Unable to write tmc spi 'stepper_z' register GLOBALSCALER
New Hybrid Core v4 build, has never worked (brand new).
Assembled machine, wired it, using custom DIN mount electronics enclosure.
Using Octopus EZ Max with EZ5160Pro drivers, but can repro it using an entirely independent Ocropus Pro 1.1 using TMC5610T Pro v1.0 drivers. EBB42 board. RatOS can see the Octopus and the EZ42b, can flash them both no problem. Any time I enter any gcode commands i.e. initial BEACON_INITIAL_CALIBRATION, or STEPPER_BUZZ STEPPER=stepper_x (using any stepper letter) I get this: Klipper reports: SHUTDOWN Unable to write tmc spi 'stepper_z' register GLOBALSCALER Once the underlying issue is corrected, use the "FIRMWARE_RESTART" command to reset the firmware, reload the config, and restart the host software. Printer is shutdown I've reproduced this problem using a 100% independent Octopus 1.1 Pro board I had (new but untested). I've swapped to a replacement EBB42 board, same issue. I swapped the drivers themselves from x and other ports, same thing (on both cards). Also swapped the ports the motors are plugged into. Same thing. I've swapped the Pi (same SD card though), same issue. I've swapped in a new Beacon, same issue. I'm running out of ideas. I 100% acknowledge this is probably something dumb I'm missing - a jumper, configuration, something. Maybe an endstop related issue? Crimps/wiring? Any suggestions? Attached are printer.config, klipper log, picture of the Octopus EZ Max board. Happy to share anything else as needed. (NOTE: Probably alot of permutations in the klipper log with various things I've changed; sorry). Would greatly welcome any ideas to troubleshoot this.
Using Octopus EZ Max with EZ5160Pro drivers, but can repro it using an entirely independent Ocropus Pro 1.1 using TMC5610T Pro v1.0 drivers. EBB42 board. RatOS can see the Octopus and the EZ42b, can flash them both no problem. Any time I enter any gcode commands i.e. initial BEACON_INITIAL_CALIBRATION, or STEPPER_BUZZ STEPPER=stepper_x (using any stepper letter) I get this: Klipper reports: SHUTDOWN Unable to write tmc spi 'stepper_z' register GLOBALSCALER Once the underlying issue is corrected, use the "FIRMWARE_RESTART" command to reset the firmware, reload the config, and restart the host software. Printer is shutdown I've reproduced this problem using a 100% independent Octopus 1.1 Pro board I had (new but untested). I've swapped to a replacement EBB42 board, same issue. I swapped the drivers themselves from x and other ports, same thing (on both cards). Also swapped the ports the motors are plugged into. Same thing. I've swapped the Pi (same SD card though), same issue. I've swapped in a new Beacon, same issue. I'm running out of ideas. I 100% acknowledge this is probably something dumb I'm missing - a jumper, configuration, something. Maybe an endstop related issue? Crimps/wiring? Any suggestions? Attached are printer.config, klipper log, picture of the Octopus EZ Max board. Happy to share anything else as needed. (NOTE: Probably alot of permutations in the klipper log with various things I've changed; sorry). Would greatly welcome any ideas to troubleshoot this.
Solution:Jump to solution
Thank you!
Confirmed I fixed this - had a mismatch of driver slot install and motor cable port. SMH, this has been killing me....
240 Replies
Unable to write tmc spi 'stepper_z' register GLOBALSCALER Once the underlying issue is corrected, use the "FIRMWARE_RESTART" command to reset the firmware, reload the config, and restart the host software. Printer is shutdownThat means klipper can't communicate with your z driver If you haven't selected 5160 drivers, not configured your driver jumpers correctly (in case of the octopus pro) or haven't selected the correct slot (you've moved your drivers around), you'll get an error like this. If you can't figure it out based on that information alone, please post a debug zip, so i can see how you've configured your motor slots.
Solution
Thank you!
Confirmed I fixed this - had a mismatch of driver slot install and motor cable port. SMH, this has been killing me.
What if you only have Diag Pins and not any other jumpers able to be set. I get that error and its on Kraken 1.0. If so is it software related error vs a hardware jumper. I dont have the option in the RatOS setup wizard to change drivers it only allows TMC2160. In the printer.cfg it has no reference, but, the RatOS.cfg is says they are "tmc5160 stepper_x". I am new to all of this But, I am not sure what is going on with two conflicting drivers being listed in different places.
What if you only have Diag PinsWhat?
I dont have the option in the RatOS setup wizard to change drivers it only allows TMC2160It should say "Kraken 2160 Plus" on the first 4 slots, and "Kraken 2160", you can't change them since they're solderen onto the board.
