tr back from it here in clearfield... I made it a router later to take all the nodes blinded by the city and aves and it seems to be routing quite a few directly to POTM which is good

Now I need to get something on Olympus and Under mt Ogden (steeds benchmark)

The snow pack is not very thick 3/4 the way up but it gets bad quick after that

While I would want my node high up. There's no where to hide it and that place has a ski resort on the other side

Which reminds me. If I can get something going cheaper than a rak.. I kinda wanna put some up and down the cottonwoods, first mtn, snowbasin and at dear valley 🤣

Really would be nice to have the back country in those areas covered in winter

At least then in a true emergency I could reach someone

[vital hemlock]

T114 v2, no screen? Cheap

Nrf52

Or this cutie

https://www.seeedstudio.com/XIAO-nRF52840-Wio-SX1262-Kit-for-Meshtastic-p-6400.html

Both of these options are “much” cheaper than raks

[vital hemlock]

@real cedar I bought a few of the seeed sx1262 esp’s before I knew they’d do nrf52’s

[leaden crow]

I'm still waiting on an nrf seeeeeed I ordered. I think the RAK is still worth it with the rudimentary solar charge controller, and ability to add GPS and other sensors easily. I just have one other almost complete setup that already has a separate charge controller. Heltec I am sure is fine, but I'll pay $2 more for the RAK because I know they make solid hardware.

I wonder how inexpensively we could build a solar node?
https://www.amazon.com/Soshine-6w-Performance-Monocrystalline-Cellphone/dp/B09ZL3F7CT/?th=1
https://www.amazon.com/dp/B08DY28BXW?th=1
https://www.seeedstudio.com/XIAO-nRF52840-Wio-SX1262-Kit-for-Meshtastic-p-6400.html
You'd still need enclosure, battery (BMS if not built into battery), $2 Ziisor antenna, ipex cable

https://www.printables.com/model/221500-meshtastic-6w-solar-enclosure-base-station-lora-no

[vital hemlock]

I bought the panel years ago to make this
https://www.printables.com/model/297089-v2-meshtastic-3w-solar-enclosure-base-station-lora

I need to actually do it

published in 2022

[leaden crow]

That's funny. I bought 2 soshine panels when they were on black friday either last year or the year before and haven't used them. That's why I linked his v1.

[real cedar]

The soshine 6 watt is a better panel for charging

I'd modify that in FreeCAD for it

Personally I'm making models for all my parts so I can get an idea of how small I can go

[vital hemlock]

Oh I'm not making that big bastard, I'd make something smaller too

[real cedar]

I mean the 6W really isn't bad for what you get.

https://a.co/d/eYlXhx5

If you built a case around that with a 3d printer

It would be good to survive the mountain tops

As long as you gave it say 10-20Ah for the days it's just not going to get much sun

That's kinda my plan over all. Only doing over sized panels in really remote locations like Lead Mtn or Deseret

Its not a pain for me to get my bike out and BST my way over to a trail head

It is however a PITA for me to roll two hours only to roll another hour to get to a trail head on bike 🤣

When I whip out the stl I'll print a few and see how they turn out

I want to make it so that the antenna can be mount straight above the panel at the ideal angle the panel should be facing south so you don't have to guess

Then I need a seperate anemometer mount that has the little wind generators (3 phase motors) inside. Lol
I'll probably leave a hole for plexiglass for the UV sensor

I do genuinely want all of mine to keep weather data and sync time with gps now and then so I'm putting 4x 6500 mah molicell 21700s in mine

I figure that's enough to survive the winter 🤣 without solar lmao

[vital hemlock]

I meant his gigantic box, I'd make something smaller.

[real cedar]

Oh.. I mean its sized for LTO batteries and true LTO batteries are very large cylinders. I have some I'm acquiring off an old EV. I'll show you them if successful.

So yes you could could smaller on the box

What's wild is while bigger. Usually one cell alone has 20AH in it 😂 or 10.. You don't really see smaller

Kinda like good lifepo4s. Headways are huge

[vital hemlock]

He went to the higher voltage panel to deal better with voltage drop from poor sunlight

I think

Because the rak modules sip power, and having a large bank...you really didnt need a higher wattage panel. But in my mind all paths are acceptable, just different approaches.

[real cedar]

Ahh yeah I mean I could do that. It would be more efficient but more costly. I'd spin my own power board and feed the batteries directly
I'd probably just integrate the rectification and regulation from the anemometer to that board. Have it feed as much power into the batteries and capacitor banks on both ends.. (Smooth out the draws on cold nights with a trade off for more parasitic loss)

Man now you gave me an idea. Maybe I could fit that all into one thing and make a RAK power module at the end of it

Probably a week to get that PCB dialed

The only real issue you run into when you drive the voltage up is suddenly shade starts to matter a ton more. So the lower the voltage, the more resilient you are to unideal solar conditions.

The size plays a role too, but you know, at the sizes we're talking lower voltage in general is going to help you in the shade. But the higher voltage will compensate for the voltage drop, however you have to ask the question. The solar panel is still going to produce right around 5 volts, even in winter, and even if the voltage does drop in the batteries, technically the battery voltage dropping helps you charge them if they can accept charge

[real cedar]

@vital hemlock

Its too bad we are committed to 51

14 looks very empty

[vital hemlock]

I dont think we're committed to anything with summer coming.

I know most of the antennas I've messed with are defintily better at 915 center

and our SNR is amazing, most of the time, so the noise floor isnt making things "too" bad for us

[real cedar]

Yeah that's why it was more of a "facetious" too bad..

Finding a narrow LF51 filter is actually kinda hard. 902-928 no problem but it's all custom made below 5M bandpass

https://anatechelectronics.com/products/standard/band-pass-184/912-5-mhz-cavity-band-pass-filter.html

If only I knew where to get one of these

[leaden crow]

https://www.digikey.com/en/products/detail/anatech-electronics-inc/AB915B662-S/15652000

[real cedar]

That's the big one that's wide

I looked for the model number on digikey and didn't find the specific one

It was 2.5 MHz wide

-70 db just a few hundred kilohertz outside the band

[real cedar]

Jank patched in the acasom on a 1 watt g2

Time to get more data
I've aligned the antenna and filter on the most vertical balance point they can have.. So no torsion on the connector and TREX taped it 😂

Who cringed at my double couple.. This is what happens when your testing too many things at once

[broken turret]

I cringed more at the metal window frame right adjacent to the antenna. That will detune it downwards.

If you can, get it away from anything conductive or capacitative that might muck with it.

[real cedar]

its plastic i assure you lol

[broken turret]

Am curious to know how you find that filter 🙂

[real cedar]

it has more insertion loss than the gpio labs but I can hear more stuff

[broken turret]

Aah. It looked like aluminum.

[broken turret]

Yeah, that's the trade-off. Often well worth it.

Is that the narrow one?

[real cedar]

Its a wide band

902-928

extremely sharp cutoff

[leaden crow]

Your Acasom has N connectors? For some reason I thought they were SMA? Didn’t notice it was an option anywhere

[real cedar]

Oh yeah there's an sma and an N

[real cedar]

Man the RAK solar mppt onboard just suuuuuckkkss

So neptunia stopped charging. And this is where it's at

After a week of charging just fine

[real cedar]

Figured it out. GPS hung 🙄.. I had the GPS set to only send and update its position every 86400.. Its a ublox m8 Zoe module from rak. So its got a 15uA sleep

Why on earth would it not sleep?

I really need to dig into how the firmware handles ublox modules. I've worked with them so often in the past it just doesn't seem right that it would stay in constant tracking instead of sleep+cyclic tracking if the firmware is handling it with the power-save mode ublox has

[leaden crow]

Jacob had that problem. He had the GPS modules on his RAK nodes. They don’t sleep, just drain the battery. I doesn’t really do anything for you on a static node. Just set a fixed position.

[real cedar]

That's a problem with programming from the original people, I think.

Or it's a problem in the meshtastic firmware. I don't know which one it is. I am just guessing right now. But essentially, you're supposed to be able to force these things into a sleep mode. And you're supposed to be able to define what kind of power and tracking capabilities and update rates you want. And you're also able to tweak that so you can do GPS acquisition in less than 1.5 milliamps, according to the datasheet. So my guess is that's never happening and I'm wondering if there's a way to reprogram those radios on board or if I'm going to have to make some JTAG Pogo's for reprogramming the U-Box.

Also, you can force these chips into deep sleep modes. Usually they only pull about 15 to 17 microamps when not in acquisition

I'll hit up my buddy from the last place I worked

See what we can brain storm up

They tell you exactly what to expect and they have a lot of errata of how to program them

What I'm going to do (time willing) is figured out the ublox pinout. Make myself a test RAK. Make a daughter board PCB that breaks out from the RAK ZIF module connector and then stacks a ublox on so I can break out the programming pins. Once I sort all that out I'll provide an updated ublox firmware for flashing that targets 1 Hz cyclic update low power mode. I'll do some testing to see if their assisted modes are worth the trade off but first I need to get a new rak, a power module for monitoring all the draws and an m8q

I'm guessing one of my hurdles along the way is going to be making it not jank, which means probably digging into the actual firmware of the project and seeing if they do send out the proper messages to the GPS module to shut it down.

[real cedar]

@vital hemlock the dc801c6 node stopped responding but im getting its metrics and its been slammed at 37 to 50% I hope my bot didn't blow a belt

Need southern eyes to confirm

[real cedar]

I also noticed routers only forward core port nums only :(.. Guess I'm rethinking how I send data to my drone. ATAK won't work if routers can't forward. I guess they could if you switched back to all

[real cedar]

Hehe it is accesible.. Thanks RAK!!

I also see issues on git that show that the power save mode was disabled because it interfered with rapid acquisition.. But that's not for the reasons outlined. Ublox has documentation on how to fine tune this for the need of the device and how to setup c functions to be able to select the various options

[real cedar]

This is what's in it rn btw

[vital hemlock]

I havent heard from your c6 node in a while. Hope it's alright

[real cedar]

🤣 Did running scripts kill the bot

[vital hemlock]

Oh, are you using the bbs?

Yes, only one thing can connect to the radio at a time. Python "and" the bot wont work.

Generally I just set the bot to restart every day or so, but it's because we arent relying on it right now.

[real cedar]

Wait so im using meshing around and I told it to use the python IP endpoint instead

That way it just relays stuff to meshd

[vital hemlock]

Oh, you are using this:
https://github.com/SpudGunMan/meshing-around

not this:
https://github.com/TheCommsChannel/TC2-BBS-mesh

I havent tried meshing-around

I'll give this a look

Installing now

[vital hemlock]

Okay, I like this a lot

[real cedar]

Yeah it's a good bit more useful

It has a lot of inter network relay stuff

[vital hemlock]

So I see

[real cedar]

Which you know with the right multi radio node. You could have a local short turbo and still get that data on the long or medium fast

[vital hemlock]

Oh, wheels are already turning

[real cedar]

All you need to do is setup a wrapper script for it and run it with your favorite Linux service supervisor. I've got mine running with full autorestart in s6
Lmao nice..

You got local IoT stuff going?

[vital hemlock]

too much, yes

[real cedar]

I'm thinking of making dual node femtofox's

Getting around the 7 hop limit

FINALLY!

Haha

I hate that limit so much. It's like the one thing I'll permanently dislike

I want relay links essentially between bigger meshes ran by a dedicated user. Like id love to reach where my parents are in Idaho. My friend in Cali wants to reach the Vegas one that's setup. I'd love to meet there and have another relay so I could chat with him 😆

[real cedar]

So what I can say and @broken turret you'll be happy to know this.. While I'm still frustrated at the lack of reliability of connectivity.. I can't blame that on much except me being across a salty lake with imperfect line of sight from 50-70 miles away from the router nodes..

The filters. They have done wonders. I now have gone from 7-9 of 35 online to 19-24 online up here. I can hear a lot of things I never consistently heard. I think you don't need super narrow filtering. It's ideal but just passing the pand itself is great.. I've also done some interesting optimizations. Instead of relying on client mute and stuff I often find I really like the talk when necessary behavior of a client.. So I've made client_high client_opt and client_silent profiles for myself.

Client_High leaves GPS on but has the talkiness intervals of a router.. It also has smart broadcast on enforcing minimal gps updates. It has neighborhood info on and allows metrics data, rebroadcast all

Client opt: turns off gps, uses the talkiness intervals of client mute, rebroadcasts all, has neighborhood info off, can broadcast some metrics,

Client_Silent: the talkiness intervals of client_hidden BUT rebroadcast local only, basic device telemetry only upon request, GPS off, managed mode on, bt le off, serial off, neighborhood info off.

