#Ender 5 To Mercury One.1
1 messages · Page 2 of 1
let me know if it does, I'm planning on using mine as well
I think I can use the cheap clone gears because to try to make it better I got a “hollowed out” pom gear, to replace the thick one. I can now say that that combination worked ok. So in my expense the 50t gear is probably crap in the cloned set
cheap-cloned set
everything might turn to spaghetti though… my merc can’t possibly print that fast… I made a plate full of things and I think there are way too many details to draw
popcorn watching the mainsail stream, waiting for a fail at any minute
I didn’t want to go hydra at the beginning, because I had a double z mod. but hydra was hands down the best upgrade I could have done to this printer. I probably would have been happy with the stock kinematic and hydra (of course, printing “slow” at 200mm/s is nice though)
I was waiting for mellow to ship my stuff in November, and I endedup printing hydra parts… than the fysetc hydra bed was in discount….
I got fysetc 50g
50t
I’ll be damned, it works like a charm
Looks sharp, too
hmm fysetc you say I'll have a look
yes, the hydra bed. I can really recommend it
ah, thought you were talking about bmg gears
ah no, burgo mentioned fysetc for the gears, I got all my gears from 3do.dk, or by shady merchants in fb marketplace
3do sell bondtech original gears, since Sweden is just next door
I’m using a cheap set of gears in the rack now, but the 50t is original bmg. the one I got with the cloned set was just incredibly bad
now, if only my kids will leave me alone for a stretch of half hour or so I could attempt to print an astronaut, make a video and get the serial 😬
Good luck! Get that first serial!
do it
it will be lame
why
because I can't edit the video in yt directly as I assumed I could... I took several short ones and I have no clue how to do this 😛
oh there are apps for that on android
I just look a long video and cut out sections
almost there
toddler woke up and I’m sure he’ll make in a way that I’ll miss the final 30 sec to wrap up the video
you don't need the full print
mine was like a 60sec segment of a stacktocat
love to see my custom purge
yep! it’s amazing
done, video processing… but I’ll be too late for 2.1.1
you never know
If only I knew! I've used a software called OpenShot. It's pretty good, it took me only 10 min to figure out what to do (the results is not pretty, but at least it's a thing)
@proud cliff congrats
Seconded!
thanks! I have never done multicolor, but this would look nice
Time for a bambu 😉
one more reason to finish that tradrack
I'm holding printing my serial because of that
yes, I'd have to stop it and change it manually... no way. it's also a nice test print, something to justify building another tool to make something :P.... now I have to build a CNC machine so I can make the missing M5x40mm dowel pin
I bought one from a user there along with some bearings and ecas
it's crazy how dowel pins are impossible to find on amazon in europe. I have to order from US or Ali!
go with ali
things have been arriving quickly
(2 weeks)
I don't believe it thouhg... I'll go on the local hardware store (bauhaus) and browse if there is something like a dowel pins they just don't call it that and make it impossible to find...
same with micro fit 3, I have plenty of jst. If I look for microFit3 kit on amazon it just shows me jst kit instead! 😩
I have dupont, despite me hating it, so I can skip the microfit for now. A kit with 480 microFit3 pieces cost 50 euros? is that expansive?
I would use jst but they really require the 4 connectors 2x2 not 4x1
@frail swallow I feel you are a microFit3 enthusiast. Are they so much better than regular jst-sm?
Yes and no. 😄
To clarify: they come in larger combinations, are more compact, and (I think) just as easy to crimp.
Comparisons:
- JST SM gives a more confident "click" connection, Micro Fit gives a higher quality connection.
- JST SM is limited to 3A, Micro Fit goes to 5A
- JST SM gets really bulky once you get past 4 pins; a 12-pin Micro Fit is the same size.
- JST SM is dirt cheap. Micro Fit is expensive unless you go Ali and then it's more affordable.
I have burned through 50e on 2.5mm² silicone wire just to not need it now with a 230v bed, along with a lot of microfit those things are expensive
References for you: https://jon-harper.github.io/clockmaker-project/wiring/connectors/#molex-micro-fit-30
vs
https://jon-harper.github.io/clockmaker-project/wiring/connectors/#jst-sm
Documentation for The Clockmaker Project.
also @frail swallow finally used the iwiss to crimp, that thing is awesome
I see. I have plenty of problems with the connection quality, but that's due to my questionable crimping.... but after 2459 crimps to build the merc I'm getting a hold on it
Right! It's like my bestie now.
Only 2459? I think you missed a few in there. 😉
I have my array of tools to crimp
scissor to remove the eccess of metal flags (the one holding on the silicon/sleeve), a plier to quickly pre crimp the wire on the sleeve, and than very carefully crimp the actual thing...
I do the sleeve side second
Uh oh. FIGHT!
aligning 8 wires in the correct order is painful
Yes. Yes yes. More yes.
after a few dozens you get the hang and almost make them blindfolded
I don't want to deal with the tiny thing, so I trust my crimping tool will push the metal in its place and make it work
Generic crimper or a JST-SM one?
interesting I find it easier to hold it with the sleeve side open
you know... I don't know 😅 .. I use it for all the crimps (a side spade/ring and big stuff)
wait a pic... it's a cheap one
Just curious what you use. I love my Iwiss, even though I bought a JST SM crimper.
this
I have to be careful how much I press on the small one, yeah I know I’m doing it wrong: so first crimp is in the middle teeth, than I pass it again on the first teeth but without going all the way. so I know it was tight. check any white stain on the sleeve, give a good pull, if all ok pass to the next
I use PA24 for my molex
pro stuff, I got mine just because it was bundled with a jst kit years ago… s it must worth nothing
I hate the ratchet crimpers for jst/dupont
Really I love them. We’ll just the first click. Then useless Helps to hold in place
i wish there was a jig you could prompt you wire, the connector to crimp, and it automatically does the job 😅. I have sausage fingers and I’m half blind and it feels every time I do a small change there is something to crimp again, and again 😅
define blind
I have -14 dioptre for eye… sound bad but it’s only if I remove my contacts
issue is that the lenses are so powerful that I struggle when I close up… without lenses I can only see close up 😝
I know the feeling, -9 something here
the trick is to use the glasses in those situations
perfect vision close up
I guess after the -6 is all the same blurr
heh it gets worse
I have been slowly increasing
once met a guy with -21 he was like a mole without glasses
yes, I can count the fingers of an ant close up… veery close up though
you can always invest in a magnifier
I was at -10ish for a while, than I got a weird issue with my cornea, being conical and not spherical.. and it made it be a little worse… but I have also free special lenses because of that… made it a huge difference vs the soft lenses I used before
I want free lenses
my left one got a damn fungus eating it right at the begining of covid due to damn confinements there were nearly no doctors, could have lost an eye 😑
I can still see but with some slight degradation that can't be fixed with lenses
Anyone up for some voice?
not atm
😮 posted in the wrong channel lol
wow, that sucks!! I had an issue with my retina in February.. I started to see much more floaters, very big and dark… I had a bleeding in the retina. no detachment but still sucks. I can still see them now.. doctors say not to lift weight and stay calm…… I have 2 kids 5 and 1.5y. it’s impossible to stay calm and not lift weights 😂
completely the kids themselves are one, and the kind that asks you to lift it
and I think they hit me in the eye/head that’s why I had that mess to start with… kids are the worst, I got 10y older in the last one
have a similar issue, the right one sometimes gets too dry during the night and when I wake the cornea is stuck to the palpebrae the result is a nice corneal abrasion, this started after mine hit me in the eye, not sure if it's the source but the timing matches
for any CANbus users, which wire should I purchase?
I think I should go with 20awg wires because my hotend uses 115W, but I couldn't find any twisted canbus wires at all on aliexpress
ah wrong chanel for me too :V
Rapido only pulls 115w at a very cold temp, once it hits 70c its only pulling around 70 watts. The hotter it gets the less power it draws.
But I run 20 for power amd 24 for signal
did you make your own cable?
also I limited the rapido power to 60%, and it’s mire than enough. it’s super fast
yes
I use 20 for power and 26 for signal… I actually have some random error during homing, But I don’t think is due to the wire size 🤔
I use 20awg for power, 24 for signal. Anything will work for the signal pair, just make sure it's crimpable. I would stick with 20awg for the power, though.
timeout error?
all I'm worried about is the shielding, it seems quite important based on the search results so
if you're having no problem and using fine, lmk
I have two setups with CAN, both using 20 and 24 pairs. Works fine for me.
Shielding = twisted! Make sure your data pairs are twisted and power pairs twisted
just normal wires?
is being twisted enough
I was talking about foil/meshs whithin the wire so
yep. That's all phone cable is. Twisted pairs.
Ahh, no you don't need that kind of shielding integrity, though it won't hurt anything. Just not needed.
nice to know that
so 20/24awg pair?