"tmc5160 stepper_x"This is correct, the first 4 slots should have 0.022 ohm sense_resistor and the last 4 should have 0.075. (in RatOS.cfg)
I am not sure what is going on with two conflicting drivers being listed in different placesYou need to select the motor slot where you have connected the stepper. Make sure everything is updated before you select a printer and go through the configuration steps. If you still have problems, please start your own thread.
@miklschmidt Sorry it does not. nor has it ever given me an option to choose drivers.
@miklschmidt Sorry. "What if you only have Diag Pins" was in response to a past post on this same issue. I am new to posting in discord.
Like i said, you can't choose drivers on a board that has the soldered on there (that's why the "integrated" tag is there). They're correct in the screenshot - because RatOS already knows those are soldered to the board.
Slot allocations look correct too.
I'm guessing you just forgot to connect the motor power?
@miklschmidt I have 2 new 24v Meanwell. 1 for the Board and 1 for the motors. I tested both power in connections at the board both getting 24.31v. the always on fan pins all getting 24v as well.
@miklschmidt I am at a loss to what to do next. I am at a clean install with all of the updates using update manager. except the Beacon. ( which would install a lessor version if I did)
can you show me a pic of your board?
@miklschmidt sure 1 sec.
@miklschmidt its a mess... I will clean it up once things are working.
Wait what's going on with the PSU's here, that's not a great idea
You have motor + connected to one and gnd to the other.
Ah i'm blind
never mind
Followed the wrong cable π
This is fine.
Post a debug zip (download from the sidebar in the configurator)
there is a connection in the bottom right where the power comes in
lol ok
it will have a bus bar and such later on.
You need to connect the - of PSU 1 and PSU 2 if you haven't
It should be done internally on the board, but it might not be
if i understand you correctly that connect is made on the psu1 on the fastend to the board
for now. it will be changed. lol
I looked but couldn't find that connection.
I'm talking about the 24V side
Also why are you using 2 PSU's? A single 350W is more than enough
both power supplys are supplying power to the board. psu1 is supplying 24v to the drivers/motors. psu2 that is sitting on the right side heat bed is supplying 24v to the board
In fact it runs on 200W.
well that cause BTT said that might be the issue.
Who from BTT said you should use 2 PSU's? ... Did they ask you what steppers you were using?
This board makes no sense for the V-Core 3 in the first place, those first 4 drivers have no place in a 3d printer.
0.022 ohm requires at least 6A RMS motors if you don't want horrible microstepping performance.
they know what the board is... it came from sorry [email protected]
where I bought it from.
The board is just one part of the equation. All that matters for power usage is what amperage you're running your steppers at (because the bed is AC, and that leaves a ~50W average heater)
and service 005@biqu3d
2 350W PSU's for 6 1.6A steppers and a heater is completely insane
Regardless. Those drivers should not be failing to initialize
@miklschmidt I planned on doing that
Yeah but use the bed output, instead of the heater.
Generally stick to the RatOS wiring diagram (which omits the AC connection, you can use the voron one for that part)
@miklschmidt thats from wiring 4.0.
Yep same deal π
@miklschmidt which they did state is was only the general idea. not to follow it to the letter.
Reason for using the bed output over the heater port:
Well that, and RatOS expects you to use the bed output π
I would rather us the bed output.
Anyway.. I don't know why your drivers are failing to initialize. Try a full power cycle of the board, they may have gone into a wonky state.
Followed by FIRMWARE_RESTART.
I used a kraken recently without any issues.
you can try unplugging the motor cables as well and see if that changes anything
Now i gotta go, it's 5AM π
RatOS wiring diagram is here btw:
(for the rest of it, if you're in doubt)
(you don't need 48V for the motors)
fans
Thank you for the layouts. I have saved them for later ref.
there is just to many sites saying that the SPI and some Max3xxx chip dont work together causing that error. then the not enough power to the motors. the screen shares the same spi or something...
its just endless directions to go to keep testing things that "may be" the issue.
oh and the misconfigured software settings.
this is the place the Biqu3d sent me for a possiable fix. https://www.klipper3d.org/TMC_Drivers.html?h=uart#i-keep-getting-unable-to-read-tmc-uart-stepper_x-register-ifcnt-errors
well. I will see if I can get them to exchange this board for another. maybe that will be the fix I am looking for.
Thanks for you help
That's plausible, sure, but RatOS doesn't use the Max31685 chip and it's a different error. It's an SPI communication error in that case.
then the not enough power to the motorsNot the case here
oh and the misconfigured software settings.Also not the case here, unless you modified stuff? yeah that's the very basics, we already checked that The drivers are not responding I haven't seen if you've done something to your config though, you could still send me the debug zip and i can take a look.
Post a debug zip (download from the sidebar in the configurator)
@miklschmidt this is a newer install. I havent tried much other than to home an axis. But, anything to do with motors gives and error.