My idea was I really liked clients ability to rebroadcast all but I wanted them to only chat when absolutely necessary in their role. And really the only reason so I wanted to quiet my local cluster down because Neptunia node in the field above a university in a tree started seeing 21% as router late and after I quieted various nodes down. Its now extremely silent up here. I've noticed I can send really quickly and receive acks better at least for now but what will this do to peoples ability to discover me if my nodes only wake up to talk if 1) there's a packet or 2) once every 24 hours (86400s).. Will this affect my ability to chat and see people adversely?

Also tall about low power usage lmao

Im receiving a shit ton of dupes though when the CH util spikes and it's coming from in the bowl not up here. My tx is only that high due to using remote admin channels

[broken turret]

Bear in mind that the early contention window (where routers live) is quite narrow, and congests easily. When you say "talkiness intervals", do you mean rebroadcast timing, or something else?

If you mean rebroadcast timing... it doesn't take many nodes in that early window to cause serious problems, especially on slower modem presets. Abuse of this is why the ROUTER_CLIENT role was removed.

Any chance of a link to the filters you're using? I'm keen to take a look at them - I get asked to recommend them from time to time, and I don't have a good go-to product that I can point people towards.

I'm really glad to hear that they've improved things for you 😁

If you're sticking with the normal client rebroadcast timing, but just tweaking how often the node will originate packets (info, telemetry, location etc) then those sound like some decent options 🙂. Most people do not optimise them, and they can get a bit overwhelming. Recent firmware versions have done a lot to improve things, and there is auto-scaling now, but there's definitely room there to lighten the load, as you are doing.

[real cedar]

Rebroadcasts info, node info, telemetry anything that it typically talks with unless I'm explicitly using it.

[real cedar]

By in large the best wideband filter for most applications is the GPIO labs 902-928 (preserves your ability to choose a frequency slot)
https://gpio.com/products/915-mhz-ism-band-pass-bandpass-filter-with-26-mhz-bandwidth

But it has a 24 dBm p1db compression point so you need a better filter if you want to do 1 watt

https://acasom.com/products/915mhz-5-types-cavity-filter-for-helium-network-filter-waterproof-lora-indoor-use-high-out-band-rejection

Acasom is decent for 1 watt but higher insertion loss

[real cedar]

What have they been currently considering?

I guess the issue is once you do set a GPS firmware on a ublox to a certain low power mode.. Its often difficult to undo that as the user. You would need to have some tools.. Sometimes an NDA but not usually

The big part is the actual onboard chip would need to be reconfigured. The Zoe m8 has an entire set of documentation in what it can be programmed to do. But the downside is once you force it to be low power and sleep.. TTFL and a few other things will take quite a long time. So you'll need multiple (3+) opportunities before a broadcast to reacquire a good lock and good locations)

Ublox on the zoe series and the Max 9 and 10 has a lot of smart modes now I believe.

[real cedar]

That's 🤣 true but why use an active antenna you don't need to if you have a clear view of the sky. I understand the benefits of a bias t powered pa+filter onboard but if it comes at the cost of power because the GPS chip sleeps but the bias tee sits and sucks like a mosquito haha.. It doesn't seem worth it since 99 times out of 100 your using GPS when outside if a bldg

But yeah Ill definitely inquire more. It's okay to forgo GPS as long as I can remote in and turn it on

I've actually thought about programming a 9 axis accelerometer to only throw a signal if it detects enough force to suggest human handling and that trigger a notification so I can turn on gps but that's a hack around

[vital hemlock]

The amount of shit going over my head in https://discord.com/channels/867578229534359593/972586878337433670 is phenomenal

They fell into board design discussion and I’m out of my depth. Enjoy @real cedar

[real cedar]

heck yeah ill take a look when im done with homework. so probably not till midnight haha

I definitely have some plans so ill look there for any hard lessons learned

[real cedar]

@broken turret @vital hemlock @maiden wraith and @leaden crow in your opinion. if someone liked frequent device metrics but just the metrics. No need for frequent other telemetry items. How often would a sane broadcast be for device and node info? once an hour?

like id ideally like to have a relatively up to date status of the battery

I guess what I'm asking is. Can you ask a router or router late to be more rude

Like can the individual change those longs 86400 intervals or is it overridden?

[leaden crow]

heh, I feel that way about https://discordapp.com/channels/867578229534359593/1302945239689990185 and https://discordapp.com/channels/867578229534359593/1008457467560214548

Meshtastic has really talented people working on both software and hardware.

[vital hemlock]

truly

@real cedar for a remote node I assume hourly would be great for long term trends. When it's new I think I'd feel nervous waiting an hour. That's why I always run things at home, in there setup, for over a week when possible so I can trust it. I'm sure you all do the same.

If it has to be configured before sending it out, not something you can edit via admin, then I'd put it where it needs to be long term and pray to the mesh gods it's fine 🙂

[leaden crow]

My opinion is that it depends entirely on the size of the mesh and channel utilization. Personally I'm fine with people playing with those intervals. If it becomes a problem specifically with overutilizing key routers, I would expect them to adjust. Kind of the same problem with hop counts. We may reach a point of utilization on LF51 where running high hop counts is infeasible.

[real cedar]

I trust nothing. I just verify in the first 10 minutes. Radio moves at the speed of light. If Im not seeing stuff after 10 minutes.. something is misconfigured

[real cedar]

Yes it does. Ive been digging into the conversation and it makes me very optimistic for its continued future. I was worried with that deal on the 915 band

[real cedar]

Where I sit.. I have a view of all but one router on our network.. so I think once an hour hopefully shouldnt flood. I saw that the stock configurations never really update more than every 30 minutes and it penalizes you to send node info out more often and stuff so what I did was node info and stuff is still rare and set to my highest interval (14400) along with other things that just dont need to be broadcast and now sensors and metrics are as often as needed..

I guess thats the advantage of living on a high point in the clearest spot south of the base up north. It certainly hasnt helped the overall distance to cover issue haha

[white field]

It like some mountains are melting out. I'm kinda tempted try the hike up to Boulter peak since it would finally enable my Skyline node to link into the mesh via POTM.

@real cedar do you have a spreadsheet you could share with a list of nodes you're considering for the link to Moab?

Assuming this link to POTM is rock solid, I think we'll be ready to place something in Henries.

I've had previously started planning out a meme link to Denver, but never figured out the continental divide.

My intent was to place a node in Indian creek.

Do you know of any options that would link easily into Moab?

[real cedar]

Thats perfect for down there. I was trying to figure that out with less nodes

Lake mtn would be perfect for most things down south that cant be served by POTM

[real cedar]

I should make something really clean huh?

do we have an agreed upon mapping tool we would like to use

It doesnt have to do LOS

[white field]

I have this one if you want to share

DM me your email and I'll add you to dit.

[real cedar]

Yes absolutely ill contribute to a community made one

one sec

[white field]

Yeah, if you can figure out how to get reliable links to/from NPR I think we can do this with far less hops.

We have been getting bad luck with NPR linking to far away nodes. (despite, but probably because it's in such a good spot)

I vaguely feel like a high power node might fix it since the problem has always been NPR rx noise.

onaqui should have 100% rock solid LOS to NPR, but it only acks a fraction of my DMs.

meanwhile, I'm getting 100% ACKs thru to POMT via NPR.

If Barton Peak can link to POMT, we'll have a 100% ack rate link to the Henries.

[real cedar]

I will do just that.. probably this evening..
Francis Peak Router who owns it?

I live literally biking distances from farmington canyon and the BST

I can try and go see if I can get to it on the weekend? if the weather stays warm
Is that tower adjacent to the radar station really abandoned.. I heard lots of electricity humming up there last summer

snowdepth at 8500 ft according to snowtel should be around 1-1.5 ft increasing from there.. I doubt its much deeper on that road and probably well packed from side by sides

Also if that person DMs me its dimensions and radio.. I might have a spare 8600 mah battery pack around to turn her back on

You can get to here before a Ford Focus sized car can go no further
https://maps.app.goo.gl/dqmj92GJcemzCunC8

its a bit muddy is all

Also theres a ridgeline trail to adams cabin
https://maps.app.goo.gl/1oXWJFD4uNGWzynMA

Above the cabin is thurstons ridgeline that jetties out farther than the rest.. if that person wanted it somewhere less accessible.. its definitely a solid place with low trail usage. Its what I was considering deploying very soon until you shared the list with me

also switch to MF51

I see the owl on Aurora Plutonis

so IDK if its dead or maybe its warmed up and gotten enough sunlight recently to pop up for air before dying again 🤣

[real cedar]

Im retarded. its hiding all the way to the right. @maiden wraith see above

If yall want to know whats in the white dome up there
its a CARSR operating on the 7172 L-Band

In the grey radome there is a ARSR-1 Long-Range Search Radar..

Theres also a Mode S IFF beacon maintained by national guard

they operate from 1215-1350 MHz

And they put out 400 kWs for the CARSR and 4MW for the ARSR1E

through their klystrons

Would it see your node.. maybe.. probably wouldnt even care tbch.. would it absolutely fucking rail your electronics if you got hit even with the sideband harmonic of that sucker.. fucking absolutely. your PA would probably be dead.. ngl. thankfully its pointed up haha

This is actually the primary area of my radio expertise
Radars, Space comms, extremely long distance and bounce communications, detection and estimation theory, encoding, phased arrays, geographically seperated multi node arrays etc.. I actually think francis peak is really cool considering all of its upgrades and how its still serving its purpose decades later without any degraded status..

its too bad I cant play with that stuff meaningfully at lower power or on my own time

[leaden crow]

It’s probably the same on Blow Hard Mountain. Which has an amazing bike trail.

I got a thing in the mail for ds nsl today.

[real cedar]

That is a WSR-88D Doppler RADAR

[real cedar]

Heck yeah.. Howd it turn out on that alpha

I find 6-8 dB is just enough gain to make the filter losses completely irrelevant

[real cedar]

Wait do you mean blow hard as in Cedar? cuz if its near cedar then yes your correct. Not quite the same setup but similar

[leaden crow]

https://www.fortwiki.com/Cedar_City_FAA_Radar_Site

https://www.mtbproject.com/trail/7005828/blowhard-trail

[real cedar]

Ahh yeah then its probably the same setup. Slightly different. It needs more receive sensitivity due to the longer range it needs to serve.. Im fairly certain that one has a unique device called an amplitron

Those were fairly common on the second variant of ther ASRSes

https://en.wikipedia.org/wiki/Crossed-field_amplifier

So I imagine if you were ever allowed inside.. just off the large feedline wave guide from the radar antenna itself.. youll find a large cylindrical vacuum tube with the wave guides attached to it

Oh fuck yeah

that page has its entire setup

LOL basically what I expected

the amplitron is that guy all the way to the right with the vac chamber and the 4 sealing bolts bulk heads

I personally would love to have one as a collectors item

[leaden crow]

Does this seem wrong? The conductor appears to have conductivity with the shield.

[real cedar]

i can check my spare

one sec

also it looks like the cedar city site does have a NOAA radar .. I think thats what you pictured .. its on the taller tower above the trees

[real cedar]

ive got zero connectivity. are you sure your not accidently touching?

[leaden crow]

yeah, tried a bunch of times

[real cedar]

okay do you have a nano vna or some other instrument you can hook it too. Ultimately dont be afraid.. a short on a radio to ground is FAR better than an open or perfect reflection

[leaden crow]

I wonder if they botched a solder job on the connectors. I've had 2 ali cables like that now. I have to check all of them.

I do have a nano vna

[real cedar]

or they fucked up and put RP SMA on instead of SMA

easy mistake to make

Oh and dagron.. it should have SOME connectivity. DO you know how a cavity filter is made?

its just coils inside some form of cavity material

thats why they can be hurt by DC

[leaden crow]

I have no clue, and I'll have to figure out how to measure this thing. I only know how to measure SWR and cable loss

[real cedar]

https://www.google.com/url?sa=i&url=https%3A%2F%2Ftektelic.com%2Fexpertise%2Fcavity-filter%2F&psig=AOvVaw3sPWgQAGW3HTPbnIwehK-c&ust=1744241302419000&source=images&cd=vfe&opi=89978449&ved=0CBQQjRxqFwoTCMjV3KTLyYwDFQAAAAAdAAAAABAJ
So heres an image of another filter. The number of elements, and the geometry determine everything

[real cedar]

pretty easy but its not a 1 port deal

you need to run a two port test

https://groups.io/g/nanovna-users/topic/understanding_nanovna_non/72768930

those are two things you generally need to understand when measuring with the VNA especially beyond 3G where your using the harmonics instead

Wrong PDF ROFL

here

https://jonase.se/WP/?page_id=337

Should give you the building blocks of what you need to do. You want to use the same antenna.. measure its performance live with another transmiter that all parameters are known about in the same room.. then you want an isolated tx and rx path and you can measure how the filter does from both ends

OR you could just do the whole send the signal through if you just want to look at the filter and not the whole feed

which is WAY easier

With that youll just do a SOLR calibration first to get your VNA calibrated for the line.. Connect your COAX on both sides to your two ports

run the calibration

Do a sweep from 700-1100 MHz

should get you all the s parameters you need to graph and get a good idea of how it responds

Expect something like that as an output

This image is not from these filters

but more to serve the point

When im back home ill measure mine

That's really all you need. And if you want to eliminate the entire feed line, including the couplers that you use to go from SMA to N, then just use type N calibration connectors

[leaden crow]

I am trying, but I am not getting anything useful out of it. my charts look wrong compared to this:
https://www.youtube.com/watch?v=F17mN5uuzGY
I can't load that jonase.se page

[real cedar]

You probably don't have it configured properly. What are your settings for your sweep?

