stable veldt
Voron 2.4 300mm (Doomed)
Beacon Probe
NH36 ADXL345
Symmetrical BFIs
A4T Toolhead w/ Dragon UHF
Everything else comes from LDO kit.
First time trying to use Input Shaper/Shake & Tune. Belts are equally tensioned, checked both with the printed belt tensioner and frequency measurement app. Belt lengths are equal. Racking was checked prior to belting and seemed okay.
stable veldt
royal forge
What number are you getting on the pf meter
You might want to install the galvanic Bowden holder
Also please post input shaper
stable veldt
Also please post input shaper
Will post input shaper when I get back to my PC in a few mins.
stable veldt
What number are you getting on the pf meter
A hair over 2
One thing I just noticed is that the x axis is not perfectly straight
These are off by about 1/8” of each other
But my buddy that ran the belts said they are 100% equal in length. If length and tension are equal what else could cause this?
Like I feel like I could maybe compensate by tightening b belt but is that okay?
Doesn’t address the root cause
stable veldt
Ok second one makes sense to me. What does fix umbilical mean?
stable veldt
Ah ok thanks!
I will do all that and see what happens and report back. Thank you for the suggestions!
stable veldt
Backing this up changed the belt tensions. They are now at around 2.5 on the PF meter
Should I reduce the tension or leave it
this is what I got when I ran it after only backing out the screw
royal forge
you need to add the galvanic
stable veldt
yep, printing now. what's your opinion on version without bearings?
royal forge
stable veldt
I'm going to print both versions so I can build one now without bearings and then order some if needed.
royal forge
for either version (bearing/no bearing) use a longer M5 screw and a m5 lock washer
stable veldt
ok good to know. luckily I do have plenty of those
stable veldt
ok so here's the result after just doing the screw and making sure the belt tensions were equal again
and here's now that I've done the galvanic
and here's input shaper run again
royal forge
Can you give an updated picture please of the entire front of your printer
stable veldt
yep one sec
royal forge
Are you measuring with the beacon adxl or toolhead adxl
stable veldt
Here’s update pics
stable veldt
Are you measuring with the beacon adxl or toolhead adxl
Toolhead. I tried beacon but there was more noise
royal forge
stable veldt
Done!
royal forge
Also check you z-joints m5 screw. Make sure it is snug
stable veldt
ok awesome! first date night with my wife since our baby was born 6 months ago, so might be quite a bit later before I'll get to do all this, but thank you so much!
I will let you know how it goes
royal forge
No worries. Take your time
royal forge
Enjoy date night. Hopefully it is better than going to Costco for diapers
stable veldt
that's a low bar haha. Trying a new restaurant so the food could go either way, but I'm sure it'll be a nice night no matter what. thanks 🙂
royal forge
Wait till number 2 pops out. It will seem like a high bar then
stable veldt
Fair enough. I want another, my wife is on the fence still
Btw had enough time to at least do zip ties. How’s this look
Oh and the toolhead screws were a touch loose
stable veldt
Ah yea good eye. Funny how easily you catch these things with experience. I watched the video too!
royal forge
There is a macro in shaketune called excite at freq.
you can excite an axis at a specific freq for a specific length of time.
Play with this macro and see what you find. I generally like it to go for 45 secs at a time
royal forge
check your belt path make sure it is routed correctly.
Your belts are uneven
are you running home/qgl/home z before running input shaper?
stable veldt
i just did home
stable veldt
so home all, qgl, then just home z?
stable veldt
ok my bad, I didn't know that was part of the routine for this
makes sense though
why the final home z?
royal forge
might as well, you just did everything else
and the toolhead needs to be in the middle of the bed to run the test anyway
stable veldt
How tight should those Z bolts be? I just checked one and it was quite loose
royal forge
They should not be loose
Snug, so they don’t;t come loose.
Aren’t you suppose to get ready for date night.
stable veldt
We are back already haha. Had to get back for bath and bedtime
We’re not quite ready to let my wife’s parents have her in water alone lol
stable veldt
I'll be honest... my wife asked if I wanted to do anything else and I said I wanted to go home and see the baby because I hadn't seen her all day lol
royal forge
and, then here you are again
🙂
I am going to go take a nap. Siesta. I will be back later
stable veldt
yep she's in bed now
ok thanks again. enjoy your sleep
making progress
oh in case you're still here. should all the bolts on the X axis be flush?