I'm looking forward to use chc pro/ 2 5015/ led so..
Yep. As long as it's got a good twist, it should be fine.
I still think it would be enough but I want to be safe
If it helps, here is a visual of the twist rate I aim for.
wait wha-
why did you twist all the cables?
oh and
can you recommend any connector sets?
I do have a dupont/terminal crimping set but I have no idea where to source those connectors for motor/thermister etc
It keeps them together and generally reduces noise.
the terminal on the board are usually just jst
There's lots of connector types out there. You'll at least need JST XH for board and such.
JST SM is a good, simple intermediary connector. Micro Fit 3 if you're feeling fancy. Here's a list I've made of most connectors you'll possibly run across: https://jon-harper.github.io/clockmaker-project/wiring/connectors/
Documentation for The Clockmaker Project.
and btw
are terminals /connectors required inclueded in ebb36 set?
@frail swallow @proud cliff
Yes, plus spare pins
I was pleasantly surprised.
oh wait
well thats good
I don't need to buy em then
xD
btw my bearings arrived
so time to start building
and they are stiff af
ono
I got a thr36, they didn't included anything extra, barely the strict neccessary (there is also a price difference with btt :P)
linear rail bearings?
yep, that's correct. 🙂
but you probably don't. Some IPA on the rods and cleanup, the relube with sewing machine oil (or equivalent). Just run the bearings up and down the rod while it has IPA on it to get work the fluid through the bearings.
There is zero need to limit the power, as I mentioned above the power draw drops off pretty quick as it heats up. I was concerned about its draw as well until I saw the chart.
lol, just remember to relube and you'll be good.
noooo. lily white sewing machine oil is best.
wd 40 is not a good lubricant for machines
FYI: be careful with IPA in the ultrasonic. Generally safest to put IPA in a bag in water in case of a short.
Yep, same stuff, I think. It's marketed as that lily white sewing machine oil in the US.
Or at least it is commercially.
should I soak the bearings overnight or sth
nah. you're super clean by now, lol
I did buy the cheapest bearing I could find so
I noticed that I limited the power when I tried another hotend... so hard to heat up, than I noticed the power was maxing out to 0.6.... but the rapido even with 0.6 max, it heats up in no time, I just keep it for peace of mind, and so I don't have to pid tune again
btw what's the rapido's rated wattage?
It's silly that I hate taking the time to PID tune, but I do.
110W peak, but they're discussing how quickly it falls off during heat-up.
tbh, not sure. We should probably dip out of here and into #general-chat or another channel, though. This is technically @proud cliff's build log. 🙂
true xD
np… no much building to log lately
Just being polite. 🙂
maybe that chart will change your mind
only takes a few seconds to hit 50c, then its pulling the same power as most hotends, or less
3A at 50C that's super fine.
also says no firmware power limiting
Now I want a Rapido, lol.
so I’ll remove the 0.6 and pid tune again 😝… After trying the nf crazy the rapido gained my respect. I just waiting for a new heatsink body… currently running with 2 screws out of 3 (I destroyed 1 m2.5 hole trying to remove a screw that broke into it
Phaetus used to have that image on their website, but they removed it from the rapido product page for some reason. But Fabreeko still has it.
strange that they removed it, cause 115w to a hotend is a lot of power... seeing this chart puts your mind at ease.
hi, what endstop are you using on your x? the gray piece?
i removed the ender 5 original endstop from its small pcb and soldered the wires in the 2 more external pins. the zeroG eva carriage has a slot for it, it sits just on top of the mgn12h
the gray piece is basically just a rail stop with a flat wall to make the endstop clik properly
it’s all stock as suggested from the bom
also I did remove the endstop from the pcb because I have a bag of those for spares, back in the ender 5 days. If I had to buy new I’d take those sold without board
and you know, you might have solved my poor flow situation. the extruder was skipping steps after a little bit after reaching ~20mm3 for a little bit. now it seems better. I have to test this, but my theory is that obviously pushing more cold plastic in, the hotend needs more power to keep up. being the power limited to 60%, it can’t stay hot enough for long to reliably push this fast the plastic out
almost for sure, the limits are usually because of that
hey @proud cliff I guess you're missing a washer at those X/Y M5×8 screws which is holding the extrusion, can you confirm?
the CAD model has a washer there. I used lock washers, though.
there was no washer in previous versions, so I never adjusted. the m5 still clamp the extrusion pretty well. it might be safer to have it though
The washer is just too prevent the bolt from bottoming out in the 2020
I wouldn't use a lock washer on printed parts though.
I mean, I can add the 1mm thick m5 washer I used for the flanges…
test fitting the side panel. I just got around to glue some of the spacers, I miss the spacers for the Y extrusions, but I have to cutout some for the y endstop, and I prefer to be really really precise on those, not to interfere with the gantry movement. Next install the side panels to sign where to send, and take measure for the back panel
I just have a jigsaw, don’t judge the shaky cuts 😅
for a jigsaw
its great
I can only get straight cuts with a tablesaw or a circular one with a track to follow
round perfect stuff only with a router and template 😆
you might want to get a blade with thinner teeth maybe
Yeah, I am doing a rough experiment if this works and how the input shaper/noise level are going to be, if it’s worth refining I’ll use the panels (a spare insulation panel I had in the shed for years) as a template of some nicer materials
my first idea was to take some thicker wood and use a router to dig the track for the rails of the y axis…. but I don’t have a router, and also never used one so….
All fun and games till it catches.
1st quick attempt to install a side panel. I started with making 2 holes bottom right and top left, square the panel to the frame, sign and make the other 2 corners. it did go half bad; first 2 holes spot on, than I managed to not make a hole where I signed and so i have to re-attempt. But I got the idea how hard it would be to install-remove each panel. it’s not easy
also there is no guarantee that the t nuts installed, wood on aluminium, will not crack a little, making the frame a little lose. I’ll add a greenhouse gasket between the wood and the frame, but I should install washers on the bolts heads
washers on screw heads should be enough
so, I was bugged because the hydra chain was always snapped off the printed when lowering z too much. it turns out, my chain was too long. I’ve removed 4 pieces (I was planning for 2 but to my surprise -4 was still long enough), and now it seems it doesn’t snap off when lowering z
But still, it should actually stay straight, without banding to the side at all. I remember the original pic from @zinc needle, the chain wasn’t sagging at all. Should I print the chain latching with other settings/material?
If you want the quick fix you can glue it
I need to stop to “work from home”. setup the panels (didn’t finish yet). The goal was to damp the noise, not make a drum box. it’s actually louder now, but I have to add the gasket betwwen the frame and the wood (it's not really wood, sorry wood) and fill with noise absorbing material (and maybe sand to round the corners to make it look nicer)
the back, front and top will be covered also, so maybe the noise will go once fully enclosed
it's basically an enclosure project on the opposite spectrum of @frail swallow one, where he carefully plan and draw in the cad I cut 2 panels with a jigsaw, bolt it to the frame and sandpaper it to measure 😂
Umm your bed wires are coming out the wrong side of the chain/bracket
all it needs is some paint
Oh, I see what TC is talking about. That's why it's breaking when Z gets low. Fixed end should be the bottom.
I just swapped it this way, before they came out from the top of the arm. I feel getting them out from the bottom gives more strain relief, especially on the sexbolt jst connector. but I can totally swap it back to make it go out from the "top" 🤔
ohhh... thank you!!
Every wire should exit through the chain
Do not run wires out of the chain like that
that's why. I had the chain with the end swapped before, I basically just changed the connector and not the wires path
funny that I printed that thing twice, I though I had it printed wrong/too weak
that was an easy fix
much cleaner even
Chain looks to short?
probably, I just removed 4 links 😂
Or maybe your wires are just too stiff for the chain
I actually wanted to open the hydra step file and see if the chain was pictured if/then counts how many links there are
add some links
probably not this link 🤦♂️
if I don’t update the log for a while, you know what happened
no not that one
GOOD game
I’ve seen trolling reviews complaining that it was the same as BOTW… I don’t know what’s wrong with that!! plus, it seems even better
it feels familiar, but not the same honesty
After measuring how to deal with the back plate, I am unsure what to do. My original plan was to use 1 single piece ~420x800, and cut out some squares for the motor and bold everything to the frame and the side panels. But, the motors would be half in half out... Maybe I can rip off @frail swallow (sorry I keep dragging you in my log) solution how to enclose out the motors, and mix it in with my side panels 🤔
@proud cliff you can either add extrusions behind and pull the motors out so they're completely out (with some gaps needed for belts) or move them in and lose travel
there's a mod for that
Drag me in all you want, dude. 🙂
What I did was like @stone valve's first suggestion: added another 2020 extrusion to the rear and mount the steppers to that. But I would recommend using 40mm longer extrusions for the Y axis (where the Y linear rails go) and moving it out a little further to avoid the wall notches I had to create (see picture).