Let me go ahead and do mine real quick

[leaden crow]

haha, I think it's fine. I had to reboot my nanovna it was just showing static. restart instantly fixed it

[real cedar]

All you probably had was the Smith chart. If you had a green trace that was making a thousand circles or you had it in continuous or CW mode, then yes, you probably would have seen a lot of static. But that's only because it's continuously transmitting and showing you what it looks like in a continuous transmit mode rather than a single two-tone or a single sweep.

And all in all, I'm glad to see that your filter is a lot like what is expected, with actually a little bit better roll-off than is advertised.

[leaden crow]

no, I was trying all the traces, it was staight up showing only static. the same as cables plugged in or not at all

probably what I get for buying the aliexpress knockoff. 😂

[real cedar]

Weird, yeah? I guess a reboot would fix that. I guess I get locked in a weird state. I have the light VNA64 which goes up to 11 gigahertz, but it's much the same hardware with a little bit improved hardware underneath in some special math to do the work on the harmonics. What I would really love to have, and I don't know if I can modify these for the capability, but I would love to be able to plug it into a computer and just live a view instead and do all the stuff from my end on Linux.

Also, you want S11 log and S22 log. You don't want the cable. You want just the filter. If you've calibrated appropriately, you should just be getting the filter response, not your feed lines response, which I think you're getting.

[leaden crow]

I calibrated it with the cables installed, that should remove them from my results, right?

[real cedar]

What the device actually does as math underneath in order to isolate the cable is if you calibrated appropriately, it will use the theory of small reflections and those small reflections it will use to characterize and eliminate the cable, which is cool. And I'm not sure if people actually know that that's what's going on underneath.

[real cedar]

Well, you've selected the wrong menu item. You're on S11 cable. You need to be on S11 log, which is just the device under test, and the logarithmic S11 readout.

[leaden crow]

oh, ok

[real cedar]

S11 cable from what I can understand on the device because their documentation is awful, but we should expect that from China, is that the device then has the S11 cable option if you'd like to characterize the entire feedline, which is, you know, very valuable if you want to do that, but probably not valuable if you just want the filters output.

Basically what it does is it adds the responses of the cables back in or subtracts and adds it to the response of what you're seeing on the display. So what you're actually seeing right now is the filter plus those cables. Given your cables are short and likely from a reputable brand, it's not a big deal. You still get most of the response, but you might actually pick up a little bit more on accident on your cables. Cables are antennas.. Antennas are just a wire

So you'll probably see in the neighborhood, say, about 60 dB rejection where you should be seeing probably like closer to 70 or 80. Personally, I'm seeing 86dB rejection out of band right now. And that's with not having the cables and S11 cable enabled.

[real cedar]

@maiden wraith So one of the problems with the ridge line up there, just talking generically, is that it's part of what we call geographically as a back arc. Wind and other forms of erosion have essentially made the mountains lean at a certain direction so the peaks are farther back from their forward ridge lines and so line of sight is actually quite difficult to get near the base of the mountain so it's unable to put something on the peak unless you want it to go along long distance or you want to search upwards like radars. So what I was actually thinking of doing there and that spot is to go down to the lower benchmarks that you see in the pictures I'm about to upload.

See how a very large amount of mountain jets out quite a bit from the actual peak. It might be better to be on the lower benchmark. You're not as high, but you have the rate line of sight and you don't get blocked by the other ridge lines if you find the tallest one. And that's what I've been trying to do on topographical data.

The only issue is a lot of those rid lines have trails and they're not trappings by all but some of the most adventurous hikers as they do tend to require some cliffside jaunts and there's easier ways up through the streams but they're still traveled so it's something that you can learn about whereas where yours is currently at it just blends in.

[white field]

fun planning map

https://www.floodmap.net/

Helps figure intuitively visual LOS coridors.

I'm vibe checking some peaks for a possible vegas link. TBH, I don't think Dessert is the best option.

Boulter or Black Crook appear to be better candidates since they can see in the lanes better.

This is land above 2200m from sea level.

oh, I guess you could do an overlook over beaver.

TBH, I feel kinda better about lower elevation desert peaks for repeaters vs 12k ft in mountains.

Winter access is impossible on those ones and it's a PITA to reach them.

This is a possible path to Vegas

• leaving SLC mesh
  • option 1: Boulter Peak - https://maps.app.goo.gl/JGZEv2nZt9R9gBCx7
    • pros: there will be far less local interference from town
    • cons: provides less coverage to urban areas
  • option 2: Desert Peak - https://maps.app.goo.gl/8XuSRx1RE3gd9AuP9
    • pros: better coverage for town
    • cons: more radio interference since it's overlooking a Toole
• Granite Peak - https://maps.app.goo.gl/wQ5SigZ8eAzbd3n29
• Burger Peak - https://maps.app.goo.gl/FC9EKfWz6BrViVK36
• Mt Charleston - Griffifth Peak - https://maps.app.goo.gl/QMXQpXqww6NfMP7x5

^ this would possibly higher power repeaters to be reliable since the nodes are all very far apart.

[real cedar]

so.. tbch if you want logan.. we need to face putting on on ben lomond on a peaks tree

and hope nobody messes with it

[white field]

Yeah, my general stance is that I only place them in sneaks spots or unbelievably remote access areas.

[real cedar]

as close to the peak as possible

I ran all the LOS

[white field]

let me take a look at the satlite images again.

[real cedar]

ben lomond or willard and a baldy router is basically your most reliable link

[real cedar]

It says that but that distance between in and ben lomond is enough not to work out likely.. we kinda already have that problem just from clearfield to NPR

[white field]

Yeah, you normally have to select a leg that has trees at lower elevation.

I wouldn't want to leave one out in the open in that popular an area.

Girffith looks like its tricky since only north aspect has trees.

I hit this problem with onaqui and got hosed on my placement.

[real cedar]

the problem is distance.. and id hate to have to make that hike again so to speak.. NPRs only 40 miles and I have clear LOS and its hard to talk to it reliably at all

[real cedar]

that too.. do you have a high resolution of ben lomond

im curious

or is that Gmaps

[white field]

some of these dead trees look decent

ridge shrubs could also work fine.

I've seen ridges like this on Skyline Drive that had the problem of no tress on the actual ridge bc of snow loading.

[real cedar]

ROFL.. yup and youll have to make that node small. you dont need a giant antenna for a lot of useful gain. Ive tested these

https://www.amazon.com/Generic-Meshtastic-915MHz-Antenna-10dBi/dp/B0D3KPFVH3

their gain is solid

its not 10

but its 8-9

out of all 8 I have

its got a load coil to make it virtually longer

[white field]

I've been using the Alpha 5dbi antennas.

They seem fine

I think one important consideration for anything long range is going to be ambient noise floor.

I've never actually outranged any of my nodes when I was outside of the city.

80 miles is light work if they don't have to deal with RF clutter from a city.

[real cedar]

so.. amazon reviews have tests in them.. their findings coroborate mine.
If you want a tiny mesh node.. up there.. odds are disguising it as a weather station will make it get left alone @maiden wraith

like if you made it small. but the solar panel was a bit larger and you had say 2-3 21700s

in a wiz minimesh ip67 case

you could make it look like a weather station

Nobody would question it on a trail as long it DOESNT look like a helium miner

[white field]

lol, just put a 🌧️ sticker on it

TBH, mine on Skyline is technically a weather station.

[real cedar]

ROFL nah id get a professionally done 3dp label or 3dp yourself but do it with a 0.2mm noz and a lot of good post processing.. make it look molded.. inlay it

aint nobody gonna think MeteoMESH with a gov warning.. aint something legit

[white field]

I really prefer to have that as my story for anything I'm doing. It's just long range, open source weather stations.

[real cedar]

ROFL

Like im putting anemometers on mine (fake.. but still)

I guess the key to getting it on a peak without worrying about it being looked at sideways is how realistically can you disguise it as something legit?

[white field]

I'm just avoiding popular summits. People know about helium miners now and I personally don't want anyone to see them unless they are going on existing abandonded infra.

[real cedar]

Yeah.. well thats why francis peak is good

kinda why I want to get up there and power it back on.. if @maiden wraith doesnt mind.. I can try but I have no way of changing its config there

Is it just a standard jumper into a rak?

[white field]

That's why I like Boulter or Black Crook over Deseret: for each - access sucks, it's not popular, there's some not-too-noticable spots

[real cedar]

yeah Im slowly abandoning deseret. Absolutely baller line of sight if it can be pulled off but I think theres another more desolate way to wendover

Desert Peak

In the HILL AFB air range.. well on its border

[white field]

Onaqui can see the hill overlooking wendover.

[real cedar]

ive thought about being a troll and have it send out a QRZ hello of "Boom Boom" ROFL

[real cedar]

so that other hill that high point

yeah

one problem

its in the southern range

[white field]

This isn't even a tall hill

[real cedar]

I dont want it nor me .. bombed

ohhh

not the stubby guy

the range before that

got it

[white field]

fak

that spot is private land

hill north of town that's super prominent is public tho

[real cedar]

LOL so i thought you were talking about this hill and I was like this is a bad day for me and the node

https://maps.app.goo.gl/hXdpC3xYGPrTeb6c7

how travelled is this

google has basically nothing

[white field]

desolate

I go there for camping sometimes

do anything you want.

It's just cows

[real cedar]

Thought so.. hide node in a pole in a tree.. then put the solar panel south facing and give it a desert paint job?

[white field]

tbh, I think the wendover side might be nicer

What coverage are you looking for?

raw distance, onaqui is taller

[real cedar]

Not bad at all. Im looking for reliable way to get MF or SF when we switch to MF or SF this summer (planned afaik).. to get the signal to goshute

or a sister peak

[white field]

hell yeah

do it

you can drive most o the way to that summimt

let me find the road I've gone up.

[real cedar]

I make enough trips to wendover itll make my list this spring and summer

the issue is I dont know the goshutes well. I dont know whats traveled and I dont know whats meth head land either

[white field]

TBH, I want to replace onaqui with this node:

[real cedar]

lol

[white field]

https://maps.app.goo.gl/jxbjZ1V2yLks3zZZ6

[real cedar]

oh damn

I ran the LOS for that

two weeks ago

HAHA

I was looking for the same thing

Where to move onaqui

[white field]

NPR + POTM + Skyline

[real cedar]

yup

[white field]

• the southern or western links.

There's also already some radio infra up there so it wouldn't be disturbing prisine env if you wanted to put something high power up there.

[real cedar]

I say go for it. Whatever has the better coverage of toele

People arent having any luck getting NPR in toele

not even in erda

and they can see it with their eyes

it got popular at the hacker space so a couple of them live out that way in various spots and have told me whats up haha

[white field]

I wonder how many hops it would take for me to send messages from Burning Man this year.

I'm on a discord and they are planning for a ~1000 person publish mesh firmware.

[real cedar]

oh shit your going to E13?

[white field]

Nah, for Burning Man in Blackrock Desert.

[real cedar]

ahhh

[white field]

It's like 60-80k people. No consistent cell service.

[real cedar]

yeah I know someone was talking about this on another forum im on

short turbo gonna be fun

LMFAO

[white field]

Yeah, our camp has an 50' scaffold tower. We ran a node on it last year.

I bet we will end up hosting 2x this year. One for a public mesh and 1x for a smaller private group.

[real cedar]

Man yall have me wanting to come out

[white field]

Yeah you should track the meshstats pages while it's happening. People will do MQTT uplinks

[real cedar]

wait.. do you live in logan.. the reason I ask.. is I was on snowbasin.. and I caught logan nodes while there

[real cedar]

where are they at?

[white field]

hmm, let me see.