I noticed these two are quite low
nvm I found in the manual, it looks like just the front idler one is the one to not overtighten
stable veldt
ok I know I'm talking to myself here, but for whenever you're back... things are looking significantly better. it still looks like the B belt is looser, but if anything it's a tiny tiny bit tighter than A according to the tension meter.
but maybe I just need to trust the shaper results because there's a little bit of play with the PF meter. maybe it's not accurate enough for fine tuning like this
royal forge
125 is generally associated with toolhead. Check for loose screws in it and x-carriage, and put a piece of filament in extruder.
Also make sure your xy joints into x-extrusion is tight
Did you take the back cover off, and make sure the belts are routed correctly?
stable veldt
I haven't done that yet. it's such a heavy printer I wanted to see if the other stuff worked first. DavidP from Armor3D built it so I assume he routed it correctly, but I will check that next. I have checked all the screws (before this last run) and there were a few that were not that tight so I've tightened everything
i will check the belts though
It’ll likely be tomorrow night at the earliest before I get to it
There may also be some screws on the toolhead I haven’t checked. I’m going to take the toolhead off to check it out as well. I’ll make sure the belts are equal out of the carriage too
meager condor
thank you @royal forge ! nice work @stable veldt glad to see some progress!
stable veldt
thank you <@432399521012645890> ! nice work <@355508118802792450> glad to see s...
Yea @royal forge has been incredible helping me. Input shaper results are already loads better than they were to start.
Learning a lot.
stable veldt
Ok I think I may have found something
But I’m not sure
Was really hard to grab a photo of it but there’s a notch in. The toolhead board mount and there was a loose M3x8 in the carriage
I noticed that when leveling Z the toolhead moves a little
I tried tightening it but it’s too low to hook up with the notch
Everything else on the toolhead is super tight
I’m running it again without the screw in case it was that free floating screw causing the issue, but not sure if I need to come up with a modification to be able to secure that point
somehow that also fixed the belt tension issue...?
meager condor
Looking much better.
I think that screw was back there for a different toolheadboard mount. But im not 100% sure off the top of my head.
stable veldt
I asked in the A4T discord and he said something has to be misaligned in the toolhead because it should sit lower
Do you happen to have the model saved that you printed? Is there any chance it was the other Dragon UHF cowling? There are two and one is specifically for orbiter
royal forge
Can you take pictures of all 4 sides of your toolhead please. So that we can see the entire side of said toolhead
Also run test with scv=5 please
stable veldt
Yep can definitely take some pictures tonight. The rear is tough but I’ll try to use a mirror and see if I can get better pictures.
stable veldt
royal forge
royal forge
you got similar graphs as https://discord.com/channels/460117602945990666/1410350087854358649
stable veldt
I was wondering the same thing! I will check when I take the toolhead off tonight
Also not sure what to do about the cable
Wrap it in some tape?
royal forge
look for an ebb36 holder that has a longer tang to support the umbilical
the danger is over time you are going to cut those wire
royal forge
it woudl be for your extruder
stable veldt
Oh right. So I’m sure one exists in that case
I’m pretty sure the cowling is right too @meager condor. They are so similar I thought maybe that could explain the misalignment that DW-Tas suspects, but it’s gotta be something else. I’m going to inspect it tonight (I think) and fix the LEDs while I’m at it
meager condor
I printed the cowling for the dragon uhf with orbiter spacing.
Those 2 screws on the back with the washer are to hold the toolhead against the mgn carriage (afaik).
royal forge
@stable veldt lets go ahead and remove those then
stable veldt
I printed the cowling for the dragon uhf with orbiter spacing.
Those 2 screws ...
Yea I figured just checking off the boxes, sorry! I could see myself accidentally grabbing the wrong model lol.
stable veldt
<@355508118802792450> lets go ahead and remove those then
Ok will do
royal forge
after that home/qgl/home-z, input shaper
stable veldt
Yep 👍🏻
stable veldt
removed those screws and running everything again. that alignment issue by the way seems to be based on the alignment of the carriage heat sets and the holes in the toolhead. when the toolhead sits flush and is resting on the hanger bolts it's too low. I have to lift the toolhead up a few mms to get it screwed on
royal forge
Y looks like loose screw in toolhead. X could be the same thing
Let’s print a test cube to see where we are
Set both shapers at ei , make sure you include damping ratio
Voron cube=3 perimeter/bottom, 0 top/infill
stable veldt
ok will do. curious, what's the difference between ei and mzv?