I can actually give you a Misumi part code that includes the blind joints if you want to order them. Just lmk.
Apologies, meetings all day 😩. I see. My goal with this build is to have the lowest possible adds on on the stock M1.1 setup, waiting for the bom for a completely new frame when it will be release 😅
Ahhh, I getcha
hmm, looking cheap (~40€)!! 5 tmc drivers. I think I can still use usb to can bridge
I also like the large heatsink
not a bad price at all
there are no neopixel pins, but I’m reading you could use the bltouch pins. there are 3 thermistors, the wifi, I have no idea if it can be used. only 2 fans though… They really implemented the basic and not overdoing. it would be perfect for hydra + canboard on the toolhead. I’m really tempted
2 fans if you're going with a toolboard is enough, thermistors are more than enough
wifi I think is more a duet thing not sure how it interacts with klipper
I wonder what "reserved I/O" means. it means they are there but I can't use it? If I buy this board it all depends if I can use PD0/PD1 😄
let me have a look
on the schematic I can only find io1
and pins don't match those
am I reading the right onw
PD0 and PD1 are linked to TMC_SCLK and TMC_MISO. I don't know if you will succeed in running CAN over these pins.
Come on dude, there is absolutely no reason to buy this 2 years old board. It's bottom of the barrel.
I see that in the robin v3 the pins are used for the tmc uart drivers. Well I like that the board is bare-bone, I don’t like the idea of having a huge board with 8 drivers and using only 5 😅
actually the perfect board is actually out there, the btt skrat seems ti check all the boxes. Only thing I really like the big heatsink of the skr pico and skr mini v3. And this eagle board seems to have something similar with an additional driver
Over heard mainly bad things about the makerbase boards. But have been happy with the thr36 so far
it’s probably true, the price difference might be telling… And I know they do take the risk to over packaging their items with wraps or extra goodies 😂
just the board in an unbranded cardboard box
waiting for this, 20€ (8€ are just for the shipping only… dk to dk 🤦♂️) for 3m of cable. I am not sure it will be ok
I am a bit confused about the 120 Ohm description
there was also the 2x2xawg20 version, slightly more expansive, but I think it might be too thick. 3A rating on the toolhead should be enough, also awg22 is what I have now and it’s working fine
120ohm is the resistors you need on the can ends IIRC
yep I would say the wire includes resistor
On most CAN boards you can disable the resistor if needed
Absolutely not. 120 Ohm is the cable's rated impedance. Due to the relatively high frequency of the CAN signals, they resemble more an RF signal (or alternating signal) rather than plain DC voltage, and is affected by resistance, inductance and capacitance. It is recommended to match the impedance of the cable to that of the circuit.
Trust me - If you measure resistance with a multimeter you won't see 120 Ohms, you will see infinity for open line and very low resistance on short circuit.
Btw, an interesting tidbit for those of you who like to chuck their wires on a cordless drill and twist it - this is a sure way to alter the impedance, for example of a stepper motor harness, which doesn't really work the same as a DC motor principle.
How much this affects its operation - I cannot tell but I'm sure there's a study that can be found somewhere.
honestly if it's to be twisted I think it's better to find them pre twisted
Indeed, because you don't really know what's the correct TPI (twists per inch) a cable requires to reach a certain impedance, and it isn't trivial to measure
but then by observation even bought pre twisted might not be consistent
A CAT-5E cable, for example, has different amounts of twists on each pair, but still within a 15 Ohm tolerance. Break the cable's impedance and you can only dream about reaching 1Gbit over 100 meters.
CAT-5 impedance is 100 ohm, btw. Not very far from the recommended 120 of CAN,
some pairs have tighter twists
With that said, I'm using non twisted cables on my CAN stuff and it works. The same as I never bought a $200 RCA monster cable when home theater started popping up, and strangely, I could still hear sylvester stallone's weapons quite well 😆
The analogy is about the same, just so you get the idea..
In an industrial capacity I wouldn't even dare thinking about using non-preterminated and rated cables, but for hobby grade - I'd even attempt to run it over cotton string with two plastic cups, just for the sake of experiment.
For the stepper motor I use silicone wires, and they don’t tolerate drill twisting, so I did only some loose twist on those. the can wire on the other hand it’s pretty tight. I checked the can network after my usual cans errors and I can’t see any obvious culprit, no errors in the network, all the mcu received/sent some data. But I did noticed that the error occurs always at a specific corner of the bed, so I think it’s a simple wiring issue (given how hard it is to put wires in the thr36 terminals and my cable is not stiff/compact enough to have a nice strain relief)
ok, but the wire I got seems still pretty cheap compared to other I’ve seen. 8€ is the shipping, so 12€ for 3m of wire is ok
I am convinced, based on the experience I have while tinkering, as well as general knowledge in electronics, that most CAN bus related woes people experience are not due to incorrect cable being used, but rather due to connection and termination issues.
Klipper isn't built to tolerate packet losses on the CAN network, at least not in a way we are used to when we deal with telecommunication in our daily lives. If the internet worked that way we could not have communicate at all. You'd never be able to watch a movie on youtube or have a cellular call without it dropping.
I was not meaning to judge your choices and expenditures - this is your hard earned money and you have 100% entitlement of using it the way you want, without ever notifying me 😁
If I was perceived that way - I do apologize.
totally agree, I think my issue is some wire in the terminal on the can board looses connection when the cable is stretched on the corner further away
not at all, I wasn’t really looking for canbus wires, I was more looking into a ready-made cable sturdy and polished enough and nice to look for the toolhead
Something similar happens to me sometimes with the reverse bowden. However, for this comparison, when there's a filament starvation you can observe a slight under-extrusion... But when the CAN signal (or power..) is disrupted, more often then not, klipper will completely shutdown and leave you with frustration.
From this aspect, the XT30+2 connector performs much more reliably that the Microfit 3, even with barely any strain relief.
Had you used a board with a microfit terminal, I would advised you to purchase a genuine Molex overmolded cable.
I wish, I have the thr36 connector, every time I want to plug l/h/+- (it happened twice si far) I spend 2 hours wasting a lot of things and time
I tell myself next time I desolder the thing and install a new terminal, but I always forget to look the specific and buy the connector
These are screw terminals with 2.5mm pitch, correct?
That's exactly what I want to suggest 😀
I don't hate the screw terminals, they are just too small for proper sized ferrules
I think it’s more than 2.5.. huu they have a step file, maybe I can check it without tearing down my printer
maybe I’ll insert the plain wire next time, if it disconnects anyway…
I don't think mine has had a disconnect yet
I might buy a spare before doing so. I’ve seen the ebb are in discount
Can't recall if I ended up using ferrules or bare wire
overall I’m really happy about the thr, if the process passes the tilt adjust it never failed so far
I suggest you definitely stock up on spares these boards. This thing is way to important to have no spares - my opinion..
Plus... they are designed as mildly disposable/self destructing after a while, so...
yes, and at a specific corner, the 3rd probing
What do you see on the CAN MCU's bytes_retransmit and bytes_invalid?
When I first setup can I used a can hat. Had to run my klicky to the main board. The can hat couldn't handle the data
I tried 500k, than I adjusted the klipper code to increase the timeout, it seemed to have fixed it. Than I’ve increased the baudrate to 1M, and the issue came up again
Why increase? Expecting better performance or print quality?
I've never heard of someone running that high
I’ll note thise next time I run it
I have no idea, I just like to test the limits 😂
Maybe you should go back to 500k since that was working?
The cats over at the voron community claim 1Mbit is recommended. That's a total BS. I run 4 CAN devices over 250K, albeit without attempting to run input shaper (I know it'll fail since not only the toolhead's ADXL is over CAN, but also the 2 motors)
yeah, I want to see if IS is better at that frequency, it might be the only thing that’s helped by higher frequency
IS is better when the accelerometer is strategically placed and runs over SPI or USB. The accelerometer in all of these boards is just a novelty.
Why would input shaper improve? Can is only on the tool head?
Or are you using the on board adxl for some reason
I tested the adxl on the thr vs the adxl on a usb board placed in the nozzle. the results were similar (very bad in both cases) so when I’ll finish the work on my printer I want to test it again, and see if there is any difference between the two boards
500k was already “working” meaning the test completed. I don’t know if by increasing to 1M I get better accuracy or what not
it debatable if I would ever finish to work on my printer though 😂
I felt asleep and I didn’t noticed the back warping 🤦♂️, a lot of wasted filament on a otherwise nice print
Sucks, I hate wasting filament
maybe also the front warped, by a fraction of mm. it’s better to try to print this on a proper enclosure.
my top chamber temperature was ~34C. So not crazy. But after I have NOT printed for few days I’m also having some bed adhesion issue. the auto offset resulted in a -[2.59~2.62] rage, while it used to be below -2.65. It might be different weather conditions (we are enjoying summer here), but after the bed is at 100C, I don’t think env humidity and pressure might influence it
curiously we're supposed to be in summer mode and yet we're expecting a storm
Don't forget to wash the bed my latest adhesion issues have been due to that
we had few days of full blown summer, it’s 24C now, it doesn’t sound much, but here is alot
I washed it, and it improved, but still warped. I was thinking to give the CF side another try, maybe slightly sand it. I know @nimble lion gave his a gentle scratch and improved adhesion?