[real cedar]

francis wouldnt see logan even if you were ontop of the radome

Unfortunately :/

Ben Lomond (im running this at 0 ft above ground as a worst case)

Kinda the problem with ben lomond becomes.. if I put a router there... Nothing... and I mean nothing should be a router in between it, francis and NPR

or it will just congest

at least in theory

Mostly because of the amount of area it would cover.. but I think it combined with francis.. could connect the whole northern area

and what my grand master plan was.. and hope it serves others plans... was to go and place something out on tangent peak there.. and then on a peak near wendover and eating the hops

@white field ... It would essentially guarantee everything from tremonton down to SLC has comms

sorry desert peak not tangent.. and I was like well the entire desert and the air ranges get coverage at that point.. put one on volcano peak.. go to the peak above wendover.. place a node.. boom

but that be a lot of hops to get there so desert might be a stupid idea

cedar mountain HP seems way more viable now looking at it cuz it sees the one id put on promontory point (neptunia)

would be way less hops.. so you have me sold

You know what would be super hilarious.. If we got that all working on MF51 and yall at the man just trolled everyone and all tried to connect through one of those poor routers haha

Ill start cobbling something together in the spreadsheet with actual locations

[real cedar]

No Oh... well you know what.. Its on my todo soon (tm).. either this weekend or next if the weather holds. I think im going to something up on city creak flats to rudys flat.. to serve the avenues.. and be kinda like point of the mountain but up north instead.. would give the city more universal coverage.. But after thats successfully up I can see about the trek out to francis

thing is.. now knowing its a bit more into the snow.. HAHA part of me is like.. should I bring some skis?

I got a backcountry back up.. could even use them to make the journey

Ill hit up my buddy and see if hes willing to make it with me

its still avalanche terrain up there and its wet slide time

[broken turret]

Thanks 😁

[broken turret]

Depends on how busy your mesh is. Take a look at chutil on your high sites, and adjust accordingly 🙂

[real cedar]

I have a lot of the routers metrics they send and their utilization is about 13%

peaking at 17

lows around 5-7

[broken turret]

I normally set mine to 30 minutes. But our mesh is on SHORT_FAST, with tons of headroom.

[real cedar]

so I got given access to the spreadsheet. Im doing a large connect all of utahs little cells together plan for Medium fast with a minimum level of routers

Ive got it down to routers can see other routers that can see unique cells but not constantly talk over each others stuff

with router lates midway down in really good spots

I think it might just work.. getting all the deets and GPS coords in and stuff.. will see what people think

this areas funnnn. I basically have divided utah into unique cells

[broken turret]

Sounds like you have it well in hand 😁

[real cedar]

Im doing my best.. this is not easy to connect up

Could use a full network engineer at this point haha

st george.. FAHGETABOUT IT.. i aint hiking in the mojave.. someone else can figure out cedar city and south into the devils asshole

[real cedar]

@white field Take a look at what I put in the spreadsheet. I made a new sheet or two. I have NOT made a full link map yet for MF51 but I have all the minimal sites we would need. Tooks me hours tweaking LOS. Feel free to verify and once we have a suggested mapping software where I could save the map.. ill use that to draw it up

No cells router sees more than 3 cells
Each router sees 3 other routers
Each router late sees 2 routers and 2 - 3 cells
All links under 5 from either end to ensure 7 hops is possible

@broken turret for a grand area total of 69067 sq miles.. 16 unique cells.. 21 routers, router_lates and clients in total all encompasing.. Do you think that would he too congested in MF51? Given what I posted a above. Just a shot in the dark guess is what I desire

This would cover from Nevada.. To southern Utah.. All of northern Utah and its deserts. A good portion of central and southern Utah. As well as a sort of dendrite out to the badlands, canyonlands and Moab area.. With routers on strategic points out there. So it's a massive and very seperated (geographically) area. Like all routers would need to be BIG pipes essentially lol

[broken turret]

Depends on terrain. I don't have enough info to answer that.

[broken turret]

Looking at the network topology you describe here, your biggest issue is likely to be collisions. How well it copes with those will depend on how much traffic is running over it.

[real cedar]

Yeah I kinda figure the farther you get from the salt lake bowl.. The more likely it is your broadcast through a couple routers might collide with something from the entirely other side haha

[white field]

Has anyone range tested anything like short-fast in zones with low noise?

If short-fast works fine in low noise environments we could run it without problems and then put up higher quality routers near cities?

Dramatically reducing the clutter problems.

[leaden crow]

@broken turret although I'm not sure about their mesh being low noise.
The more I read Bay Area Group discord, the more it sounds like filters are the answer for noisy environments. It would be amazing if C6 solved a really tight low cost filter. #meshtasticd was talking about a $2 Qualcomm saw filter module yesterday, but I'm not sure about that one. The datasheet says max 20 dB input power. I'm not an EE, so maybe they have a plan to make it work.

[white field]

@real cedar , one thing I think we ought to consider is viable end users.

I'd love to see us focus on providing links to zones that have a lot of land users AND no cell service.

Indian Creek fits this perfectly since so many climbers spend weeks there without cell service. As a long term project, I'm a little tempted to setup a dedicated link to there. (which translates to a discord server and then doesn't have any other traffic from the city mesh)

I think it would require me placing a dedicated mountain node that can link back to the grid where it can connect to an RPI. (either managed at my home or remotely at someone else's)

^ultimately, this was my goal last year and I was intending to rely on the public mesh infra for it.

However, I'm not 100% confident that the required 7-8 hops link will be stable with city mesh traffic.

• SLC backhaul to discord (home)
• SLC mountain router
• Boulter Peak - we can drop this if we can a good filter for the SLC mountain router
• Mary's Nipple
• Mt Ellen
• cliffside overlook in Indian Creek
• Indian Creek repeater - local only
• personal device

[leaden crow]

It would be amazing if we could link meshes with MQTT and have the node doing the relay strip out all telemetry, etc., so only messages got passed.

[white field]

I kinda like it being viable without internet tho. 😄

hop_limit: 8, message: how's the climbing? When are you coming home? We are hosting a bbq on Sunday

[leaden crow]

OK, so make the relay between presets. have a dedicated long range link that connects 2 meshes. The point being that having a relay filter all the telemetry. Yeah you won't have a node map for remote locations, but who cares if just want to send / receive messages?

[white field]

Yeah, I guess your local mesh would be insulated from city trash.

So your local comms will always work.

I mean, you get that for free tho.

nobody is setting hop_limit: 8

All the packets drop off by default the farther you get from town.

The only node which should have traffic reach the edge is going to be the internet backhaul.

If you're back on grid, you should reconfig radios to have hop_limit: 3.

[leaden crow]

You could have that relay also reset the hop count so it can have some distance in the remote mesh, just make sure there is no way the traffic gets reflected back to the originating mesh with new hop count

[white field]

ah, I talked to the firmware devs about this.

If you increase the hop_count on a node you will fuck things up and result in infinite loops. (crashing the mesh)

You can do transparent backhaul nodes that have inlet/outlet nodes tho.

• inlet: public mesh
• serial transform connection
• link: private mesh
• link: private mesh
• link: private mesh
• serial transform connection
• outlet: public mesh

[leaden crow]

I kind of assume this, I'm sure other people have had same / similar ideas.
I bet somehow it could be done. I think the key would be to have only a relay node config that can do it, it would only increase the hop count one direction, and it would probably have to keep a state tracker on messages it reset the hop count

[white field]

yeah, you basically end up building a repeater chain on a separate mesh config.

There's a github issue asking for a way to link nodes over serial to achieve this.

You need to be able to TX with two separate radio configs to keep from interfering. You also need to make sure that the private mesh as more bandwidth available to it than the public mesh.

Anyways, devs haven't done it yet.

If I follow-thru and take time off work this fall/winter, I might look into it.

[leaden crow]

makes sense, you could solve that with different presets, but it seems like you'd want the preset with the higher broadcast power on the private mesh

[white field]

Tho with stocks crashing this week my savings are fucked.

[leaden crow]

haha, everyone

just be happy you're not trying to retire this year

[white field]

yeah, I was like, "maybe I'll quit this month"

oooffff

yeah

"Maybe I'll keep working this month"

Suddenly motivated to do this project at work. 😭

[leaden crow]

125% tariff not going to be good for my aliexpress habit. 🤣

[real cedar]

Saw filters are limited by their circuit size. Cavities are the way but expensive. It's on my todo to see if we can make a tight one for FS51

[real cedar]

I got ahead of it all. Kinda why I hit the bullet testing all the stuff

[real cedar]

I out a good few of those in the spreadsheet along those thoughts in the minimum routers setup. I can add a backcountry section to discuss how to get stuff through the cottonwoods and uintahs. It's a super tough problem and requires a lot of LOS and prop mapping to find a good site though I like to start at any historical record of a radio site there in the past

@white field
I want to use https://qgis.org/ and/or https://kepler.gl/ to do the mapping

As for links that you want to be stable.. I think our best option is meshing-around+femtofox dual radio relays.. I think at some point we have to bite the bullet.. Accept the node limitations and connect networks with mesh interlinks on BBS? Like city traffic isn't an issue as long as our routers stay under the utilization. A lot of our problems are geographic. I think we can solve them at least within 250 mi of SLC..

You can get the cottonwoods with 3 routers

You can get heber with a Park City Node or even just Clayton Peak at Brighton.. Haha

Ubiquitis software works too but these are free and we can just share a file I think

[real cedar]

Yes.. But here inlays the problem too. If node a in city wants to talk to node b in the boonies and that takes 5 hops.. And b has a hop limit of 7 but a has a hop limit of 3. They will never be able to talk. I don't understand this part from all the dev talk I've heard

I feel like hop limit is dependent on the size of span of the mesh too

Would love to have a more clear answer on this for our particular scenario

One of my issues is because it take 4 to 5 hops to talk to some people.. My hop limit is set to 7 on my device unless I'm actually close to all those people

If I set it to 3. I can't talk to anyone except people on a DIRECT hop from 38fb ot NPR

So when I keep hearing from the devs don't change this limit but I can't talk without doing so.. I'm not sure what my options are 😂😂

[white field]

I mean, we can fork the firmware if we want it higher and it works.

[real cedar]

I'm guessing that limit is there mostly because of the managed flood multi-hop mode, right? Or am I missing something?

Anyways, I think what I will do, if you look at the spreadsheet sometime here soon, again, is that I will focus on the core population area first and see if those router setups work. And if they do, and they work really well, I think that's when we should start to branch out with medium-fast on frequency slot 51 from there.

I might very well be going to put something up on City Creek Ridge this weekend, so we should have some better coverage of that area that just doesn't really get any love.

[vernal oracle]

According to a White House press release, your $1.00 SMA connector will have a duty of $1.25, with a minimum of $25.00 per item. That means one male and one female will cost $52.00.

[real cedar]

I'm happy to let y'all know.
Bosch env sensor
Waterproof baro and temp
UV sensors
Rain sensors
Fram
And RTC modules

ON RAKs do not contribute in any meaningful manner to power draw. In fact my smallest solar node has GPS (off), and 4 sensor modules on it and is actively charging at an appreciable rate

Making a router be a weather station is totally possible with minimal power draw. The catch is.. I've only tested this with the RAK ecosystem

[void iris]

Most tarrifs have been suspended today. Except for China unfortunately. It'll be interesting to see what happens.

[real cedar]

@white field so your idea earlier about connecting back country. Let me show you what I came up with for the cottonwoods. (people are less karen about xmitters in the canyons because theres lots of them for various other purposes)

You DONT need these to be routers.. They can all be router lates or Clients tbch.. Yeah that probably seems a little strange but because their entire purpose is to just branch out to extremely hard to get to local areas.. router late or client is best

So Mount Raymond sees NPR. So does Wasatch Crest.. But mount superior doesnt. However this layout and of course with more clients comes better coverage in the area.. you could serve this area without many hops

Its difficult. One is also going to have to accept a peak top node.. like theres just no way around that in glacial and slot canyons haha

In fact the tallest peak hurts you more when the zone you need to cover is at the bottom of a canyon haha.. so Maybe a better route is to find somewhere midway up which is what im running now for JUST big cottonwood as I know it best

Like ill do something like this. Find a good campground for a permanent client node to relay stuff.. and kinda see what it can see and low and behold we have some ridgelines.. so if we were able to put something where it can see BOTH sides of the ridgeline then thats a good repeater or relay node

and voila.. we suddenly have a nice candidate for relaying something at the hard to reach winding road level

AND it sees mount raymond which sees NPR if we put something there. This is the fastest way I think we could survey and deploy something to places without cell signal and stuff

Without having that 7-8 hop problem

ROFL that canyon node sees POTM.. thats OP and a nice surprise

Its extremely touchy.. just 250 ft up.. your entire road is now blind.. wont work

If you want to cover the Uintahs.. Its viable with mt tabby and Ice Cave Peak but its a VERY difficult basin

I ran about 70 of these heywhats thats to find this

Its about your only reliable link (barely) (both antennas need 15 ft of elevation above ground..). From there it would be about strategically positioned clients and router lates but the mountains themselves are a foregone conclusion. Even if you put something on kings peak. 2 13,000 ft ridgelines and your signals gone

[real cedar]

ROFL hell yeah I got one too

I'm waiting on a flywheel

If you want to ride up 2x2

I did a 212 stage 3

Figuring out the jetting on the mikuni still

140 main seems a bit much.. Great for sea level but here it's a bit much

So this should start getting passed over long fast. More as a test. See who eventually sees it to see how well I can be heard through onaqui and npr

Unfortunately UV only worked through serial.. And it was flakey

I hope the devs add support for more Rak sensors

https://meshtastic.org/docs/hardware/devices/rak-wireless/wisblock/peripherals/?rakmodules=User+Button

[broken turret]

Take a look at the WISP tool on HeyWhatsThat. It lets you see viewshed intersections for multiple towers. That way you don't need to run 70 single-site calculations to find a viable spot. Give each tower a different colour, and the overlapping areas stand out really nicely.

https://wisp.heywhatsthat.com/?user=free5

[real cedar]

I'll give that a try. Is it best just to signup for the tool?