Configuration is correct. Any issues is either electrical or actually dead drivers on the board.
@miklschmidt I never had magic blue smoke or anything seems as of there is something broke. everything has been smooth up until having to home any axis. I just figured that I was missing something really basic that I wasnt doing or clicking on.
Wait, did you not update package through the update manager?
I did
Aight good, then board is most likely DOA
yeah, so far I have not hear back from them on the return
And you tried disconnecting all the stepper cables on the board right?
It's not unlikely that the chips won't boot if you're shorting something by accident
yep I get the error even if I disconnect everything except for the thermistors. software must check this to make sure that they dont thermo runaway.
had that issue on my MK3s
I do have a hotend fan on all the time... except on reboot.
it does yes.
Sounds dead to me unfortunately
just the drivers though.. which is weird
I just placed another order, but this time on Amazon. I figured that BTT will not do a return.
You didn't get another Kraken right?
I'd really advise you to get the Octopus Pro H723 instead
It's a much better board
And for 24V all you need is 2209's which are dirt cheap
The Kraken won't run your extruder with any sort of accuracy whatsoever
isnt the Kraken a H723 as well?
the first 4 slots won't run any 3d printer stepper with any sort of accuracy either.
yes but the sense resistors on the first 4 drivers are completely overspecced, they suck for anything less than 6A per motor.
The latter 4 wants at least 2A.
I don't know who at BTT thought it was a good idea
yeah, I thought that a bit much. I see they launched a new Kraken motor as well.
No that's some dude, has nothing to do with BTT. It's a 3A motor, so still won't work properly on the first 4 slots.
should I use 2209's on the extruder as well?
yes
It runs 0.707A.
there's no point in going 2160 or 5160 for that (same driver except 5160 has a built in motion planner that we can't use).
the 2160/5160 are only really useful for high voltage. But you need to order a version of the 2160/5160 that can actually do that.
ok, stopped the order and return when it comes.
If you can get the TMC2240's those are better than the 2209's
ok, octopus Pro v1.1 with TMC2240s
ok, done. with 2 extra 2240s just incase.
Never had a board without the driver being built in.
First printer other than Prusa.
ok, thank you for the help. I am going to download the pdf on that board and read up.
@miklschmidt ooh does this need a plugin wifi too?
@miklschmidt ESP wifi module? I hope not Prusa had one on this mk4 when shipped and it was worse than dialup... yes, I used the BBS long long time ago.
What do you mean too? You have a raspberry pi, you've got all the wifi you need π
yeah... that just dawned on me as you said it.
lol, nvm!
haha no worries π
@miklschmidt is there any chance that Ratrig will make a Phaetus Rapido 2 Pro UHF, Beacon Rev H for the upgrade 1.0?
You mean the toolhead?
@miklschmidt yeah, they dont have a Beacon rev. H version
I believe there are mods for this
Printables.com
Rat Rig Toolhead 1.0 Duct and Fan mount for Beacon Probe by LogieBe...
Rat Rig Toolhead 1.0 (Vcore 3.1) modified duct and fan mount for the beacon probe | Download free 3D printable STL models
Printables.com
Rat Rig Toolhead 1.0 - Beacon mount with frame-safe cable routing b...
A twist on the back of the Rat Rig Toolhead 1.0 for stable cable routing away from pinch areas. | Download free 3D printable STL models
@miklschmidt none of them say for UHF
which adds like a 1/4 to the hotend.
looking at the cad files to see... the UHF is a dark nozzle with a square extension that is screws into
the RR Toolhead was only made for Rapido UHF
Yes that's the default,
The RR toolhead never existed in a HF version. At least not officially
@miklschmidt Ok, thank you. I will make sure to try that out. The one issue with 3.1 version that I have is the belt is angled just a bit looking at it from the front. Much better than the 4.0 tool head I tried. that was made for a different gantry all together.
The one issue with 3.1 version that I have is the belt is angled just a bit looking at it from the frontThis should not be the case at all. That would be an assembly issue or part incompatibility. 4.0 toolhead is not compatible with 3.1 (that's why it's 4.0 and not 3.2)
@miklschmidt yeah, I ran into that before i looked at the rest of the 4.0 changes. I just really like the 4.0 v-core 1.0 toohead combo. "I didnt think" that it would be an issue. tisk' tisk' tisk' that will teach me. lol I will make sure to reprint the "PP000346-rr_toolhead_vc3_front_V01_20240531.slt" and not any other.
@miklschmidt and those belt tensioners... yummy! I hope that they work well for others then I will have to get me a little v-core 4.0 300.
@miklschmidt Where is the display config located. I dont see anything in printer.cfg or RatOS.cfg. I know it running on the PI. shouldnt there be some kind of ref. to it somewhere?