I have seen people recommend a slight scratch even on PEI with the of those dishwashing sponges (the abrasive usually green side)
@stone valve, completely OT, do you also have an electric car? I remember you talking about solar panels charging your car 😅
yup
Nice! we were considering adding some panels for charging, but it would not be very convenient in the short term (2 cars all electric here, 41 and 75 kWh)
for us works in the summer because it's just 1 car and it's at home during most days
in case of a long trip we eat from the grid up to 50 or 60% (75kWh battery) and the next 2 days solar filla to 80
4kWh of panels for reference
also you need some kind of automation to start stop charging when you have solar excess
feel free to DM if you want to exchange some ideas
@proud cliff not sure if you're still looking at that maker base board but this is relevant #electronics message
Aka wifi won't work
thanks, I am looking those wifi slots more as pins to use for something else rather than actually use wifi :). I think I made peace with myself that my setup works as is, I’ll possibly swap the board around, since they are connected via can I can use the board driving AB upper in the build, eg beiind the tophat enclosure: use the extra thermistors for getting A/B motors temps as well use the 2 spare drivers for the tradrack mmu
so I’ll have more free space in the bay (my primary goal when looking fa or a new board) and simplify the wiring
I was adamant to waste more filament and not wait for a better enclosure (also I think I have to revisit the thermistor I use to get chamber temp… later). So I sliced increasing the temp to 265 (and later to 275 on mainsail, you can clearly see when I did that by the discoloration), increase fan to 40% and lower the switch offset by 0.01mm
at least no warping
so now I don’t really know what did the trick, but I think the higher temperature (265 was enough) might be the thing that avoided warping
Still warped at the back under the cable chain. But should work just fine
Good looking print other than that small lift
yes, I also noticed in the picture, but I can hardly see it on the piece, I wonder why this particular shape tends to warp so much
also, @solar bridge , am I too close now?
it is at the limit of a “nice” first layer when I print a square, I can get nicer looking layers if I lift the nozzle, but than some lines don’t stick on the bed during a print (eg the skirt might get loose …), it’s a bit of an hit or miss. Also I’ve noticed a much stronger adhesion within layer when slightly squashing it: If I tear the square by hand I have clear tears, squishing in it need much more strength and it tears not on the filament path
ah! right I almost destroyed my cf side playing with z tuning yesterday 😩
not almost, I really dent the thing badly
I wanted to fine tune, reset by mistake the offset on mainsail, than on the console messed up the calculation by 1 full mm too much
Opps
I just baby step using the zoffset box
And save if I want to
Let mainsail do the math update
maybe I'll try to add this in the inside of the panels, and see if they dampen the noise and vibrations
rubber tiles, 5€ each
If not I am sure I can use some rubber tiles in the garden... They should also be rated for some sort of heat, in the summer under the sun must be over 50c, so I think they are safe to use
Though on the cabinets: I am still not sure if my cabinet will be good enough. I designed it for the ender 5 moderate vibration and the stiff-but flexible work plate worked to absorb some printer vibrations
but with the Mercury One type of vibrations it's a totally different music, it doesn't seem to move that much when it's in the ground, but when on my cabinets it manages to move everything, So I've stiffened up the table plane, added a massive concrete tile, and I was thinking now to add those rubber tiles as well (under the concrete). I hope this will do
@proud cliff have a look at the feet I found and posted on my log, I'm considering giving them a try
Might work for me as I'm not using the electronics part of M1 but need to raise it because of hydra
I’ve seen them, they look very nice. The hydra feet do a very good job on absorbing the vibration, but for this to work the work plane need to be solid. on the floor it works amazingly. If I could leave the printer open air on the floor I would just do that
but I can’t 😓
Bolting the table to the wall?
I ended up with a very stable table
Now the movement is from the damn enclosure
my walls are not that solid, I need to catch a beam to have the more stable setup, I’ll get to it if all others attempts fails 😫
wood walls?
in the meantime, v1 of sound proofing side panels is ongoing, hopefully the spacers installed like that will further help absorbing any noise from inside the printer
and the gasket on the side will prevent/mitigate the frame vibrations
wood house 😅
I wanted to go with 2 lines of gasket on each side, but let’s see
I think the drywalls are the fist layer, the walls of these houses are multilayers, and finally the beams structure behind, I’ve managed to install sinks and wall mounted cabinets so far, but the printer will actively pull
yeah you will need beams for those
it needed a cutout for the motor mount, and I countersunk the holes (so the screws could easily reach the t-nuts, since the gasket adds some bulk) but it seems to work!!
I ended up using 4 M5x30mm. that’s the easiest fit for now. it’s really solid when shaking. I tried more holes but given that this is not a precision job, it gets longer to catch all the t-nuts
to set the rubber tile on the wood panel I used silicone, and it didn’t worked very very well (I think if I pull most of the tile detaches). For the other side I tried wood glue, and it seems to give a firmer grip on the panel. On the top of the printer I’ll not take any chances that the rubber tile will detach and fell on the toolhead. So I’ll glue it, wrap on a cloth and bolt the cloth on the wood from all sides (it will also look nicer than the plain rubber) and finally bolt trough the rubber and cloth to set the horizontal top beams of the enclosure (that will press on the rubber tile)
I don’t want to waste that many fasteners for the rubber, it’s really a strong material, but it get flimsy when you brake it (holes/cuts). plus, I feel if I fix the rubber that way I’ll somehow transmit more vibration to the wood panel
but if anything, I’m increasing the mass of the printer, and that alone will decrease the vibrations
it’s heavy! I can hardly move it around
indeed, I still need to fix things (a panel has a loose corner, how much I press I can't engage the M5 t nut) and need to install a back and top panel, but I never had a Y resonance as clean as this (I know it's not perfect yet)
so this was my last IS in mid march:
yes I did change to 1.1.5 and maybe tightened things, but I also added more stuff that could vibrate, such as the nozzle brush and the nevermore, and maybe I've squared the frame a little, but I haven't done anything specific to improve the resonance yet
comparing old and new Y resonance, I can see that all the old peaks are there, but they are really tuned down by the mass of the panels on the frame (must be), and the real Y resonance at 43Hz stands out now
I need to be careful to mount the back panel a bit further away: if I install it in line with the rear 2020 I’ll sacrifice some Y movement
Oh, my yes. That's close.
sure, it’s from @gleaming trout than modded the one from the 5+
mine being on the bed is not helping
It’s posted in #1007998760967737364
yep, rapido and crazy have similar if not the same final z height… maybe the crazy goes a bit more down, but still works with this brush
perfect
#1108320860915388446 message
so far yes, I got one for 70€ from triangle lab… apparently the discount was because all 3 Ti screws that secured the heat block where rotten/rusted, and 2 out of 3 broke before I could swap any nozzle…. I should have opened and swap with new ones… but even if crippled, it a great hotend
I prefer the rapido to the crazy. the crazy is a order of magnitude more sturdy though, the rapido prints better and it’s faaast
so satisfying when things click when you press them into their place 😆
I love WAGO nut holders
STOP calling them nuts
chips?
wago connectors
I feel the resonance analysis can be more useful. I mean, thinking about it, it should be possible to trace the aetiology of a specific odd peak to wiring/nuts/loose frame, right? given that the frame is kind-of the same for all our printers, the resonance of similar size object should have the same frequency
there is a macro script or something to make the printer vibrate with a certain frequency
Wago Things That Shall Not Be Named
boooring
Too many syllables. I get confused.
I was thinking about putting an ADXL on my print bed away from the supported edge. Always wondered how much the bed jumps/moves at high accelerations.
true :), for the stock bed it would be nice to know.
it’s a tank, 500 mm/s and 30k acceleration
that filament path really bothers me
it’s “temporary”
its been temp for to long lol
at least run a reverse bowden tube
and stop the filament from just getting slapped around
I did! but I had some cheap pla that wasn’t sliding nicely in, so I removed it, it’s somewhere on the floor now
but I’m building a tradrack, so hopefully in a week or so I’ll have that in top of the printer, serving one or more rolls
@proud cliff 3ID tube
so 3mm internal diameter?
yep
and how do you fix it to the extruder?