Damn! 60T what ratio you running?
I got a jugg-a-verter 5.5:1 slapped onto that stage 3

I love the way it rides but it's in a rebuild right now due to some port metallurgy on the crank. It's on my todo this weekend or next to fix

[real cedar]

Yeah I can hit 63 MPH.. Do I need to most of the time no.. I traded a small amount of torque for it overall. Mostly because I wanted the speed if needed cuz I definitely go to some shenanigans meet ups down town a lot

Disc brakes are NOT that expensive haha. They mount to the RB200S, B200 and I think BT200x natively for the rear. Not the CT sadly so yeah I see why your on drums.. CT requires heavy modification for front and rear.. Where as the others just require the correct fork and your good

I have plans to do that with mine. Disk brakes

There's mine. I've got a custom tank that fits the bt200x on the way

Rofl Yeah it's fooled cops until I went with my black and cyan paint

"Anubis" paint job

[real cedar]

Guess whats in the plans.. Mikuni 32 LMAO.. and a fuel pump.. and ported heads

Waiting on the port.. Too.. The way the stage 3 is setup is I shoved the longest stroke piston and crank I could so I could get useful low end torque. Strokers are the way 😂. It's hilarious I can do all this to a lawn mower

Shits gonna absolutely LAUNCH anyone thats on it.. and I cant wait to have to controlled wheelie off every stop light because of how much imma jam pack into the 212 haha

The only step after that is stage 5 but thats a lot of money I dont want to spend right now with the chinese tariffs

[real cedar]

So TBCH if your doing a billet rod and flywheel.. Move to the 26 lb valve springs.. rockers are meh until 9800 rpm

if anything chromoly push rods help once your in the stage 3 territory

ive blown multiple engines. I figured out what works haha

3d print the airfilter box

curl the air filter back under the frame and voer the top of the valve cover with a scoop. provide a point lower than the carberatuer with a drain hole for water and your pretty much set

when i get my true and full build complete ill post pics

a lot of the little things like my airfilter box are for comfort. if Im flying off the back a bit or need to extend my legs back... i definitely do not want to break an air filter for the 40th time haha

Long exhaust because I despised feeling any of the heat come up. Im planning a heat shield under the custom under the top of the top pipes of the frame so the fuel and heat shield protects me from any heat above.. When I remove the tank im planning some shrounds to funnel the air along the engine when riding so it gets cooled really well

Gonna get all the harnesses sleeved too. I really want it to look slick and everything be far out of any kind of impact spot.. The juggaverter is one of the best TCs ive ever run. No issues

We do shenanigans in Delta most years. if your interested and it happens this year I could hit you up before it occurs

[real cedar]

Sadly you suck dust in no matter what. I use that maxmima air filter oil for all my filters. I put a new filter on every 3 months

Does a fairly good job.. Predators were cheap now they aren't

Get while stock is still in supply before the cost goes up haha

If you ever need a spare block I mean

@leaden crow @white field

Forest camo green seems to blend best into our watch front vegetation

Khaki for deserts

I got the slate one but it's too dark even for rock mounts haha.. Definitely stands out

Oh I think I didnt covey what I meant. Im making an air box that fits a Suzuki DRZ400 K&N Filter. Theres not going to be really much dust that gets by that thing

https://www.amazon.com/Suzuki-Kawasaki-Performance-Replacement-Filter/dp/B000E28JLG

At least not when properly maintained according to maxima and K&N

Like Im making a housing thats quite pro. Ill post pics.. youll see how the scoop will suck neither water nor much dust. Its not a blank scoop without a filter

I basically am just creating custom bolt on stuff with proven moto tech. Not the amazon cheap pit bike stuff. Seems funny to do to such a cheap bike but We ride the BST a lot

dust will screw up my engine in no time

[real cedar]

Yeah flat dark forest came out super nice

Highly recommend that color

[leaden crow]

2.6.4 killed all my settings. I'm guessing that happened to D801 node this week?

[real cedar]

I don't auto update meshtasticd

So maybe.. Maybe not.

I won't be there to mess with it

[real cedar]

@pseudo kestrel seen your base station on the 999. Your in a dead zone

Your low and got a lot of concrete around you

[broken turret]

The link I provided there gives you up to five towers at no cost, and can save data to a cookie.

Subscriptions are expensive, but help a lot if you want to put all your towers on the map rather than just a handful of them. Subs also save to the server rather than a cookie. They have been generous and gave us a free sub for our mesh here, which has been incredibly useful.

[pseudo kestrel]

Is there any infrastructure up around Kamas that I'm likely to see on my trip today?

[leaden crow]

Probably not, but if there is anything it's likely on frequency slot 20 (the default for Long Fast).

I installed the Acasom and Alfa at ds nsl today. My first trace up north returned, so that's a good sign. I'll run some traces overnight and see if my SNR and successes has improved.

│ 17 │ Aurora_Neptunia │ !452b53bf │
Route traced towards destination:
!eb5de3e2 --> !452b53bf (-6.5dB)
Route traced back to us:
!452b53bf --> !d36b38fb (-6.75dB) --> Unknown (?dB) --> !eb5de3e2 (-8.25dB)

[white field]

Btw, has anyone her done any coding with Cursor yet?

I'm shocked at how good this shit is.

If you have ever needed to do professional codebase chores, it's a total mindfuck how good it is.

It told me how to setup credentials for github CLI, searched for codebases that needed to be updated and then is now steppign thru each of them updating a dependency I had been working on.

I suspect this will trremendsouly good if anyone wants meshstastic custtom tools

[real cedar]

Yep, I saw your packets. Hey, how do your nodes handle all the heat in the box and the batteries?

My boxes are already getting to 40C on a spring day.

I realize sunlight is a factor, but at the same time, the sensor is actually in a shaded box with some airflow, and it's the one that's on my patio, which is cooler than normal. So, this is concerning. As lithium ion batteries do not like temperatures above 45 to 50 degrees Celsius.

25AH LTO battery 🤣🤣🤣🤣

My charge controller supports it. Im highly debating doing so

Would he a higher panel to battery difference too so easier to charge

[vital hemlock]

https://tenor.com/view/glory-charge-civil-war-civil-war-charge-war-gif-25539430

[real cedar]

Im thinking of converting mine to a 120 mi range Electric mini bike. With those foldable solar panels i want to be able to fast charge it at a camp

[real cedar]

@vital hemlock @leaden crow you know those 17 cm generic whips.. that I said were good

they are stupendous

its what on neptunia

Im curious to see if its better than a big high gain

6 dBi gain average

Probably will not go Nearly as long or as far but good for discreteness

[leaden crow]

You should test against the Gizonts. I think they are just rebranded and overpriced Aliexpress Gizonts. I have a few if you want to compare.
Edit: according to the docs, they are the same:
https://meshtastic.org/docs/hardware/antennas/#compact--handheld

[real cedar]

too bad tariffs make it not so fun to get them

where as I think at least for now amazon got a lot in stock

[leaden crow]

Yeah, true. I think the Gizonts were like $5 when I bought them last year.

[real cedar]

How are they for a router. We are considering it

they seem good enough

Its whats on neptunia atm

[leaden crow]

I could be wrong, but I thought it was mentioned they are a monopole and work better with a ground plane. whereas the Ziisor, Alfa, etc are all dipole and don't need the ground place as much.

[real cedar]

So attach a sqaure plank of copper LOL

On the SMA line

[leaden crow]

I could probably use one on my RV node, that one the antenna is mounted to a large aluminum junction box. I would have to change out my N connector for an SMA though

[real cedar]

Alright chonking antenna it is

So you know the northern part of the bowl where there is antennas on ensign

theres power pole infra

and if you make the device look like it belongs on the pole. Most people arent going to say anything

Since it looks like nearby stuff

Just was hoping to use less of a Gigantic antenna haha

[leaden crow]

I don't have an answer on the batteries overheating. I haven't had temp sensors on any of my nodes. I suspect this one:
#1197577977781821541 message
Overheated a few times last year and internally shutoff the battery pack:
https://voltaicsystems.com/v25/
It still works, but it was already kind of old, so I replaced it with molicels and didn't have any more problems with it.

[leaden crow]

Haha. The last time I was going to buy a gun. I spent the money on a motorcycle instead (exhaust). I guess we cancelled each other out.

[vital hemlock]

I’ve got a 23 XSR 900, love that bike, but your 1000R would likely bully it. 😆

I do love my naked bikes, would never go back

[real cedar]

I want an MT10

I don't do sanity

[leaden crow]

Love that bike, but I'm angry Yamaha is not bringing the XSR900 GP version to the US.
It probably would be closer than you'd think before the exhaust and tuning. Honda stock tuning on these is notoriously bad.
https://www.youtube.com/watch?v=d2xjd0TPNms

[real cedar]

https://www.yamahamotorsports.com/models/mt-10/specs

Saving the 14 grand atm

[leaden crow]

Personally I think that bike is in an odd spot. The Yamaha triples are a good value, but the MT-10 not so much. It's too close to Tuono V4 price.

[real cedar]

Yeah it's in a weird spot and honestly my choice is totally shallow

Everything in my place has a cyber punk aesthetic except my beater

The bike must fit

I've thought about scaling back to the mt09 or 07 but I like the 4 crossplane

[leaden crow]

You need way more than $14k for an Arch. 😁

[real cedar]

🤣 we are going custom at that point

[leaden crow]

The crossplane sound good, but they do not sound V4 good.

[real cedar]

Some wild friend of mine thought up an i6 bike

And I was like how are you gonna

Fit it in there

🤣

[leaden crow]

CBX?

Watch youtube video of the CBX. They do sound amazing, but I still prefer V4

[real cedar]

Nah he was thinking up a whole bunch of stuff that adds weight and how to mount it inline with the chassis instead then convert the output and all this jazz and I was like bro just settle for the i4

Though unfortunately in some ways I'm still debating an MT purchase. My priorities have shifted to wanting a larger dual sport

Enduro. I spend way more time off road these days

[leaden crow]

This is the result after running traces overnight from ds nsl to NPR
Ziisor antenna:
76 success, 18 fail
80%
SNR average -13.3dB

Alfa antenna and Acasom filter:
65 success, 7 fail
90%
SNR average -7.8dB

[real cedar]

Got a link to those Alfas?

Still trying to find the best router antenna that isn't 39 inches long haha

[leaden crow]

https://store.rokland.com/collections/all-helium-antennnas/products/alfa-aoa-915-5acm-5-dbi-omni-outdoor-915mhz-802-11ah-mini-antenna-for-lora-halow-application

https://www.amazon.com/dp/B08H8J6ZV6

If you get it from Amazon, I would do the "ships and sold by Rokland"

I got one that measures worse than the other two I have bought. I wonder if I got it from a different seller and it's counterfeit?

[real cedar]

Type N?

Looks to be

[leaden crow]

Yes.

[real cedar]

Have you found mounting it upside down to be problematic?

I kinda want to put panel on top antenna on bottom. Box strapped to pole for a couple setups

[leaden crow]

I haven't tried, but I think a lot of people have. The enclosure I'm working on for inexpensive nodes, I'm pretty set on mounting the antenna upside down.

For practical reasons, I want to keep the footprint as small as possible and not have an external antenna cable.

[real cedar]

I think I'm going to do it for this one because it's going on a powerful but it's going to look very very official so even alignment probably won't even think much of it. It's going to blend like any other box that's ever been posted on the light pole.. There is some power infrastructure that runs power lines down to some radio stations that are near Ensen and it's nearly perfect and I don't think that as long as you have a safe setup, then anybody is really going to concern themselves with your box, especially if it's painted khaki brown and official looking with official stickers.

I think that's what we're going to do, and we're just going to risk it, and we're going to get that really sweet avenues connection, finally. And we're going to have a point of the mountain-like router on the northern tip of the darn bowl, because that's going to be the only way to make this work pretty efficiently. And I will figure out Olympus when it's a little bit more melted.