What exactly do you mean by display config? Do you mean OS level configuration of the screen at a device level, or do you mean KlipperScreen configuration, which is an application running on the pi that shows a UI on the screen, not the actual screen?
@miklschmidt I was reading over something last night that was talking about the printer config and was showing the pin ref to where it was to be connected on the board. Given Raspi is doing the leg work, I was wondering if there was some bit of config in RatOS.cfg or one of the files that is links too.
@miklschmidt either way. no worries. Biqu is giving me the run around on the return. "Have you tried changing other drivers and configuring the driver files? Please provide a screenshot of your order and order number, and record a video of the error during use, and submit your configuration file so that we can help you feedback to the company."
What kind of screen are you talking about here? Like an old-school OLED with an encoder wheel?
Have you tried changing other drivers and configuring the driver files?:facepalm: ... they're soldered to the god damn board BIQU. You can tell them the config is correct and verified on several other krakens - you're using RatOS. Just provide them the klippy.log (download it right after reproducing the error) it has all that info in one file, including the configuration. Well it doesn't have the order and ordernumber obviously.. I suggest you go to BTT's discord instead. That sounds like some first level ali express rep.
@miklschmidt MOTOR_POWER or Main Power? Main I assume.
GitHub
BIGTREETECH-OCTOPUS-Pro/README.md at master Β· bigtreetech/BIGTREETE...
This is OCTOPUS Pro open source material. Contribute to bigtreetech/BIGTREETECH-OCTOPUS-Pro development by creating an account on GitHub.
@miklschmidt I also assume UART jumper settings for the drivers
@miklschmidt oh, no. its SPI only. operation mode, right?
yep
The 2240's that is
Same for 5160 and 2160
2209's are UART
Set them to main, no need for separate power
Rat Rig
13. Wiring, Firmware & RatOS
Step by Step Guide on how to assemble and wire the electronics on the V-Core 3.
@miklschmidt step 5 is using different jumper settings. I assume due to a different driver
correct, stock is 2209
@miklschmidt lol, i just got done taking all those off... cause of Ratrig doc. "Remove all the jumpers on the board before starting. Insert the jumpers on to the highlighted pins."
@miklschmidt ok, now 4 jumpers on each driver.
@miklschmidt flash UDF or SD
dfu
@miklschmidt Error Toolhead configuration cannot be deserialized, please check the configuration. partFan: Required
@miklschmidt after clicking skip on T0 Toolboard
@miklschmidt given i do not have anything pluged in to the board. not even the drivers.
Which printer did you select?
@miklschmidt Octopus Pro H723 v1.1 no toolboard
@miklschmidt
That's not a printer
Hmm this does not have a toolboard by default.. Did you already have a toolboard selected and selected a toolboard connected fan?
@miklschmidt I should get a skip option here. no toolboard
Anyway, you can fix it by reselecting the printer in the Printer Selection step (just click the navigation to the right)
Yes but you managed to create a broken config, maybe you selected a different printer and when reselecting the v-core 3 you chose "merge" instead of "start fresh" ?
@miklschmidt
Yes i can see this in your previous screenshot. Select a different printer, click "start fresh", then click the 500 v-core 3 and select "start fresh" again.
That will solve the error.
My guess is you merged a default printer config from a printer that by default comes with a toolboard into the v-core 3 defaults - which does not have a toolboard, and thus broke the config because a toolboard connected part fan isn't valid on a printer that doesn't have a toolboard selected.
If in doubt, reselect printer and click "start fresh".
@miklschmidt where is "start fresh"
When you select a new printer, youβll get a modal in your face
@miklschmidt hmm If I select the v-core 3 is the same error no matter what.
@miklschmidt I never get the option to skip installing a toolboard
Like i said, you need to select a different printer then reselect the v-core 3
@miklschmidt ok
both times, click "Start fresh".
@miklschmidt I think that I am past it now... just have the rest of the wiring to change over. Some need to be longer wires.
Klipper reports: SHUTDOWN
Unknown config object 'beacon'
Once the underlying issue is corrected, use the "RESTART"
command to reload the config and restart the host software.
Printer is halted
lol. nvm
@miklschmidt ok, I need to get a clear pinout diagram for wiring
Look at os.ratrig.com itβs the same as the other octopus pro models
@miklschmidt yep, I am using that. There are a few BBT pinouts for this that are green and white. very painful to read.
@miklschmidt So, I have a few things I need to wire up due to this being a 500m the two front z motors are to tight to run through the configs yet. plus an extruder fan is to tight for my taste.
@miklschmidt yeah, it doesnt like to heat the bed. "Klipper reports: SHUTDOWN
Heater heater_bed not heating at expected rate
See the 'verify_heater' section in docs/Config_Reference.md
for the parameters that control this check.