3x4mm tubing is pretty standard for reverse bowden
I was running 2x4 but filament got stuck a lot 500+500 tubes
I'm running 2x4 inside the enclosure 3x4 out
I find it easier to load into the sherpa with 2ID
right, I used some capricorn tube, it’s 1.9 or 2 I think… that’s why it was working so poorly
capricorn slides quite nicely though I had issues with the transparent one
need a quick favour can you measure the top front extrusion?
so I’ll get some 3mm id ptfe 😓… reverse bowden and also for the 5 spools (I think I’ll ever use 2 maybe 3… )
got 5m
so I can’t even start counting the meters needed
link?
sec its ali
so another month of waiting 😂
yes, but DK is the last butt hole of eu when shipping from china
capricorn is tighter than normal 2x4, to tight for reverse bowden
I have been using it with great success
man Portugal is not a lot better
would you use it for the small piece needed between the extruder and the hotend?
instead of regular ptfe
I have a very bad flow (max 18mm3/s) maybe it’s that 🫤
@proud cliff can you measure extrusion?
ok and I see you’ve set 30mm3/ in your speed profile
I just set it higher than my hotend can flow for stupid speed benchies
so it’s not that my issue.. are you using a cht nozzle?
cause flow would limit speed
I'm loving transparent one, not as good but I like to see the filament
normal quality speed profile has a proper value
I do and run the reverse bowden with it
yeah, I set it slightly less than what the hotend can handle
I guess so, there is a flow test, but my half ass test is on mainsail, I extrude 50mm increasing the flow until it skips and doesn’t extrude 50mm anymore
ah that test
you change the federate, and it estimates the flow below
I use orca slicer to test flow now
I understand the choice of installing the motor further back, this rear belt/motor tower is a very busy place for making a nice clean enclosure 🫤
but I am not doing that
elaborating an action plan
- install a panel in the back from bottom to top extruder, the rubber tile will be set in a way to absorb the vibration from the motor towers.
- install the wire guide that was on the central 400mm 2020 on the panel (I will remove a/b cable and install the new canbus cable I got)
- find out how to cutout a panel that sets around the a/b motors and yet bolt on the side panel (this top part might not have enough room for rubber insulation all over the place)
- move a skr mini right behind the a/b motors, in an alcove that will create by the awkward shape of this rear tophat. connect it via the canbus that’s already passing by (the skr mini will serve the a/b motors as well as 2 motors, 2 endstops and 1 servo for the tradrack
of course instead of focusing and finish I slice benchies: testing turtle speed profile I get this
this is already twice as fast that I've ever printed a benchy, and I doubt it would just do it if I send this, but I wonder if I could do it faster, this is already 40k accel and 600 speed
and my max 19mm3 flow will make a mess when it's mostly needed a 30mm3 flow
wow, it feels a bit excessive for a nozzle and a gear...
at the same price I got a full hotend and a clone cht… let’s see
I have found a eva 2.4 mount for the cloned bambu hotend, and will compare it to the rapido and mosquito… I have to stop to zip tight my hotend wires 😂, it starts to be environment unfriendly
Recycle the nylon and print new zip ties?
I did spend the money on one (before clones), no regrets so far might buy another but .6
I will eventually get one, I got the clone primarily to compare rapido and nf crazy with the cht-like vs the cloned bambu hotend which came with a cht-like nozzle
21.37 euro for a BMG gear? Yikes.
That was the cost for the whole Sherpa mini gear set
Extruder gears, the big one, shafts and bearings
yeah, the other that makes this is from mellow, and surprisingly it’s not cheaper
@proud cliff trianglelab also has IIRC
What about buying a BMG clone and using the parts from it?
that’s an integreated gear with the shaft, so you don’t have a grab screw. the deal is that the grab screw slightly pushes the gear off center and when it spins performs an uneven force to grab and pull the filament… which it’s noticeable in the print quality (eg vs Bowden tube prints)
Gotcha
maybe the lgx and orbiter don’t have this problem at all
I still use a BMG, lol. No idea. 🙂
So enclosed half of the back. it doesn’t feel a particularly precise work, it’s actually not precise at all, bur I don’t have any serious tool, so we’ll see if I do more damage than good
I made layers of packing foam and covered in the interior with some insulation plastic sheet. it’s easy to cut with a knife, but it’s otherwise hard to break.. (I found this in the discount cuts pieces in the diy shop, don’t know the name)
the top part is unfinished, for now I just cutout something around the motors
the plan is to add some thicker spacer below the motors and on top. this will create a space for a mini electric bay just behind the motors, where I’ll install one if the skr mini e3. moving it out from the electronic bay. this board will primarily drive the a/b motors. I am also looking into move in the “top board” some endstop and fans. but in this board I need at least 2 endstops and 2 motors for the mmc unit that I’m building a side
a little bit more details of this wip unrolling abomination. luckily my life doesn’t depend on woodworking 😅
now the side panels are more stiff also on the top part, where no aluminium extrusion is present
a close up on the rear part, this thing really attracts dust, maybe the black, maybe the material… anyway I’m planning the back “cabinets” so hopefully I’ll be able to be more precise here
it also needs to be accessible if I need to rewire things, so I think there will be hinges and a locking mechanism involved. I think for simplicity all the back panel would slide up and down, so I could also reach motors/towers/belt if anything needs intervention
That makes a lot of sense
also planning a door. I have a 5mm plexyglass sheet, which I was planning to cut to size and add magnets all around, so a detachable magnetic door. the reason for this is that I don’t have space in the cabinets for a door that slide opens to the outside, I can detach it and set it aside (maybe add some features that I can attach the door on the side if the printer, if it remains smooth, no roll holders or similar)
my initial plan was a foldable magnetic door, that looked like a stained glass artwork, where the frame would feature the zeroG logo in various sizes, and it could be folded in a zeroG logo itself (brilliant idea while walking with the dogs around). I then realised I don’t have the slightest idea how to do it nor the technical skills 😂
on a technical note, I have to redo the back panel step: with the sides only I was able to achieve a good z-tilt (meaning at every cycle the adjustment was working progressively), after the back panel i can’t get a better range than .16mm. I tried spinning the printer around as it usually I can find an orientation where the floor gets me a more favourable tilt
the hypotheses are that either the back panel stiffen the frame, but it also slightly unsquared it, the other is that it’s just what happens when you stiffen an underlying unsquared frame
also, it’s possible that my floor is actually terrible, and the weaker frame was able to adjust more than now. So first thing I’ll try other floors, I have a concrete tile that might be a good test, so before tearing down things I make sure the fix requires intervention on the printer
the tilt issue was the pei sheet was slightly on a build-plate-centering screw… it’s not the first time that happens 🤦♂️
5mm plexiglass will insulate very very well.
found a bunch of industrial looking locking hinges, they are from the shipping box of a sequencing machine I’ve bought years ago for the lab… 😜
love the look of those
and my rapido is complete again, new heatsink arrived, installed the 3 screws (carefully!) and after everything was bolted in it gained a 0.5 mm3 in flow (asa). So the slight bend was actually detrimental, curious to see when the bambu clone arrives how it compares
I’ve also found a 70w heating cartridge so I can try again the crazy
3 meters of canbus cable, and 8€ worth of packaging (the zip tights must be very expensive here in denmark)
2x2xawg22
I am going to have the wiring for the backpack first, this requires crimping few connectors and measure to length backpack to canbus board on the toolhead and another one from the backpack to the terminal block in the electronic bay. So I can install the skr mini and replace the current can wiring in one go, and have the printer operational if needed
once the can network works with the new setup I can remove one skr mini in the the bay and reassign the drivers in the config
plan to move in the backpack, the y endstop, the chamber thermistor, the nevermore fan, use the hotend terminals to switch on/off 2 circular led hotspots
and of course the a and b motors, leaving only the 5v led and the canbus cables going passing in the back of the printer from the electric bay
getting some help on assembling stuff
Cuuuute
that's awesome man
my helper loses interest quickly
and not with the printer, next thing I need to print is a wall-e
he started playing with the m5x10, for which I don’t have spare, so I wised up and put back everything for later
Dude, imma send you a care package. Like, 50 M5x10 BHCS and 100 SHCS. I have soooooo many. I don't know why, either.
Wonder how much it would be to overnight 150 screws across the Atlantic...
more than you’d think 😅
My mom's business sends boxes by two day air to Europe, but once overnighted it. It's insane.
Even snail mail is super expensive
don't forget customs
luckily I don’t really need any bolt, I stopped the kiddo before he could lose anything 😅. I had barely enough fastener for the tradrack, I stole 1 m5 button head from an aluminium brackets kit, I should have enough to play with… but I really need to order spares 😅
There are no good gifs for this, sadly.
My spares of M3 and M5 screws keep growing, but it's all in the wrong sizes (like M5 x 10) 😆 .
I got one of these for my kid https://www.infento.com/ takes like 4 days to assemble something that should take 1.5h
but he likes it 😄
Dude, this is super awesome! Never seen it before..
which kit did you get?