It will have enough solar to keep GPS on so if someone does mess with it, it will pretty much find out.

[leaden crow]

I got the stuff to build more E22 hats, so I'll probably have one for you next week or the week after

[real cedar]

While waiting on parts I'm fixing the mini bike and getting it Going like a raped ape.

Oh, that would be great. I could use one more. I don't know if I'm going to deploy those to the mountaintops yet unless we have a plan for how we're going to power it. I don't know. I haven't planned for powering a Raspberry Pi 02w yet.

I think all my nodes are really sized for the Rack 4361 right now, and I'm trying to do things on 24 dBm to avoid having to go up to 30, but I'll go to 30 if it's the way to go for the really high nodes. I just don't know how to keep power usage low.

[leaden crow]

I think if we have heat problems with the D801 node, we should swap that out with a 02w. Have been slightly worried about that.

I think femtofox is really the way to go if you want a solar powered 30dBm. The smaller you can keep the solar and battery, the easier it will be to hide

[real cedar]

I've kept my raspberry pies in black enclosed containers without any kind of ventilation in the summer with just a large heat sink and they have survived being at 60 degrees Celsius or 70 degrees Celsius all day long. They don't have problems with that. It's the batteries I worry about.

I'm down to deploy femto foxes. I just only have two of them and they're kind of getting more and more expensive and with everything going on with the tariffs, I imagine he's having some trouble right now.

I'm gonna use one of them for my node at home here soon, and then I'm gonna use the other one on the drone. So my two are pretty much spoken for, but any e-bites left over definitely, could be used for the mountaintop nodes.

Here's a thought though. If we could find the same rock chip and the same form factor as the Raspberry Pi two zero-w, and glue that to a E-byte, that might be all we need.

I don't know if such a thing exists. I haven't searched. I'm just throwing ideas out.

[leaden crow]

The luckfox they use are like $5 - $10 on ali. We could roll our own, but it's more components than the rpi hat. At minimum we need a voltage step up to 5v for the E22.

[real cedar]

Well Buck converters are dime a dozen.. Are we serious about a spin cuz I'll get KiCad going and if we are serious about it. Do we want to baloon the cost a bit to include a ceramic cavity on the PCB itself?

As I understand it the ebyte has a version without an antenna ipex connector.. Direct feed to filter. And IIRC it is an sx1262 with the Return PA?

These would be frequency limited to what we choose

[leaden crow]

My immediate plans are to build RAK or XIAO nRF52840 based nodes. I don't think it's a bad long term plan to build our own femtofox based filtered board, but I'm happy to start getting simpler lower powered stuff working this year.
I also think the filter won't benefit my wasatch crest nodes much. There isn't even cell service where I plan to place them.

[real cedar]

Yeah the rak is also well supported and feature complete. It's very low power. Except meshtastics way of putting ublox Zoe modules in sleep is sub optimal. So that's a fun quirk I'd like to put my own efforts into figuring out. But I need a router module on the city creek ensign peak area to grab the avenues.. Lots of the blind spots east of npr. All of nsl, centerville.. Bountiful all the way to hill. It's a great spot. Just gotta make it blend on a power pole

[real cedar]

Anything you can do to help your rf environment is good. Filtering rarely harms but sometimes it's just not necessary. Like I might not filter Cottonwood canyon nodes in the canyon. They won't hear the noise of the valley

I would filter Mt Raymond or Mt olympus lol.

[vital hemlock]

Gah, just realized the same happened to me. 👎

Something is up, feels extra broken. Might downgrade

[real cedar]

😂 staying on 2.5.x

I knew it!

[leaden crow]

It also kills your meshtasticd config.yaml, but the default one appears to have the meshadv settings.
It’s been fine for me since reconfiguring all the settings I lost.

[real cedar]

Hurry up and melt I miss summer nights up there

(Taken with Google pixel 6 pro)

https://www.youtube.com/watch?v=rJps0n83DbI
Why I want a long range autonomous drone 😂. I want to map all my camp sites and ski resorts and back country areas

Why I'm building one rather. Will figure out legalities as I go

Quad Plane (VTOL capable plane) is the arch I'm shooting for btw

Meshtastic is going to be very useful for me here soon

Need proof of concept for a grander thought I'm mostly keeping silent on until feasibility has been proven

I'm honestly more big on meshtastic enabling peoples harder to do more creative ideas than just comms. Hence my investment

I have another idea. @broken turret there was an older more or less abandoned meshstatic location idea

I want to propose an idea that might be a moonshot and thus for #1217252717844500480

But

https://github.com/a-f-G-U-C/Meshtastic-ZPS

Forget zero GPS

Meshtastic Based Augmentation System (MBAS).. Each router knows its precise location. It's precise location could in theory be used as GPS Augmentation for hyper accurate location without the expense of the sheer power needed for multi gps receiver augmentation

All one would need is each GPS to use the almanac. Then you can measure the relevant differences and augment out all the channel conditions.. Additionally you can correlate the location data based on meshtastic SURVEY LOCATE data from the zps

Also you can augment zps with PPS timestamps. Thus you could just use GPS briefly to get all the time synchronized across the known routers and almanac synced.. Then turn it off.. And use zps augmentation for good location data 😂. Although this could definitely be weaponized. Be respectful people (if you saw what I saw)

[vital hemlock]

That's sweet. I use this when I'm ready to camp

https://eos.com/landviewer/?lat=40&lng=-111

I get a nice day old view of the area

[white gyro]

Someone launched a drone with a mesh relaying from it. They had it flying long enough to send the mars desert research station some texts from 11km away. We should be doing this kind of test ourselves! If they followed the rules, they would have gone no higher than 400ft. That area is really socked into deep canyons.

[pseudo kestrel]

wow, the import/export config options in the app do basically nothing 😬

[leaden crow]

I can't speak to the Android / iOS app, but my experience is the CLI is wonky in this regard too. It seems like if I do meshtastic --config it never applies all of the settings, it only gets some of them.
Also tried with multiple set:
meshtastic --set lora.region "US" --set device.role "CLIENT" --set lora.modemPreset "LONG_FAST" --set lora.channelNum 51
Same thing, seems to only apply some of the settings.
I have just been doing the --config and checking what settings it missed after restart..

[real cedar]

Approved haha

[leaden crow]

I just configured a new T1000E this way and didn't experience a problem, so YMMV.

[real cedar]

Ahh see I want to go one step further. Figure out drone autonomy. Pass missions to it over meshtastic

Might use BBS for that not sure yet.

Ensign peak router hardware almost in the final form. Once that's done I'm going up there to mount it before busy busy traffic

[white gyro]

You probably have to build the drone for it to have that kind of integration. I want a drone that can park on a cliff and go low power without turning off. That also needs a custom build.

Maybe the mesh radio could power on/off the main drone

[real cedar]

I'm basically doing so lol

Do you know how epic it will be to have the entire state worth of areas as testing grounds. Also the ability to go get what is really hard to drone map

I have a router ready for ensign

Testing out solar tomorrow and tonight

GPS

[pseudo kestrel]

Is firmware 2.6.4 having poor battery life on the T1000-E for anyone else?

[real cedar]

No mine's fine. I'm still getting about a week as I was getting before. What I will say is when you go to the firmware, just go ahead and do a factory reset after everything's back online. A full erase is just better for this upgrade. It kind of sucks, but it's not that bad.

Back up your settings before you do so and then just re-upload your settings when you're done and the floor should get rid of any weird quirks.

Also, coming to the avenues near you. We have a node going up at Instant Peak on a power pole here, probably this weekend. That's ideal. We'll see how it goes. I gotta talk to my buddy. But once we get it up there, everybody, including you, Rex, will be able to come out of your dead zone and be able to talk. It'll also help me talk directly to Salt Lake City without jumping through all of the other routers.

I'm gonna put it on 2.6.4 and hope things go well. I do want it to be newer. I want it to be the newest router in the area. And hey, if it helps the hops and it helps the network then great.

Also, here's a life hack for people with routers. Not only should you learn managed mode and figure out how to do all the admin key stuff, take a full, DFU level backup of your firmware, which contains all of your meshtastic settings, and then have that good config on your phone, and on the Android phone have the NRF-52, and this mainly works for RAKs... But have that, and you can do this with other microcontroller based architectures for this stuff. But the point is, here's the example, have that full level of firmware on you at the time. So that if you do lose your settings and you do have to go back up there, you can simply flash the good thing back. That known good image can be flashed very easily and you just need to be able to connect via Bluetooth again.

It's very convenient and all you've got to do is sit up there for five minutes while you transfer eight megabits of firmware over the pitifully poor speeds bluetooth low energy mode.

It works, though, and that's what I'm doing.

[real cedar]

Also for the router nodes, I'm finally just giving up on GPS. It's too buggy for me to actually fix in time to put some of these up. And I've held off long enough. So I'm going to remove GPS as a capability entirely and just hope they don't get fucked with.

[pseudo kestrel]

Theoretically couldn't you triangulate with some pings if worst comes?

[real cedar]

Yes. I would have to put it in a mode that sent out uber regular messages to distinguish it from others but yes

[pseudo kestrel]

Unrelated, but I just ran my location through the site simulator 💀

[real cedar]

🤣😭 can you see ensign peak radio towers

We are putting a node up there saturday

It will see both the upper aves lower aves and anything ensign can see

Aurora Uranialis. Router

Which is clearly needed as northern hops literally would rather take the road to Provo then to SLC before finally going to a node in the bowl and I'm tired of taking the scenic route

If you still aren't served by that node, which hopefully is going to help me in Rufio, then what I would suggest you do is you find a high point that you can commonly see that you can see NPR as well. And I would put a client node on that spot.

If you need to use a building downtown, what I would suggest is you make a drone delivery mechanism and you put it on top of one of those exit guard doorways that will never actually see someone go on top of it ever. It usually just has drainage. And if that building can see NPR and you can see that building, there's your relay node.

Permanent magnet on one end, electric magnet on the other, pretty much the easiest way to do it.

Is it legal? Eh, one could say no, but is anybody gonna care? Probably not.

I was thinking of fixing a deadzone in down town by doing that with that weird spire building. Painting it white. Magnetically attaching it to the metal up top. Self contained antenna.. Smooth 3d printed body.. In a white cone 😂

With flexible solar panels on it 360 degrees. I thought it would be a hilariously novelty and have been chipping away at the design

[pseudo kestrel]

those ones?

[real cedar]

Yeah there's power infrastructure going all long the ridge line that you can't see with the camera

Its the plan

It looks pro

For reasons that are obviously I won't put its final look on discord

But to any passer by it will look like a bog standard power box

[pseudo kestrel]

I've seen rufio on my radios before, I think they've been poking my bot as well xD

[real cedar]

25W solar panel + 4 A charging circuit.. Power metrics onboard. 4x 21700s. Should be set and forget

[dreamy ledge]

Hey everyone, just started looking into meshtastic for the first time. The online maps that show nodes don’t typically show much in Utah, but it sounds like there’s a pretty decent network. How do you find out what’s existing and where to place new ones?
Also what’s the program you use to generate the maps with the red areas showing reception?
Sorry for the super basic questions.

[pseudo kestrel]

I use https://site.meshtastic.org to see estimated coverage 🙂

[leaden crow]

The default (Long Fast 20) in Wasatch Front had 200+ nodes last year. I would guess even more now.
If you check the pinned posts, slcmesh will show you roughly the coverage we have on LF51.

[real cedar]

I've got a good idea of where you live hombre. I was passing you on the 999 and your SNR peaked to 9 🤣 at one point in my location history. But I'll keep it to myself. It sucks though. I got a clear picture of what your dealing with and its exactly why I am putting two routers up. Olympus and this Saturday ensign. (AUR)

[pseudo kestrel]

I just love how there's a whole half of the valley I can't see xD

[real cedar]

See there's this problem with NPR. It's great if you don't live in a dip or have an other wise clear view of the oquirhs.. But if you can't which is more than 37-39% of the city. You basically need to hope you see potm which 28% of the darker zones don't. All of which have I. Common.. ensign, university, and Olympus

So I went for the higher two

And Olympus is gonna be a fucking challenge

It has to be compact

It has to be discrete

It needs solar

It will legit be a challenge. What I might do is find a place only climbers go. Vault over a bit to somewhere useless (and stupid).. And anchor it.. With a muzi gazont antenna or whatever

Its got one job. See NPR and other routers. Grab the other half of the city

Ensigns the other crappy situation spotfix. Thanks to 🌄🌄🌅🏔️⛰️🏔️⛰️🗻 GeOgRaPhY.. We have a mountain plateau in our way looking north.. With a quarry in it. So we need a convenient way to find nsl and everything south of the base more directly.. No sense in it going to Onaqui

The crappy situation stems from Farnsworth. It drowns out links between it and NPR. Kills receiver sensitivity and received SNR. So as ideal as NPR is. Farnsworth throws a wrench in that. I have the ensign radio ready and filtered. Using an Alfa antenna

[real cedar]

So Rax, Proof of this is right here. I'm standing on a mound for all intents and purposes that can see both onaqui and NPR. And you can also see pointed the mountain. So for whatever reason, it takes all of these bounces and it has to go through my network to do that. And I've even configured some of these as client hidden. And when I've done that, I've lost connection to the network. So for some reason it's necessary. And the only explanation I can have for that is that NPR is blind for some reason. And the only thing that would bind it from where I'm standing is Farnsworth Peak.