Once the underlying issue is corrected, use the
"FIRMWARE_RESTART" command to reset the firmware, reload the
config, and restart the host software.
Printer is shutdown"
What do you need that for?
I don't know what this means?
This is bad wiring.
@miklschmidt I am not sure why the bed isnt heating. But, as far as the wiring... I am just waiting on a package to get some longer front right and front left motor wire. I do not want to run any configs with bed moving down cause the wires are right there. I would rather run the wiring as it should be before going further. Plus I would like to double check the pinout and make sure. So far all lights on and no magic smoke. Thats something, right? π
@miklschmidt current setup. only temporary.
@miklschmidt Good news! I can home X and Y. Once I get longer wire for my to front motors I will home Z.
@miklschmidt I am so happy that I made it this far. This really means a lot to me. Thank you! I could not have done this without you! I does mean the world to me. Thanks!:chefkiss:
bed won't be heating as long as the heater isn't connected at least π
@miklschmidt I did have the heater pad installed directly into the bed connection on the main board.
@miklschmidt Was hoping that I wouldn't have to use direct mains power to power the bed. But, I am getting timeout errors. 24.31v on all 3 power in, and 24.31 on power out on the bed.
That doesn't work. The bed connector on the board should be connected to the DC side of the SSR.
It's a 220V AC bed. It'll do nothing at 24V π
@miklschmidt Keenovo 120v 1200w
American? Well 120V then.. Explains the timeouts as that would overload your PSU at 24V i can't math, that makes it a 12 ohm bed, which works out to 2A at 24V.
regardless, you'll get heater errors without connecting it properly
@miklschmidt Remeber to leave a full box of heat shrink tubing on the bed when running a Pid tuning. π
#imverysmart
@miklschmidt Drugs are bad mkay!
So letβs say you already own a Kraken, and were considering using it, would using running the steppers at 24v help this problem any more? Iβm not an expert on stepper drivers and how they function.
No, there's nothing you can do unfortunately. Even if you watercooled the steppers and overcurrented them, you wouldn't be able to reach anywhere close to the current required for the currentscaler to not drop under 16 with those ultra low sense resistors.
it's just a huge designflaw.
@miklschmidt so for the 2504 or 2804 what drivers do you recommend? I can do 48v or 24v. Any place you can recommend to do some reading on this topic?
First of all, forget the 2804, they make no sense on a machine like this, they don't have enough torque for you to get any benefit from the 1.5x torque drop off point.
2209 are fine on 24V. 2240 are marginally better (but best of all they have an integrated vsense circuit so they don't need the same kind of manual tuning regardless of current).
For 48V get the 5160 (and make sure they can handle 48V), i don't think there are 2160 high voltage step sticks out there yet. Make absolute sure the 5160 drivers you pick are using the standard 0.075ohm sense resistor.
Any place you can recommend to do some reading on this topic?TMC Datasheets It depends on what you already know. The TMC datasheets won't make much sense to you if you don't know how chopper drivers work or if you have no experience with basic electronic circuitry / signal processing. It's really hard making recommendations without knowing a persons level of understanding The best way to learn, is to learn how to acquire new knowledge and how to identify what pieces you're missing to understand the topic you're reading about. I'm not entirely there myself yet π
Learning how to learn is an important and difficult step π
Def not an electrical design engineer. More of a research and learn from folks like yourself. Tmc data sheet is where Iβll start
@miklschmidt hah well dang, now Iβm wondering if I could locate and change the sense resistors. Iβve heard of others doing it on other steppers just never understood why until now.
you can!
BTT publishes schematics, that should help you locate it π
@miklschmidt relatively affordable it appears. Just need to take the heat sink off later and see how crowded it is and if I found a correct replacement for what I think is on the board.
https://www.digikey.com/en/products/detail/yageo/PS0306FRP7T0U75L/15219933?s=N4IgjCBcoExaBjKAXATgVwKYBoQHsoBtEGADgDZS4BdXAB2ShAGU0BLAOwHMQBfXGAAZSAVnggkkNFlwFIxQQDoA7CIAEAWwDyACw0BnELRAMmAVQ5tkWgGYBZTAEN96VJj78SRFpg77Mam76bPrIeKhGuOTibAAmTAC0YIIg9IyQIKkgyACedO4ZHkA
@miklschmidt hey, I have a concern. After doing the "RatOS: Beacon calibration" the system did a restart. Was taking longer to start up and the bed was very very hot... like thermal run away hot. Am I missing something? I read something about this being an issue. I just don't know if there was a fix?
the system did a restart.It's not supposed to.. That sounds very bad.
Was taking longer to start up and the bed was very very hot... like thermal run away hot... Extremely bad, that should not be possible.