@gentle garnet Genius kit
Ah.. why cheap out?? The legend kit is the price of a modest 3D printer and is the only kit I’d buy 🤣
I'm not that rich 😅
may upgrade in a few years
Wait, you don't live off of loans and credit like everyone else? 😆
while I do have debt I try to be responsible, house and car thats it 😁
LOL this definition definitely makes me a financially irresponsible person
hahaha
that’s the price for a bambu x1 here, is it that a modest printer? 😅
maybe 100$ short of it
I’ve somehow cut the top panel in a way that I can attempt the sealing of the gaps, and it press a little the rubber. the mmu (blue:green thing) will be installed around where is places now, I think. now I have a rough idea of how to cut the back panel lid
I am not sure if I should bolt it down, or instead I should have a removable clamping mechanism, so I could work better on some parts for maintenance (eg rails or smilar
I'm a fan of anything that improves serviceability. Saves headaches later.
I finally gave in and ordered the expensive gear from bondtech 😓. I needed one 4mm ball, so I had to order a 1.5€ bearing -to be dismantled- and the cart was so empty to justify the checkout…. Damn you consumerism!
I’ll print the sherpa with the RIDGA mod, and see the difference. It should remove print artefacts due to uneven extrusion -due to the grab screw pushing the gear slightly off-center-
I also bought some b-stock asa (13€ 1Kg), let’s see what color arrives this time
At least it's cheap
I love random color cheap filament
all my red parts were dictated by the random color choice 😜, it seems the random color b-stock is always on at 3do. asa, petg, and they have also great deal on tpu
I might have over posted this, but it’s amazing
it always gets me the excuse to buy some stuff from them
no issues so far (ok maybe the end of the spool of the gray look like white instead of silvery of the beginning… but I printed super hot, maybe it was that)
My cad skills proved again to be the worst 😅, meshed together various pieces for the din clips into this. The noctua is there because I can’t find screws to fit a sunon 4020, but maybe the noctua will be good enough for this. The 4 corners will screw in the mount in the backpack lid
and the 2 heat insert in the front are for the mcp2551 CAN transceiver
looks cool
few layers before the top there was some under extrusion. it’s nit the first time it happens. It could be the filament getting stuck or some slicer issues… I don’t care for this print
that reminds me that I also need a place to put the can transceiver
but hopefully I’ll find out in the future
they are small little pcb, maybe it can stay dangling if you use some big wires on its terminals, but yeah, it’s better find a place for it. in the electronic bay I just bolted to a couple of generic din clips
U2C? there's an OmniBox mount for the BTT one.
https://jon-harper.github.io/OmniBox/0.9.10/support/lower_bay/#bigtreetech-u2c <-- forgot to backport into the 0.9.9 docs
Documentation for the OmniBox 3D printer electronics case.
no check my latest photo on my build log
IT TINY!
yes
just because, instead of doing anything useful and procrastinate instead, started a 24 min benchy
the skirt took 10 sec, it was quite impressive
the brim… whatever it’s called
noise level in the other rooms is good, I hope when fully enclosed to keep it even lower
it’s going to be more 30 min benchy
but still, my preparation routine takes forever
nope, got off the bed during the roof bridging
not bad quality overall. would like to know what are those artefacts… at least I have a general idea now
Curios to see your results after you replace the hotend with X1C clone. How soon will you get that?
Looks good for a 30min benchy
next week… hopefully. you never know if customs keeps it extra longer or all the stars aligns and it’s quick (I still have to see this)
I am really looking forward to finish/decide the next few steps. I need to figure out the toolhead wire and ptfe tube paths. It involves making a couple of holes, so I am procrastinating until I can’t avoid doing it
the ptfe tube is the easiest, it probably coming straight fro. the top, so a hole in the top lid to mount a collet, maybe just a pass through (so I don’t risk the tube to drop on the build plate while printing)
@proud cliff check how I did mine, can provide the parts
it's just a 2 piece print that screws together from both sides and the bowden attaches like in the sherpa without collet
that’s pretty nice, I was thinking something like this actually. How thick is your wood panel?
9 or 10 don't remember exactly, I can just redesign if you need different size
so you have 2 separate pieces of tube or is one tube that pass through?
or a can drill a pocket to fit it, I think I have 13mm thick panels
2 pieces
the "screw" has a wall to ensure that
you will need to be able to drill a hole wider than 10mm
I have mixed feelings about the sherpa ptfe slot, it fails over time, mine is a bit loose right now
I have some drill bit that makes 19-22 mm circles
and “gently” scrape everything in the middle
mine still works fine, but my sherpa was SLS printed
on the other hand, I have plenty of spare ecas04 collets
these that I printed it's PETG , pretty tight
let me get to the computer and I'll adjust it for 13mm
Just went to the vet, routine checked on my oldest cat (take some blood samples for controlling how she’s handling the chemo). Apparently not all the cat react like mine, I have a “difficult cat” apparently says the doctor (she’ve bitten me by mistake, she was aiming at the nurse 😂)
Oh dear how is she?
she’s a rock, almost 1 year after diagnosis
@proud cliff still with my dimensions no time yet to adjust
wow!! thanks!!!
I’m also making a grommet for the canbus cable, it’s probably easier than I thought the decision where to make the holes (removing the impractical places the choice was obvious.
I have tpu, but I never used it 😅, I am not sure if an asa grommet would be as good
TPU is fun when it works
@proud cliff did not forget you but didn't make the time yet to adjust
no problem, I don’t have time to print anyway, going for a small trip with the kids tomorrow 😅… maybe I’ll loose my sanity for good after that
do you still have any left?
I lost mine even before having kids
nope, good point
they can't make you lose if if you already lost it
but they can try!! they surely try
I am also dealing with running a scad file to make a grommet for the canbus wire, my idea is to have a screw in nut that will fixthe cable in place. I think asa will be fine
ah, right, the mystery box from 3do was another kg of red asa
Better to have a spare spool of what you already have than run out
yet another new sherpa mini. But this one is featured rich: RIDGA bondtech gear, a ptfe collet and the a filament sensor (it just need a endstop switch on the 2 holes and an m3 pressing it on the lever)
I don’t know how to remove the test ptfe tube from the collet now 😅
and also this tiny little 5 lanes filament switch is finished (mechanically)
I could make a funny spinning logo to make use of the protruding shaft of the ender 5 stock Y motor 😂
do it
or a knob… I am not sure for what use, maybe if I want to manually drive the selector 😅
Can't unsee!!
Is there a trick to printing cable chain ends? I have terrible luck with it.
I had good results printing a whole drag chain with Nylon. Turned out much better than the one I printed with ABS
oh right! I forgot about it. I have to find the right size that match the underlying heat insert, I had the 3 screws one, like for hydra, so I printed this one with 2 holes, but they are spaced ~2mm more than necessary
maybe they specify the exact model in the bom 😅… but how’s reading documentation anymore
actually even less than 2mm
I thought those things were standard in few sizes, but apparently there are subtle differences… Actually, I think this what’s happened: the bom chain is 10x10, I have the chain leftover for hydra that it’s 10x11. I suspect the stl I printed was just increase the size of the object from 10 to 11, with a result that the holes don’t align with the standard 10x10
@stone valve I’m printing this https://www.printables.com/model/296348-bowden-guide-with-print-in-place-360-degree-joint-
as I realised I am not planning to split the bowden tube from the rack to the hotend
oh that one is nice
the original mod also have a version with the sherpa-like hexagonal inlet
the usual sloth pace progress on this enclosure. But I decided on the ptfe inlet position and made the holes. I wasn’t sure if I wanted a 20mm hole on the top, so I just made a 20 mm counterbore-like thing to allow the wobbly coupler mod to spin and move around but only a 4mm hole fully passing through the panel
and installed the top bowden mod with a 10mm coupler
a ring with heat insert on the other side to secure the thing
failed attempt at installing the skr in the back of the printer… I spent 1 hours to carefully cut and make holes to a piece of plexiglass, thinking I had a wooden bezel to screw into behind the soft plastic panel, but no, I don’t have any bazel... I felt harder when pressing.. that was my judgment call to go ahead with the cuts 😅. this because I have to rush things and work only in the nights on this… during the day I make plans without the printer on hands and I imagine things
So I wanted to attach the board on the enclosure body, so when I open the backpack lid all the wires and electronics would stay there, but it seems it will cause a firmer installation on the lid, so when I open it all the electronics moves out 🤔
but at least, the filament path seems to work
maybe I can try harder to install the electronics in the non-moving parts, it’s just more manageable with wires not moving all the time
Which drag chain .stl do you use ? I’ve tried several times and always get horrid results.
how many links and which size are you aiming for?