[pseudo kestrel]

this is all so fascinating to watch, I've never been interested in radio stuff before now

in a fun unrelated note, I got a ping from my own base station out in the wild today 🙂

(granted it was only a couple blocks away, but given how bad my location is, I'm still tickled pink)

[dense basin]

Hi everyone, new guy here excited to see ya'll

[leaden crow]

Hey, I have an enclosure I was going to toss, but you might actually be able to use it.
It's a chinesium pelican knockoff that some device came in. The ID is approx:
13.375" x 8.5" x 5.5"
Depending on your bus bars, it might actually fit 2 rows of those batteries (12 total)

[real cedar]

😆😭 do you want to see the jank that is the quickly done up ensign peak rtr?

[leaden crow]

oh, is it up?

[real cedar]

Not yet

Well not in its spot

So everything is screwed down

Then mounting puttied

Then super 35 outdoor etape

And now tacked on jbweld around the battery pack 🤣

[leaden crow]

haha, I've done worse. one of my nodes the battery cables / board were all just loose. Not even taped.
as long as it's super 35 or other good 3M / scotch tape and not cheap garbage that leaves adhesive residue all over stuff.

[real cedar]

It works wonderfully. Doesn't even dip below 99%!! 🤣😭

I tried. I made GPS slam the network with updates every 30s. Which BTW it handled it fine. And overnight it did not drop below 99%

I've removed GPS. There is literally zero chance of dying at that draw vs capacity and the fact I can dump the full 25W into the batteries

The tips of the MPPT output to the green screw terminals are tack soldered onto the screw terminals after screwing it down. Worked great. Should keep them from ever falling out?

[leaden crow]

I've never had problems with those screw terminals falling out, but extra insurance never hurt.

[real cedar]

41 degree hangs under the box antenna up top. Painted brown like power pole boxes

Should pass anyone's muster. Has a vent. Oh and I got some hilarious stickers

And top it off with the weakest lock humanity created

[leaden crow]

haha, nice. I think we should do stickers that say atmospheric measuring device for toxic lake dust.

[real cedar]

I was going to do that for the neptunia router location on lead mtn. It's the one that made most geographical sense

[real cedar]

Shitty locks and stickers. Keeping honest people honest since before electronics!

[leaden crow]

I bought a 3d printer a couple of weeks ago and have beating my head against a wall trying to learn freecad. I modeled a bunch of stuff in sketchup like 15 years ago and thought this would be easy. 😁

[real cedar]

FreeCAD is simple.. Watch Shawn hymels intro

Then open the program. Use the default layout to learn why you hate it. Switch to open-inventor.. Choose shapes.. Draw shapes.. Enter tolerances Extrude shapes.. Enter tolerances.. Chamfer shapes. 🤣

I tell people a lot. Don't over complicate CAD. Think basic geometry what are you doing with a chamfer. Your drawing a cylindrical bottom cone

[leaden crow]

My problem is that I'm trying to hack 2 .stl files togther, which is a bad starting point

[real cedar]

Anyone got suggestions for anchoring a PVC pipes into the ground. Maybe a screw auger. Water bottle and some light tiny amount of concrete?

For the Neptunia location it's a desolate peak with an even more desolate location with no rocks or vegetation so I need to make it mounted and official and not easily moveable

But the means need to be hikeable

[white field]

How remote? Any chance there's abandonded cattle fencing nearby?

You just need a fence puller if you can find a fence post + rock on site.

[leaden crow]

With no rock or trees, I think the auger is probably your best bet. Maybe guy wires if you can't get it very deep in the ground.

[real cedar]

Extremely remote. About as remote as desert peak on the way to wendover

[leaden crow]

I think it would be better to use a stainless zip tie. That luggage lock might look out of place for a commercial utility box.

[real cedar]

Super super desolate

Is there anyway to change POTM to router late? I think I'm going to make AUR (Ensign) router late

[leaden crow]

Ask @hallow dagger
It would need firmware upgrade first. ROUTER_LATE is new

[real cedar]

Here's my plan that I've kind of put in our Excel spreadsheet just kind of what I'm thinking as I get closer to actually deploying these nodes and after I've learned so much from people here and in the rest of the metastatic groups.

I’m placing standard routers on mountain tops at 10 to 12,000 feet because they rebroadcast all messages right away, ensuring fast coverage across the mesh. There was a warning that we need to worry about collisions. So I think we should really maximize the mode offered to us in the new version of the firmware. For the ridge lines at 5 to 8,000 feet, I’m using router_late nodes (ensign), which only rebroadcast messages after a delay if the high routers miss them, acting as a backup. In the valleys, I’m using client nodes that can rebroadcast but usually don’t if another node already has, keeping traffic low. At least in theory. Everything that's inside a building is basically client hidden at this point. It serves no point to make everybody's traffic route through a building. This should maximize the amount of airtime that's useful, get you from point A to point B in the least amount of hops, and also help you route down into bowls efficiently instead of jumping around them.

Also since my routers have Bosch sensors sending telemetry data every 30 minutes, I’m keeping rebroadcast settings on all packets so everyone gets the environmental and air quality info. I think having good temperature data no matter where you're at is kind of cool. These broadcasts won’t clog the network because they’re not frequent. If I see congestion, I’ll switch some valley nodes to client or client_mute or bump the telemetry interval to 60 minutes. I’m avoiding core port numbers only for rebroadcasts since it might block telemetry from reaching everyone. This plan uses high routers as the main hubs, mid nodes as secondary routing, and keeps the valleys quiet unless someone wants to do otherwise.

[real cedar]

At this point, I've exhausted every bit that I can study on this, and I think this will give us the least amount of collisions for the most area covered. Thoughts?

[vital hemlock]

If router_late has good visibility it will always add to the noise floor because it will "always" retransmit. As long as we except that, it's fine

I have no issues with it

[real cedar]

Yeah it's the consequence of another router sadly. Router Lates should be ultimately helpful. Really what we need to avoid collisions is good routing strata. Clients and client mutes will never be ideal. But router + router lates.. If those nodes see 100% of the network amongst themselves then in theory say Node a wants to talk from Ogden to a node in SLC. Well it broadcasts. The routers above it rebroadcast before clients.. But because the router late catches the nodes not all seen by 1 router it at least somewhat guarantees the message gets to clients in time to halt them from broadcasting while the router lates try to rebroadcast the reply back to the routers and back from node B

At least that's the theory anyways

Its the timing we want to get right to avoid collision and structuring the timing with the given roles might be the easiest first try

And in that way, we actually have a reference now. Instead of choosing a client mode based on, you know, where it's located physically itself, we could do that in relation to where it's located compared to a router or a router late. In that way, we can actually guarantee in some manner that every node that sees a router directly can probably be a client. One that doesn't see a router directly but sees a router late directly can probably be a client mute and one that only sees a local client probably should just be client hidden. I don't know. I've tried to think of a way to structure the mesh. It's a hard thing to do because it's a mesh in its nature. It's not structured, right? So the modes offer us a way to try to do that. At the end of the day, though, for clients, everybody should just use whatever mode works best for them. But this is kind of the general idea. The problem I found was that when I really dug into the documentation, it wasn't very clear how this actually works. Until I dug down into what actually triggers rebroadcasts in the code, and I realized, oh, you can set up pseudo-strata. What would throw a wrench in that, of course, is if someone put a router where they weren't supposed to, or you had a hidden node. It would however minimize the hidden node problem. Ideally, if most people were polite and followed this, then the hidden node wouldn't really broadcast that often or present the problem.

I decided to obtain scientific data so I put seven radios on channel 37 and I went and put them at various heights and the highest one was a router one that could see some of the local nodes but not all was a router late and I mixed and matched various modes and I observed the channel data for 48 hours and ultimately a router plus router late combo worked really well and only added about One percent to the channel utilization and what I noticed when I tried to do it with clients and a bunch of mixed modes was not only did the channel utilization go up, but you had a loss of telemetry data as if there was some interruption to the packets around similar times. So I assumed those were collisions.

@vital hemlock @leaden crow @white field One other thing I suggest and it's just a hunch that I'm going to try out. I am not convinced that routing all of the traffic by read broadcasting all instead of core port numbers is going to save our network in particular very much channel utilization or airtime utilization. I think when we do routers, we should let them re-broadcast all so that everyone's telemetry, if they want to pass it between certain devices, can be repeated by routers, instead of it just being localized to a cell. The only thing I'm convinced will actually increase the traffic significantly is MQTT, which we can just keep off.

Apparently, also the admin channel for people who want to admin routers remotely does not get repeated unless you're on all from what I can tell. So if we want that feature to work reliably, we should repeat all.

[vital hemlock]

Is the core ports only a newer feature? Interesting

[real cedar]

No not that I can tell

Its 2.4 and newer

[real cedar]

@maiden wraith couldn't get up to francis. Easily high centered the mini bike.. 🤣🤣 and they just got another 1.5 feet and the tops

Might be a late may trip

There was a point where I was walking and leaving the mini bike far behind. Wasn't worth potentially having the bike stolen. You can walk the rest but it's very slow

If your serious about allowing me to do what I want with that RAK. I legit will dedicate it to helping extend to Logan since it is yours and that seemed your goal

[real cedar]

@leaden crow @vital hemlock I've been thinking of the wrong word. It's not hub and spoke. It's tree topology. Meshtastic uses CSMA/CA, but collisions increase with overlapping coverage and dense connectivity. To reduce collisions you need sparse placement: where routers are 30-207 km apart, reducing overlapped cells, with terrain (mountains) providing natural isolation. The tree topology is some backbone like a Public BBS mesh centrally located in SLC everyone posts to... In a weird way it imposes branch limits.. So less message flooding compared to a full mesh. Because what has to happen is you have to communicate only with the BBS mesh node to update it. So it goes between you and two other routers and gets to it. And you update it and then it sends its updates to everybody else, but it only has to go along the mesh to that person. So you yourself do not have to send the message from point a to poiny b. You send it to courier B and he gets it to C so to speak.. Then people set their hop limits to however many they need to reach the BBS mesh point reliably.

Can we set this up?

I feel like it's gonna be the only way for somebody in Moab to talk reliably to somebody in Logan. Or somebody in the Uintas to talk reliably to someone in the Cottonwoods. First off, if we actually ever get that working, that would be so fucking lit. 🤣👌

[real cedar]

Sounds fine by me. That's if I beat you to Francis. Tbch the next two weeks might have me very busy.

I was considering relocating it to one of the other potential router spots. I mean it's a rak4361 right? Old 19001 or 5005 baseboard?

(Like lead MTN)

We are assuming it's even reviveablw and not full of snow and ice haha or knocked down a cliff somewhere from the intense wind this winter

[real cedar]

Yeah I know precisely where it is located haha

Also everyone

🫡 Aurora Uranialis Router (Ensign peak) going up soon (tm). On bikes on I street.

[real cedar]

Nice climb and views for a casual bike ride

[leaden crow]

I should have skipped my kids soccer game and pedalled up there with you. 😁

[pseudo kestrel]

here's a once in a blue moon occurence...

[real cedar]

Fuck yeah

[real cedar]

We cooking with fire boys

Everyone sees the node. It trace routes great in the aves.

Sees the whole bowl. Has -10.25 db SNR to my house up near the base 🥳😎

[real cedar]

The anthem of shenanigans must be played
https://www.youtube.com/watch?v=oXI20uaOUCo

[white field]

bahaha, I think he got the link when I put up a node in the window at Publik Coffee on West Temple... but now I've forgotten it and it's still re-transmitting. 😂

(gonna have to fetch it soon)

[real cedar]

😂 is that the new guy in the aves?

@maiden wraith we got batcave. Your on the map if that's you

[pseudo kestrel]

Is there somewhere I could find fun new ringtones for my tracker card?

[real cedar]

Does 38fb just have a 10 bajillion gain antenna or something 😭. I have literally solid SNR to AUR but yet still I go to 38fb for entry.. Exit through AUR. Why 🤣

38FB gets beef cake early node placement of the year award

Yeah, we finally got you thanks to my router placement. Now if I can just get the fracis peak. Although, after running a lot of surveys, I would love to still have it there for short-fast, but for medium-fast, we really only need something on promontory point for all of us.