Am I missing something?Most likely wiring problems
I read something about this being an issueIt's not
I just don't know if there was a fix?Depends on the problem
@miklschmidt https://keenovo.store/products/delixi-ssr-40a480vac-solid-state-relay-input-3v-32vdc-40da that is how I had it setup temporally. IF the RatOS installs or reboots after a config. in my case the PID heat bed saved settings and then rebooted. when that happens the heat bed is still heating until PI is booted up. In my case it hung up on boot. Now i've done it again with the tool head on Bed Mesh. It saved settings and hung while the hotend keep heating up the toolhead. Both times I had to do a complete shutdown.
Keenovo Flexible Heaters Store
C-Lin SSR 40A 440VAC Solid State Relay Input 3V~32VDC 40DA
Description: This listing is for a brandnew C-Lin SSR of premium quality as shown in the pictures: Input: 3V~32V DC ( 3 and 4 --Polarity: 3--> Positive, 4--> Negative) Output: 24V~440V AC ( 1 and 2 --no polarity) Capacity: 40 amps( strongly suggest consuming only 50% capacity) Qty: 1 pc This unit enables you to u
@miklschmidt my wiring matches this https://os.ratrig.com/docs/boards/btt/octopus-pro-h723
BIGTREETECH Octopus Pro STM32H723 | RatOS
When the V1.0 version of this board boots into DFU mode (which happens during firmware updates), the heater pins on HE0 and HE2 will be set to high, meaning your hotend will heat up for the duration of the firmware update process.
Side note: Make sure the SSR is switching the live wire and not the neutral.
in my case the PID heat bed saved settings and then rebootedWhen you say rebooted, you mean klipper restarted? That's supposed to happen, i thought your pi rebooted which would've been really bad.
when that happens the heat bed is still heating until PI is booted upSo it did reboot!?
In my case it hung up on boot.What? I need to see what's happening Please record it
It saved settings and hung while the hotend keep heating up the toolhead. Both times I had to do a complete shutdown.How are you verifying that it keeps heating? and what do you mean it hangs?
@miklschmidt blown 2 fuses wiring it the way ratrig has on its page.
I doubt that π
I don't doubt you blew your fuses, i doubt that you wired it like in the instructions π
I'm gonna need to see some pictures / videos to help you here
You made a mistake somewhere, and it's hard to find without physical inspection
Miswiring AC also explains why stuff is hanging and acting weird.
@miklschmidt
@miklschmidt unless the power (line in) and bed has to be switched... that is the only thing I can see in question.
@miklschmidt maybe the fuse itself.
@miklschmidt So, It must be blowing the fuse due to the 5A
yeah the bed itself will pull over 10A.
Make sure your house wiring actually allows that. Typical american wall plugs are 10A IIRC.
you need to make sure the circuit is on a 15A breaker as well
Those 1200W beds were discontinued years ago because it was hard to run for americans.
EU is spoiled with 220v mains
@miklschmidt yep, that 15A shouldn't be an issue. But, I will keep it in mind. The voltage and measurement difference really becomes a pain. The evil Imperial empire didn't do a good job at getting everyone to switch over.
I'm still curious about the hangups and the continued heating etc you say you're experiencing, that should not happen. If you can show me what happens that would be very helpful.
@miklschmidt Well, I will give it another try once I get another fuse. I think that plug was repurposed from long ago and would like to get something that I know is made to connect to 15-20A vs 5A. Plus... I would like to look into getting Din rails and making things look nice and clean. I can change the wall outlet to be 15-20A GFCI and the breaker to be 15-20A.
@miklschmidt that's the look I would like to have.
Sounds like a great idea!
DIN rails are great π
@miklschmidt I should be able to update system node.js by using update manager even with Oct. pro. not being powered on correct? It is updating only the PI, right?
@miklschmidt
yes
you'd have to digging in the syslog to figure out what happened. Probably easier to just ssh in and run
sudo apt-get update && sudo apt-get upgrade@miklschmidt ok, no problem. I was trying to keep to updating only using the update manager in the RatOS
Which normally works fine, not sure what happened here, but if your pi has lost power multiple times it's likely a corrupted filesystem
They need to be shutdown properly before removing power
@miklschmidt I use the power in the RatOS to shutdown then power off.
@miklschmidt I know i've used ssh into this pi. I can use 'sudo apt-get update'. But, when trying to use 'sudo apt-get upgrade' I get error 'E: dpkg was interrupted, you must manually run 'sudo dpkg --configure -a' to correct the problem.'
@miklschmidt fixed
@miklschmidt node.js versions are different. 18.20.5-1 vs 18.20.4-1
@miklschmidt node -v v18.20.5
That's odd
It means what it says pretty much, i refer to my previous warning about the lost power.
you also said your system rebooted when you ran PID tuning
And you've blown fuses etc
So pi = dead in those cases?