stable veldt
toolhead bolts all seem good. I took the toolhead off one more time to just snug up the carriage bolts and I adjust the hanger ones a tiny bit further. got the best result I've had so far I think
going to run shaper one more time and set the ei values and get teh cube printing
will take pictures in the morning
I do have a quick question. where do I apply the results from vibration profile? I don't know if I'm just losing my mind, but I've read through shake tune docs about what everything means, but I don't know how to actually use those values
I get that some of it is helpful for how to orient parts for optimal quality, but for example the green zones. I don't know how to use that. there are gaps and the test only goes up to 200mm/s
do you know of any good resources out there to learn more about this stuff?
royal forge
You use vibrations profile to set int/ext perimeter speed in slicer
stable veldt
Ah ok thanks 🙏
stable veldt
So it looks a lot better. I just took that last picture as a reference against a really clean one (left - diff printer ofc)
It also looks very similar to the one I printed with the original input shaper settings before all the tweaks we've made
The one based off the really messy graph we started with
Which I was pretty surprised by
royal forge
The last picture (right one) looks like it may need a PA adjustment
stable veldt
hmm... I've done PA tuning using Ellis' PA pattern
a couple days ago I calibrated the extruder, first layer squish, and PA
I will run it again though just to be sure
stable veldt
I'm at work atm, but I have a new cowling printing. Without disassembling everything it seems like the easiest way to identify why there is a misalignment of the mounting holes. I feel like some of the vibration may be due to the fact that when the toolhead is assembled it is not resting on the hanger bolts at all because I have to raise the toolhead a few mms to align the mounting bolts. if the new cowling has the same issue then I may need to reprint the x-carriage to perform the same check. If I can fix that issue it will also secure the toolhead mount misalignment. I'm also modding the toolhead board mount with a longer tang for the NH USB cable.
I will probably try to get all of those adjustments made, re-run everything to see where we're at, do PA tuning again
stable veldt
this is the new toolhead board mount. hopefully that'll help extend the life of that cable
royal forge
try this one. Use the upper ziptie places. You want to zip tie against the black and not the individual wires
stable veldt
ok so don't use both sets, makes sense. I had left them in there thinking i'd use both, but it makes a lot of sense not to ziptie around the original spot directly on the wires
I also am going to run a new cube because I was looking through the config files and I think I may have forgotten to save the most recent input shaper values because they are both set to MZV instead of EI (and not the right values)
stable veldt
Looking a lot better with the correct IS settings actually applied. Looking a lot closer to the reference cube (honestly better in some ways)
In fact I am pretty happy with these results overall. Tempted to leave well enough alone, but I think the right move is to fix the toolhead so that it mounts properly to the carriage. if for no other reason than piece of mind, but I imagine that if I can get rid of that peak on the graph I'll improve my speeds some.
stable veldt
Confirmed it was the cowling. Printed a new one and it fits perfectly. Not sure what but something must have happened to throw off the original print by just a few mms
royal forge
Your cube shows no resonance.
stable veldt
Yea but I want to fix the issue with the cowling anyway. Won’t I possibly get more speed if the assembly is more stiff?
royal forge
Potentially, but you will hit a wall at some point
stable veldt
Right yea that makes sense
Is it normal for one axis to be half the max acceleration as the other?
stable veldt
Ahhh right
meager condor
Yea but I want to fix the issue with the cowling anyway. Won’t I possibly get mo...
What's wrong with the cowling again?
stable veldt
What's wrong with the cowling again?
When it’s resting on those bolts coming out of the carriage the mounting bolts do not line up
You have to lift the cowling up a a few mms to get them to align so then it’s not resting on them. And there’s a third one in the back that attaches to the toolhead board mount that you can’t use because the toolhead board mount is too high
meager condor
Maybe it needs shorted standoffs for the toolhead board?. The bolts on the back of the toolhead with the washers are needed to clamp it on the mgn carrier
stable veldt
Yea those actually aren’t a problem. I’m going to reinstall those with the new cowling. The problem is definitely something went wrong with the cowling print. It’s the actual holes in the cowling lining up with the ones in the carriage
The new print fits perfectly. And it’s impossible to see an issue with yours in that regard. I couldn’t even hazard a guess as to what could have happened
I can take a side by side, maybe you’ll be able to spot what I’m missing
I’m actually not using this one either. I was trying out that infill but it looked cooler in the slicer than the actual print lol
stable veldt
Any idea why Z makes this sound when going up?