[real cedar]

I didn't even know you could change it. All I did was shorten its duration because my god if it rings for five minutes one more time, I'm going to chuck it to Mars. Now it only rings for two seconds, which is nice, but I think I might just disable the ringing because it's definitely had some awkward moments.

[white field]

Its an alpha 915, 5dbi. (Probably on the wrong side of a tree from what’s optimal for you 🤣)

It’s just setup in a zone with practically no RF noise, so it’s super reliable.

Let me know if you want any settings changed for it.

I’m gonna probably do Boulter Peak in the next two weeks given the current weather. (Finally getting that SLC to Henries link viable)

Once the snows melt, I might swap boulter peak to Loafer Mountain.

[real cedar]

You probably are sadly or increase your hop count to 5. I'm 4..

[real cedar]

No, by all means, keep it as it is. That's freaking awesome that it's doing it. I mean, it's irritating for me, but by and large, if it's that good, keep it where it's at or wherever you were going to move it.

I just find it to be hilarious. And also, I definitely think we should exploit router late. Hopefully sooner, and hopefully in time for medium fast, I don't know when you guys want to do that, but I'll gladly ride back up Ensign Peak. It's a bitch of a ride, but I'll do it to switch it over to MF51 and Router_Late

I'm honestly convinced that we will be totally fine going to short fast. Everything I've looked at tells me that that's only a handful of DBs, and considering we're all trying to run very high gain antennas on high point routers, there's no reason why we shouldn't be able to get short fast working.

We do have the perfect test environment now. I mean NPR before anybody switches that at al we could do one or two dry run router lates with the more accessible routers which are POTM and AUR.

[white field]

Have you heard anecdotal tests from anything g but long fast?

NPR was horrible for many people, we need to get a filter on it

[real cedar]

I've heard some things, but I wonder if that was because of earlier firmware and poorer placement.

[white field]

The penetration was apparently horrible.

We might need new hardware to supersede it

Maybe an yolo 3w node with a filter 🤣

[real cedar]

Yeah, I really hope someone can go do that. In fact, I honestly will totally buy you guys the 10 dB antenna. If you're willing to put it on that thing, because I think if you cranked up the gain and you filtered it really well, it would do awesome.

I am very much considering going to Black Mountain. There's a lot of people in toele that it would help, and honestly it's a pretty good router late spot.

[white field]

We should move NPR to a new location. There’s a lot of infra there but it’s currently discretely sitting on an unused fence post

Anything bigger would need to be at a new spot or with approved placement

[real cedar]

If you think about it too, a 12 degree beam width, only on the vertical side, but 360 degrees horizontal. All you're talking about realistically, if I am completely frank with you, you're talking about a 194 feet of blind spot underneath it, if it's only about 8 to 10 feet above the ground.

I would highly suggest moving it. Did you see where I put Ensign Peak router? (AUR) 🤣🤣🤣

[white field]

It’s probably 3’ off the ground

Its on a steel fence pipe

I would suggest not making it any more apparent unless it’s in a new spot.

Tbh, if we got a filter on it, it might be able to direct link to skyline. Mary’s Nipple node can get traffic from it

[real cedar]

Well, that is problematic. Three feet's a bit low. That's why we got a 10-foot pipe for AUR. It's in the best mental blind spot you could think of up there. And yeah, we don't own the site. We're not inside their fence line. We're totally obstructed from view even when they have to maintain the place. So the reality is, being well filtered, I don't think we're interfering with them. So I just don't think anyone's going to ever mess with it.

I would definitely consider a new spot. Where were you thinking?

If you were thinking to move away from the infra, then chances are you're going to have to make some acceptable sacrifices. Like, you might just have to disguise it as a weather station and 3D print and actually post-process an anemometer or something. Make it extra convincing and put it on a Sch 80 PVC pipe With a proper mouth, and label it something like Nelson Peak, whether telemetry station.

if you're going to hide it away from infrastructure, the best thing to do is to actually make it pass muster, and I mean pass muster pass to the people that are tech literate and would actually take a second look.

I guess the way I see it, if it looks official enough, even on the second or maybe even a possible third look, odds are most people are just going to walk by it, go on with their day, maybe take a picture next to it, or slap a sticker on it. Really the people you just hope don't fuck with it are the people that got stupidly drunk but at 9,700 feet I don't know if I'm really worried about a bunch of people having beers up there all the time.

[white field]

Tbh, I feel like filtering the node and a small power upgrade would really make the difference

I’ve previously looked at options and the only option was going official and get the a placement

There’s no trees on that ridgeline that straddle both sides

[real cedar]

In that case, yeah, just do the power upgrade and filter it. I would suggest getting the asacom

For 30 dB

Gpio labs for less than 24

[white field]

Yeah it’s not my node.

I might get motivated if I run out of hops later

[real cedar]

30 WATT 🤣 I keep doing that

[white field]

I still have 3 nodes to place before I can chat with friends in Moab or Indian Creek efficiently

That might be the final cleanup if nobody else gets to it since it could reduce hop count by -1

[real cedar]

I totally get the feeling. So, what's the plan to get to Moab? I'm actually very curious. And honestly, I've really thought about that and I think that the BBS mesh helps so much more Then I thought... Like, if I'm understanding it, right, you update your BBS node, and if someone wants to check that from somewhere else, like say in Moab, then they can actually be more than seven hops away from you, as long as they're not seven hops away from the BBS node, right?

[real cedar]

What's part of the final cleanup?

[white field]

Hmm, tbd what configs I’ll do. Indian Creek was my winter plan, but climbing season is ending there

Cleanup is filtering NPR so I can delete boulter or whatever intermediary node I need to setup

Which should also make onaqui redundant for certain uses

Still tbd on all this, just a big fun experiment for now

[real cedar]

Oh, I see. So reducing hops. Yeah. That's kind of been my goal. And, you know, as much as I want to put something I've been lovened as a point of pride and because it would be the easiest way to get to Logan in one hop. Having something on promontory point makes that mostly redundant and the easiest way to get to Logan from there is to put something near the Logan radio station on the tremonton high point.

So, I kind of get what you're trying to do. Especially with Francis Pete getting an upgrade here soon I think and that's going to make Ben Lomond Redundant for 90% of the areas, seas, and honestly, it's kind of hard to hide something up there. It would be freaking hilarious. I would love it. I would giggle every time I saw it on the map, but I don't know if it's necessary.

[real cedar]

Yeah, it's a big fun experiment, mostly for me on the academic side of the challenge. And also, because I am going to use meshtastic to do other things, and I have certain areas I want to be bulletproof so that I don't have to worry about the network's functionality interfering with my testing and development. That area stretches from basically north Ogden all the way down to, I would say, right there at Point of the Mountain and over to Toele. Like that area is the core of what I would like bullet proof.

Cottonwood canyons is my moonshot. Nevada is more my watch this shit 😂.

I actually have plans to interrogate a remote satnogs station via mesh and I think that that would be super cool since most of the data that you get from satellites is very low data rate anyways.

For the promontory point location, I think I'm going to be testing A very new design. I have 25 amp hour reliable LTO cells. So 30W new powah, drok buck converter and and 1 of those cells, A rak4361, a 10 dB antenna and a filter.. Now I have an off grid node that not only charges well -60C to 100 C 😂.. It basically will last a year without sunlight hahahhahaha. I wish I had a reliable non femtofox 1 watt setup with a filter but I dont

I'm actually going to use one of the GPIO ADC pins on the rak to read the battery voltage through a two-to-one High Z resistor divider. Then I will tweak the multiplier and the software just a little bit and get the right readings while directly hooked up to the battery So I at least know the battery level as well.

My weird goal is also more for lols. If all of us up and abandoned them or for whatever reason dropped dead. Would they be able to go until their silicon degrades or their panel disintegrates or good chances... Absolutely smelted by lightning

I kinda want mine to have a good chance of being able to just keep on doing there thing without battery replacement or anything ever

[white field]

TBH, my fear is always wind loading on the antenna connectors.

150 mph winds during a winter storm 😬

Onaqui 38fb is running a $5 chinesium grade 5000mha battery pack from Amazon. It never dropped below 93% all winter.

The Mary's Nipple node is currently on a smaller tree/shrub that looked like it's top was sheared off in a wind storm.

[real cedar]

😂 yeah my batteries are all oversized to give me a chance to get up there if the MPPT dies or something. Like sized big enough that for whatever reason it lost solar power and it got slammed never slept it would last 5 months transmitting non stop it should still survive.

Wind loading is definitely a concern, I have that concern for larger solar panels.

[white field]

Yeah, take a look at those photos I just DMed you.

This was apparently able to survive the winter, but let's see how it does in the summer T-storm season.

My nodes all have similarly discrete minimalist placements where I try to reduce visual impact. I suppose that also has a nice side effect of making them less prone to issues from the weather since they stay small.

I'd really to finally establish a link to that one since it's a really neat spot to get temperature readings from. 😂

[real cedar]

Its a really unique spot to get wind speed readings from 🤣

Anemometer will end up in the washes

I'm worried about tstorms. It's why I'm going the non conductive pipe route when not near an existing installation

Under the paint of the solar panel is a rubber insulator paint. Lol. I basically was like if there's no connection to ground. It won't be very attractive

Inb4: Breaking News! The cause of yesterdays BLM fire was lightning striking a home made weather transmitter. 😜🤙

[dense basin]

Did someone get a node rolling on Francis Peak just now? Suddenly I see ZDK2 sporadically tonight with a killer coverage radius centered right up there

[real cedar]

You mean Zed2?

That nodes actually in the weber canyon area. It's caught by my new router on Ensign

If Francis had a new node it would be extremely unlikely that I did not see it

[dense basin]

I wish I could see towards weber canyon but I can't ever see past the AFB - i just raised my house node up another 20 feet today so hoping for something. Also built a bunch of these little guys wanting to sprinkle them along the east side of the mountains here above the shore line but below the "constant snow" line

Actually maybe that zdk2 one is near bountiful peak i guess it could be some dude with a node up there. Just haven't seen that in the limited time I've been on, looks awesome. Also I'm excitedly following your adventures, that Ensign rig is badass

[real cedar]

Is ZDK2 on 51 or 20?

20 is such a disaster

[real cedar]

I would be careful Don't sprinkle too many on channel 51. More routers won't be helping anything. At least for now. Unless you made them a client. But it really would be a waste. I would suggest discrete nodes like that to put them in places where it's hard to reach like campgrounds and stuff or up the inner part of Farmington canyon haha so you don't lose signal in there

Because the higher router nodes on channel 51 do most of the heavy lifting haha

On 51, you should be able to see Hill Air Force Base. You shouldn't be able to see over yet, but I'm working on that. I have a plan to make sure that the entire area in the north can generally get to Salt Lake City in one to two hops. That node is actually next, so hang tight and you might be able to talk past the base on channel 51.

LMFAO

Please be nice to me I help people

🤣

Have you had issues with people taking nodes?

[dense basin]

Oh yeah no way would I make them routers, and I was gonna do channel 20. Already too many routers in my corner of the mesh. They would be clients

[dense basin]

No lol - but the kid in me understands that some might see one and wanna throw rocks at it or hit it with sticks

[dense basin]

That would be awesome!

[dense basin]

20. I didn't even know about 51 I gotta check that out. But yes, it is a mess. Silly question, how do I even get on channel 51?

[dense basin]

Oh frequency slot 51 - ok I see in the search and pinned messages. Gonna try to get on it. But yeah "ROUTER" role where it isn't needed/agreed on is a sin against nature

[real cedar]

Part of me would have laughed if you trolled channel 20 a bit and sprinkled 7 routers on a mountain side 😂 would be funny to see what it does

[real cedar]

So Frequency Slot 51 in your lora config.
I pop onto 20 now and then to see what's up

The two networks have different goals. 51 wants to he a functional long distance minimal hop network. 20s a free for all (a bit)

[dense basin]

Yeah that sounds more like what I want. 20 here in davis county is a wacky situation and often I can't get messages out for all the congestion

It's super inefficient

[real cedar]

Yeah, that's why a lot of us changed which channel we're on. We don't really safeguard it too much or police the airwaves, so to speak, but we have the general agreement with each other that we want a useful network amongst us all for our projects

Actually, to be honest, a lot of repeater networks that span multiple areas tend to do the same thing. It's a local area as its own frequency, and then there is a central repeating backbone that uses a different frequency.

[vital hemlock]

So, for now, if you want to show up on this map you'll need to enable the "okay to mqtt" in your lora config per radio

https://mesh.thebadplace.org

U: slcmesh
P: slcmesh

We have a few radios that are uplinking to a local mqtt. If anyone wants to use that same mqtt server for their own channels I'm glad to share the info. Reminder, your channel's custom key prevents the mqtt server, or anyone else, from decoding your payload. This is however decoding the LongFast channel because it's a known/shared key.