A reflash is properly not a bad idea if you have further issues, in case of filesystem corruption.
@miklschmidt Nope, pi is new. working fine. Do you think a clean install of RatOS would be worth doing? I have a new 32g SD
I mean pi = dead as in pi lost power.
Yes
@miklschmidt PI fan still on, on board lights on and power switch is on.
But it reboots? and hangs?
(that means it lost power)
it should not be rebooting
Almost seems like a surge or temporary short.
impossible to know without observing the event
@miklschmidt once I have the fuse and ill test it again. I would like to add a 2nd fuse for the heat bed itself.
@miklschmidt Well... maybe sometime tomorrow. Gotta love Amanzon... years ago... (this is a years ago story) we used to have stores that had everything you would want to build a computer. not any more. á¿Γ
@miklschmidt Now it's time to Pew! Pew!
@miklschmidt Which order did you want people to upgrade in the update manager. ratos-config, RatOS, system, etc; I does it not matter?
RatOS first, but it should work just clicking "update all".
@miklschmidt Want me to use RatOS 2.0.2?
@miklschmidt I received this twice using RatOS 2.0.2 Update Manager - RatOS first, then Update all.
No π
2.1
2.0 is over a year old
this is not an error, it's just information
@miklschmidt installing image "2024-07-27-RatOS-2.1.0-RC2-raspberry-rpi32.img.xz"
why were you on 2.0.2 all of a sudden?
yesterday you were on 2.1
@miklschmidt I was installing to the new SD card and picked up a clean version of the software. searched RatOS and it took me to "https://os.ratrig.com/" and then I downloaded the newest version. "https://github.com/Rat-OS/RatOS/releases/tag/v2.0.2"
Yeah that's because 2.1 is still in the release candidate phase
there are still things that need changes before a real release can be cut.
@miklschmidt I really like "2024-07-27-RatOS-2.1.0-RC2-raspberry-rpi32.img.xz" very nice looking and best config and UI out of any that I have seen.
@miklschmidt I should never update Beacon due to the currently installed version is newer, correct?
No, not correct, you should update all packages.
@miklschmidt ok, update it is.
@miklschmidt Running 'BEACON_RATOS_CALIBRATE' so far, so good.
@miklschmidt It seems to be doing some sort of 'touch off' with the extruder tip. I would assume I should run this again once I have the magnetic sheet and flex sheet installed.
@miklschmidt π₯³ Very low gantry twist: 38.117337ΞΌm. No beacon scan compensation needed.
yes that's a good idea
@miklschmidt why a 4 wire 4028 and not a 2 wire. lol, now i have to figure out which wire is which. one would think after all this wiring would be the least of my worries. lol
The ones that work well for 3d printing don't come in 2 wire configuration
@blacksmithforlife that would explain it. I am still trying to figure out what naming schema is used to name other fans and which fans are triggered and which are always on. 'toolhead cooling fan' seems to trigger once you give the extruder a set temp but, shows on at 100% all the time,. but isnt. 'Fan' which i assume is the 'parts fan' is triggered so far with a slide bar. 0-100%. Controller fan I dont have installed yet. but, seems to be runing 100% from start.
These are all defined in the klipper documentation
@blacksmithforlife yeah, but over layed RatOS
No, in this case it is just klipper
Fan, controller fan, etc are all the names from the klipper documentation
https://www.klipper3d.org/Config_Reference.html#controller_fan
https://www.klipper3d.org/Config_Reference.html#fan
https://www.klipper3d.org/Config_Reference.html#heater_fan
@blacksmithforlife https://www.klipper3d.org/Config_Reference.html?h=fan#fan
@blacksmithforlife hmm. the information describing the fans function is a generic.
What exactly are you looking for?
@blacksmithforlife 4028 4 wire split to two wires. I gave the part cooling fan in. now i need to add another like it for printers drivers and such.
Sorry, I am not understanding. If you are using a 4 pin fan you have to supply positive, ground, and PWM. The pwm is what goes to the boards IO pin
@blacksmithforlife https://os.ratrig.com/docs/guides/4028
Connecting a 4028 fan | RatOS
4028 server fans are becoming a popular option for part cooling because of their light weight and great pressure and flow rates. In this guide we'll go over how to connect one and use it for part cooling. 4-pin fans usually aren't used in 3d printers, and many boards only provide 2-pin ports, but fear not! There's always a way.
@blacksmithforlife sorry. new to all of this.
Yes, that article says what I just did
@blacksmithforlife https://d3t0tbmlie281e.cloudfront.net/igi/ratrig/wg3AAMX6sPAYe6US.full
It describes getting power from the always on fan ports. Those just supply 12v
@blacksmithforlife yep, I did it once already for part cooling. But, I need to add a second fan just like it.