I’m about to connect an end-stop to io29 and use it as filament sensor… hopefully it would work (can’t find any docs and I’m lazy to look the schematics)
nope
but I suspect they labeled the pins upside down... I guess it didn't matter for the x/z endstops, but for this yes
crimping a jst connector so I can comfortably test which has 5v.
you don't want 5V for the endstop
IO and GND pins only. 5V = smoke
The "5V = smoke" lesson was nearly "5V = fire" but I cut the power in time...
the thr36 is not turning on with 5v shorted
tested and I was actually shorting 5v… I did this before with the thr, that’s why I was quite sloppy on it. I know the the doesn’t turn on when shorting pins (I haven’t tested all the pins… so don’t blame me if your blows up)
so the labels in the schematics are actually inverted… I know the squared pins might have been telling, but I don’t know this things… now at least it is not smoking and it’s on with the pin connected… bolting all in again and actually see if the makeshift filament runout works
thr36 pins pdf is wrong, the schematics is correct. at least for the MT-DET (j13) header
yes, it’s actually 5v g io29
that’s probably why I couldn’t make io21 to work on the z probe header
good thing this board is full of protections and endstop pins 😂
everything on and connected
I added a new klicky fan duct option which should support uhf with unklicky
If that's something you wanted to try
huu, I've seen it, I actually have a couple of volcano nozzle, and I checked the unklicky (this morning when I've seen your post)
I somehow lost my uhf parts of I would have tested it by now lol
it seems something reasonably doable, I should be able to re-use the klicky mount with unklicky, practically, it only differs in the longer probe offset
I never had it, but I might test it without the bolt 😅
first I have to finish this thing, I'll make a comparison vs cht nozzle next week -hopefully- so I can test volcano next
I ordered my second rapido with it
is a v2?
runout sensor works
But I can’t load the filament properly anymore 😂, I think I need to push the ptfe tube more, it’s 3mm Ø internally, maybe it’s too big
tbh the v1 rapido is pretty good already. I had few issues but it never had downtimes (even if not fully functional it always managed to print fairly good)
nope, new extruder is stuck, I suspect I bonked when retracting fast the hot filament (eager to test the sensor)… I haven’t time now, but I need to disassemble the thing 😫…
I think the crazy is a very good hotend, the rapido is slightly better, faster and more consistent… but it’s not on another level
W00t!
yeah! it’s not the most user friendly setup, you have to adjust the screw so it clicks just when the filament leaves the gears (all works depending on the spring of the thumb screw)
how is it compared to the trihorn? does it get a better cooling angle?
Is hard to compare directly with klicky at least since it takes time to swap and adjust z again
But I don't think it's any worse?
@solar bridge I was picturing how an awd setup would be implemented, the front towers would have an additional idler, leaving the filament path hopefully unchanged, and the motors would be mounted upside down, on a corner mount just on top the hydra leadscrew (it works with the stock front bar) if only I wasn’t a complete mess with cad…
I don’t know why nobody tried it already 😅, I think that would step up the performance at a near -k3 level
I know 😭
yeah, the kinematic is pretty simple, each rail 1 motor, I guess the simplest the better works for everything
k3 at x/y and hydra for z 😛…
there was some junk filament leftover stuck in the sherpa mini, I tried to pull it out with some pliers… but than my better judgment stepped in, took the smaller Allen key I had and poke it down where it belongs, melted where it should be
apologies for the baby monitor in the background 😅
after that the sherpa is back, testing the filament runout it’s pretty reliable
planning for when-hopefully-if I'll receive the bambu clone hotend, I need to try and print this
not the most polished and slick mount as I was used with the ones from turtle... I also have terrible luck with removing support
I’ve seen a voron and setup. Had the front motors just inside before the tensioners.
yes, this one?
this is really weird in the context of a flying gantry, in our setup it would be easier since they are fix on the top extrusions
looks heavy
@proud cliff What colors and brand are these? Perfect color combo
the blue and the red? are both ASA from 3do.dk
dark blue and gray at the very beginning
ahh same, asa 3do 🙂
Makita blue, right?
3DO ASA is a UV-resistant and weatherproof copolymer designed for industrial uses.It has a high level of process dependability and is well suited for mechanically stressed components. It's an electrically insulating substance.The layer adhesion, thermal stability, flow characteristics, and warping tendencies of 3DO ASA
first time I see a cf other than black! https://www.fabreeko.com/products/3do-asa-cf-0-8kg-gray-limited-run-carbon-fiber
What is it3DO ASA is a UV-resistant and weatherproof copolymer designed for industrial uses.It has a high level of process dependability and is well suited for mechanically stressed components. It's an electrically insulating substance.The layer adhesion, thermal stability, flow characteristics, and warping tendencie
the standard 3do asa grey printed smooth AF. I am almost out of it and received 2kg of red for 25€, so I decided to change the accent to red 😅… but I loved the gray
They also have it in the eu shop… damn you gas and electricity bills, we got 6k $ of bills for kids/house/insurance in July 😩… well, hopefully they still have it next month
3do also got fancy colors with fiber glass, magenta and green ratrig… 🤤
I can’t avoid some downtime today, I can’t finish it in a reasonable time snd it’s 3 in the morning, so… new canbus cable connected to the toolhead; rear cable guide secured sturdily with the enclosure
next is remove the old canbus cable from the electronic bay, and connect the new one. test if everything works as before
than, install the skr mini in the backpack (the plexiglass already have mounting holes for the mcu printed parts (skr mini+mcp2551+cooling fan), flash it with can bus communication, move the ab motors, y switch, nevermore fans, chamber thermistor there, and test
cleanup the wires, remove old skr mini from the bay, rearrange electronics to have more room in there
That cable looks nice
it’s super shielded, and pair twisted. I don’t think it’s vastly better for our application, but I think/hope it may be reliable out of the box
And thanks to @stone valve, I used the 0.5mm ferrules, and it worked this time (after pushing a bit and trimming). I had to use a bit of fantasy in the color choice, hopefully I’ll be careful in connecting the wires properly later on
worst I did on mine was using red/black for AC
black is usually Line blue Neutral
here black is Neutral red Line
aka very swapped
(good thing AC has no polarity)
but then a wise man said "your build your rules, just remember them"
yes…. that last part is the issue 😅
true my plan is to set it very carefully close it and hopefully maintenance will be low like replacing a component where I don't need to think much either
just take this out, put same thing in
I seriously doubted I could pull it off (I’ve spent 1h trying to screw back a wire that got loose from a din terminal) but I made it! so nothing exciting happened, the printer is still in downtime -because I haven’t thought of making a proper wire duct in the back panel, so I had to remove the a/b steppers wires as well otherwise the canbus cable wouldn’t fit-
I’m half thinking the purple cable could pass trough something loosely fixed on the top panel (like a soft lace or something). it doesn’t drop loose, but I’d like it more vertical 🤔
here is how it moves (also note the dingling hole mount)
What if you made a larger version of your PTFE coupler that also passes the wiring through?
Might be more work than it's worth, though
but they could be one close to the other
I would make it enter back center at an angle
pretty much how mine is
I'll get a photo
only thing is unless you have the electronics enclosure on the top, it’s a lot of extra wiring to run
why not like this?
it looks cool
you can make it enter with angle from the center
night not work very well when the toolhead is at the back
ok, I’m hitting an issue with my old skr mini e3 v3. the one that lost the microusb. My plan was to use it in canbus mode via PB8/PB9 pins, as all the other skr minis (v3, v2, v1.2) I have/had in this canbus network, but it doesn’t came up. I have a fever today (also the little one… it very nice to have kids, 😒) but I think I wouldn’t mess up a make config… the board flashes but just doesn’t list
summoning @gentle garnet , maybe he knows how to debug if the canbus pins I configured are somehow faulty or the fact the the usb port was damaged influenced also those pins. I used this board in uart mode for months without issues
Hey there
First, can you share a screenshot of the menuconfig items used when building the firmware?
of course, So I’ve set up the board in the “backpack”. I first thought the wiring was faulty, and I somehow messed up, so I swapped an skr I have in the electronic bay in the backpack and installed the v3 board in the electronic bay. Now I can see the board in the backpack but not the one in the bay.. so the harness was good.
I tried also 500000
and even usb to canbus (go figure, the other v3 I have is set usb to can bridge…)
I mean I have a working setup now, but I really wanted to reuse this board. I like the big heatsink of this version 😅
How did you wire the CAN transceiver's RX and TX pins?
like this
Legit for both. Bus measures 60 ohm?
I swapped the board around keeping the tranceiver where they are, and I made the dupont terminals in a way that I can’t make mistakes… well it’s harder to make mistakes… the board kept not working
yes, It’s 60 ohm, If I remove the mid-point junction and measure each side gives me 120.