Otherwise the new cowling worked out well. Everything lined up nicely.
Made a minor improvement to the belt graph, no more peak at 125Hz
Overall it’s pretty similar though. The other things we did (galvanic, tightening screws, fixing zip tie locations, etc) made a bigger impact
royal forge
Where is that sound coming from? I think a s motor but which one?
royal forge
It sounds like it is skipping.
First check config. Make sure that motor us the same as every other motor
Then visually check to make sure it looks the same
Then check screws and make sure it is the same
Tension appears to be on the bottom. You will need to make sure it is transioned the same
stable veldt
Visual all look the same, screws are same and tight on all 4, tension is same on all 4
Checking config now
royal forge
Look at max z-accel
stable veldt
it all looks the same except for a few extra things on the first section for beacon
step_pin: PG9
dir_pin: !PG10
enable_pin: !PG13
rotation_distance: 40
gear_ratio: 80:16
microsteps: 32
endstop_pin: probe:z_virtual_endstop # use beacon as virtual endstop
homing_retract_dist: 0 # beacon needs this to be set to 0
position_max: 245
position_min: -2
homing_speed: 8
second_homing_speed: 3
## TMC2160 configuration
[tmc5160 stepper_z]
cs_pin: PD2
interpolate: false
spi_software_miso_pin: PC7
spi_software_mosi_pin: PC8
spi_software_sclk_pin: PC6
sense_resistor: 0.075
run_current: 0.800
stealthchop_threshold: 0
## Z1 Stepper - Rear Left
## Connected to Motor_6-S6
[stepper_z1]
step_pin: PG11
dir_pin: PD7
enable_pin: !PG12
rotation_distance: 40
gear_ratio: 80:16
microsteps: 32
## TMC2160 configuration
[tmc5160 stepper_z1]
cs_pin: PA15
interpolate: false
spi_software_miso_pin: PC7
spi_software_mosi_pin: PC8
spi_software_sclk_pin: PC6
sense_resistor: 0.075
run_current: 0.800
stealthchop_threshold: 0
## Z2 Stepper - Rear Right
## Connected to Motor_7-S7
[stepper_z2]
step_pin: PB4
dir_pin: !PB3
enable_pin: !PB5
rotation_distance: 40
gear_ratio: 80:16
microsteps: 32
## TMC2160 configuration
[tmc5160 stepper_z2]
cs_pin: PA9
interpolate: false
spi_software_miso_pin: PC7
spi_software_mosi_pin: PC8
spi_software_sclk_pin: PC6
sense_resistor: 0.075
run_current: 0.800
stealthchop_threshold: 0
## Z3 Stepper - Front Right
## Connected to Motor_8-S8
[stepper_z3]
step_pin: PG15
dir_pin: PB6
enable_pin: !PG14
rotation_distance: 40
gear_ratio: 80:16
microsteps: 32
## TMC2160 configuration
[tmc5160 stepper_z3]
cs_pin: PA10
interpolate: false
spi_software_miso_pin: PC7
spi_software_mosi_pin: PC8
spi_software_sclk_pin: PC6
sense_resistor: 0.075
run_current: 0.800
stealthchop_threshold: 0```In the config in [printer] there should be a max z accel what is it?
And max Z speed
royal forge
Make it 15 and 250
Then rerun
If that is good what you are going to do is increase accel by 25 till the noise appears
However, it could be making this noise because you are not qgl’d. So you will have to check that before you start these test
stable veldt
Am QGLd and it’s still happening with those speeds/accels
stable veldt
15/250
stable veldt
Oh and on longer travel it starts to sound more like a creak/groan
Kinda scary like something’s about to snap
M5s I just tightened the other day. M3s I could tighten but they are already pretty tight as well
royal forge
At this point I would ask you open a help ticket
It is more involved than resonance issues
On top of that I have not played with inverted z motors
Either here or doom. Doom discord maybe better. I do not know
stable veldt
Ok yea I realized after that it’s beyond what I came here for help on. I asked in the main v2 chat as well. Appreciate all the help getting the resonances taken care of. Learned a ton
stable veldt
@royal forge this is what I get after fixing the Z belt issue