Cool. How did you confirm that the board actually flashed the firmware binary?
so, let’s say I flash canboot, communication in uart mode, and than go ahead to flash klipper to use cabus… would canboot allows me to debug what happens?
the sd is empty, and give me a firmware.cur after I let it flash. Also the led blinks slightly the first time it boot with the sd
Just to confirm - you placed the klipper.bin on the microsd card, named it firmware.bin, booted the board and after checking the microsd card you see a firmware.cur file. Correct?
yep
And you're 100% certain the binary was the one you created after make clean && make ?
also I have the 24v and the 5v circuits separated, so I can shut down all the mcu leaving the linux host alive… I tried to reboot the network few times with the klipper deamon off, and I can only see the 3 mcu I had, I keep missing the new one
hmm, I can try again now. I am sure, but 100%? I assumed 😅
Would be a shame to proceed to next steps without being certain on something rudimentary I guess 🙂
Can you post the content of your /etc/network/interfaces.d/can0 ?
Sharing ground between PSUs?
yeah, so many times I went to great lengths to almost ruin something that just needed a simple fix
cat /etc/network/interfaces.d/can0
allow-hotplug can0
iface can0 can static
bitrate 1000000
up ifconfig $IFACE txqueuelen 1024```
yes
Looks good. Which command you run for CAN bus query?
I removed the out folder and trying again
$ ~/klippy-env/bin/python ~/klipper/scripts/canbus_query.py can0```
make clean would do the trick 🙂
Your "Main" mainboard connected to your SBC is running CAN bridge mode, correct?
I did make clean just to be sure
yes
they all use the same setup
same transceiver (well one had a 120ohm shorting l/h)
the one in the usb can bridge
They don't have to - only one board should run CAN bridge, the rest of the devices should be "CAN bus on xx/xx" as you've configured here
Refering to this sentence, can confirm?
confirm the bin was the corre one.. nothing happens
no no, sure, actually only one is canbus bridge mode capable, the older skr mini don’t have the option
and the new v3 is set as normal can communication
I meant hardware setup
Ok. How easy would it be to isolate the main mainboard and this one? Also, can you find a 120ohm line termination resistor for this transceiver?
same wiring schema, same ttainsceiver
clear!
to isolate I just need to unplug a jst connector
the 120ohm is harder
I need to solder it in the pcb 😓
I see. Well.. It may work without one, but not necessarily and may not be reliable
I'm only attempting to see if this board comes up on the query without the rest of the bus, so not anything permanent
It only finds the first board
going back to the fondamentals
I made the fat32 partition with linux… I mean, how sure are we the the .cur file means the flash happened correctly?
well, we assume 🙂
That's the piece of code in the bootloader of the mainboard in charge of flashing and renaming the file once it's done
Gotcha. We are unable to be sure if this is due to the missing termination resistor, or due to another issue..
Any ST-Link programmer? You can install CanBoot bootloader on the device, but not without DFU over USB (which obviously you can't) or ST-Link using SWDIO.
well I could swap the old board there and see if it list just the 2 of them
...old board?
sorry, blabbing, the board that I took away from the el bay to test if it was the wiring or the board
the “old” board is now installed in the backpack, and I can see it when connected
OK... with the amount of boards you use even I'm getting confused 😁
so this are the 2 skr mini v3 i. the bay
and this one is in the backpack
when everything works I’ll remove the additional one in the el bay
May I see a closeup of this PCB?
tried a different sd, same stuff
@gentle garnet can we go save this man? Hahahha
As messy as it is it’s awesome you know every wire
I'm spotting a 120 ohm resistor on this PCB
Ah, not with the better photo 🙂
there shouldn’t be
and imagine that I tried to be tidy!!
but also I don’t have cable guide, I am thinking to print some
It probably is tidy. For you and many others. It works that’s all that matters and it’s sorta cool looking like that but drives my ocd crazy
hmm, I’m not going to do that now 😅
Now, it's a bit confusing for me to understand which board is what, from the 2 photos you sent.
the one not working is the one on the right in the electronic bay
I haven’t sent you close up of that one
let’s try
What do I look at here?
it the canbus cable, it has a metal sleeve and some plasic/cotton fabric… it’s really messy in there. I covered with a loose heat shrink (the blue thing)… once everything is sorted I’ll cut the excess of things and heatup the shrink
and this is the corresponding transceiver for the not working board
but the transceivers are confirmed to be all working
It's quite a challenge to follow your wiring, but it appears to be correct. I can only suggest isolating two boards, and continue from there. From software/firmware aspect your setup appears to be correct as well
is it possible to debug the pins?
ok, I see… I should take it out, flash it in uart
and see if the pins do something if I configure them
With an oscilloscope..
l only have a cheap multimeter
😂
worse case scenario I’ll use this board in uart mode for another build.. let’s say a v0 😝
Don't worry. Any I2C device? temperature probe, etc? Those pins have various configurable alternate functions
Should work as a digital output too, if you want to test it this way
as an [output_pin]
great!! so I’ll set it aside and debug it this way, if I find a fix I’ll just swap it in the backpack
but now I have to go… I feel like when I had corona, or it’s just manflu
that tomorrow morning my 18month kid with fever will be hell of a lot more active than me with fever
be well you all! It's past bedtime for me as well
but it doesn’t look I’ll fix the can issue, I might diagnose an issue, for curiosity
I don’t have the capacity/equipment for fix anything…
no actually I wanted to solder the usb , I bought a bag of those.. I completely forgot about it
it might not do anything, but who knows
It's not lost as you can still use it in another way. Soldering the microusb without proper tools can be risky, and it also depends on the condition of the traces and the pads. I wouldn't do it.
will not do it .. I tried it already with the old microusb, but failed miserably. the traces where pristine… before my first attempt
now they are bad
but it worked in uart mode after that
well, it somehow works 😂… my crimping technique improved along the line, but I just hate the DuPont stuff, if it works it’s good for me, until I have to recrimp
I might have also soldered the header not in the best way, so if a pins in the header is bent, than the connector can’t go all the way down
you should look at the last one that i did, it looks much better https://cdn.discordapp.com/attachments/1044248147473481828/1123029659907534958/image0.jpg
It tiny!
Seeing hotends on the websites then getting it in person at first blew my mind. They look bigger and then you get this tiny thing.
Only heard good thing about those hotends, I just wish the heat sink was symmetrical
I like to swap nozzles so there's that
Easy or hard on a Bambu?
not doable
On the clone you can
ohhh
Bambu just tells you to swap the hotend
2 connectors and 2 screws
Takes a minute or two
Okay, that's fair
The fact that the whole hotend is cheaper than a revo nozzle doesn’t make it that bad
LOL
how much btw?
I'm thinking of that $999 hotend now
the slice one?
slice is insanely expensive, nf crazy cost me like 45 ssme performance
Probably with heater, thermistor, etc.
0.4mm assembly is cheaper too, it’s AUD$15
how much flow does it handle?
I think this would compare with a rapido. it’s 40$ including a bunch of nozzle swaps (2 normal and 2 cht)
I still need to print the eva 2.4 mount
rapido should handle more flow
wich sucks, it will need support
I expect to have around ~30 with this thing
i get 20 with the rapido, stock
this is no a 0.4 nozzle spare, included in the bundle
I installed the 0.6
I don’t need more flow than 20, but it would be nice being able to go higher if needed
I don’t think this will replace the rapido, but I like to have spares
and maybe will use in future projects
I installe fusion 360, just because I am home sick and bored, just copied the stepper moror mount and brought forward. The motor pulley is already perfectly aligned with the front idler
auch! I'm looking at the stacks of the motor tower itself... I need to remove that
but from another angle they look aligned
ok, from this angle
I need to remove all the uneeccessary stacks from the mount, leaving only the pulley and a single flange stack
and make some mounting thing on the extrusion
Are you considering putting a stepper where the tensioner goes?
yes, leaving the tensioner as is
Ohhh
and slim down the rear stepper tower to don't interfere wit hthe tensioner
AWD CoreXY?
it only need a pulley and a flange stack
yes, but wait
I haven't considered which side of the belt the pulley is going to bite like this 🤦♂️
Ah, I think I can see where it might be trouble. Hmm.
What if the motor is the tensioner?
Dang, I need Fusion open
it's the wrong side of the belt.. so it need an additional loop of flanges
I think that would ideally be better, so for the mods one doesn't need a sligthly longer belt... but it's going to be hard to tension moving the all wagon
I think that's not proposterous, vz are doing this, k3 uses 4 motors (different kinematics), but basically it just gives more control other than more speed, potentially, This means less ringing at the same speed as well (provide one fixes al lthe extra vibrations)
Just curious if that's what you're doing so I understand.
What if it's not the whole stepper assembly, but just one of the idlers?
something like the tesioner from vengeful enclosure, right? the thing is fixed, but the idler inside/behind moves
the belt path should do something like this
You will lose a fair bit of travel
true, I have though about this but maybe there is a way to mitigate the thing
I feel like there's an elegant solution here somehow.
exactly
Motors outside the frame
The motor should be outside the frame, like the steppers on the back
Like they are in the back
jinx?