The blue is the pwn and is how the fan is controlled
Then use another always on fan port
Or directly supply from a 12v PSU
i.e. just need to supply it 12v somehow
If I recall correctly, the octopus board has 2 always on fan ports. So the port next to the one in the image you showed should also work
Double check by cross referencing your board's manual
@blacksmithforlife depending on which board, the type fan and which version of printer that you would be building. they change.. IE 3.1 vs 4.0 Ratrig shows many different ways to do this. none exactly matching.
Which is why Mikel provided many wiring diagrams(which has been a real headache to maintain I might add)
@blacksmithforlife oh, I am not blaming anyone one. I am just telling you that things are not that easy for a new person to just jump right into understanding everything.
@blacksmithforlife for example. I dont know if this 4028 is a 100HZ or 25KHZ. I have not been able to locate the information online as of yet. I just know that I have a hotend fan on 24/7 and dont know if it should or should not be. There isn't any reason to be on its not cooling anything down. But, maybe it's a safety feature. I dont know cause I am new and learning as I go.
@blacksmithforlife I might have to just flip '!' but, it didnt work when I saved in in printer.cfg and did a restart.
Correct, server fans (which is what the 4028 is designed for) go to 100 per percent if the pwm is disconnected. So either your wiring is bad, it's connected to the wrong port because the config is telling it a different one, or the board's port is broken
Yup, just imagine how it was before Rat-OS made it easier
Be sure to look through other postings, many times it has been solved before for people
@blacksmithforlife yeah... I love the wizard setup and the UI. super nice
@blacksmithforlife #############################################################################################################
FANS
#############################################################################################################
Part cooling fan
[fan]
4-pin fan connected to 2-pin header on Octopus Pro H723 V1.1 - digital pwm
pin: !PA8
cycle_time: 0.00004
@blacksmithforlife
What's the question?
@blacksmithforlife pwm is PA8 and my postive is and neg is in j56, fan7, ground. So, should I have the tachometer_pin as PA8 and have the rest as FAN7/GND?
https://os.ratrig.com/docs/configuration/includes-and-overrides
So you need to upload the entire printer.cfg - like the pinned post https://discord.com/channels/582187371529764864/1233461645167169576 rule #2 says
Includes & Overrides | RatOS
RatOS uses a modular configuration that heavily takes advantage of the config file include and merge logic in Klipper. For this reason, the order of includes and overrides are very important, do not change the order of the configuration unless you know what you're doing.
@blacksmithforlife yep. I read about overrides. didnt think I would have to use them for a fan. But, I'll figure it out at some point. I have to many things open with each pointing to somthing else to read or do.
That's the world of of a DIY printer
You get to know all the details of the machine
@blacksmithforlife yep, the Kraken pin outs and schematics were a little better than the octopus.
Tangent - I assume your username is a reference to wargames movie?
@blacksmithforlife yep. original Wargames. never seen the new version. the new is blasphemy!
IDK I enjoyed it Β―\_(γ)_/Β―
@blacksmithforlife these young bucks dont even know what it was like before the internet. Heck I remember that BBS were the best thing ever.
I remember when we got the Internet in my house
@blacksmithforlife 9800 baud... oh when we got a 286. we were the talk of the town
Anyways, it is getting late here on the east coast so I'm going to go. If you have more questions you might want to start a new post since it probably has nothing to do with the original post
@blacksmithforlife ah yes... they good old days. You could load am image and goto luch and it would be up once you get home.
@blacksmithforlife nope I am good. i'll figure it out. just no magic smoke and ill be happy. Thanks!
@miklschmidt so our conversation about the btt kraken let me to another thought. If we ran two steppers off one driver at 3.4a how would the driver respond since we are in the run current range you mentioned in the past that would make more sense and that would be shared between the two stepper motorsβ¦? Just ideas Iβm brainstorming.
It's better, but still not enough, you need twice that to avoid the currentscaler dropping below 0 on a 0.022 ohm sense resistor 2160.
Furthermore you're introducing further current measurement inaccuracies since the two motors may not act identical due to production tolerances etc.
Generally it should be avoided running multiple steppers on a single driver, there are only downsides if you have enough driver slots available to run them independently.
how did you fix it man? Same prob here. 2209 all works, swap drivers redo config, same error. No cables have been changed etc.
Iβm using the Octopus Max EZ. It has a bunch of ports. I didnβt have the stepper motor cables plugged in to the ports that matches the drivers. I.e. there are like 8 ports (from memory) and I had offset the cable plug from a stepper driver by 1
@miklschmidt I have ran "BEACON_RATOS_CALIBRATE" and works perfectly all directions X,Y and Z. But, the motors stay engaged after calibrate is completed and temps on the drivers continue increase to 90c. I am able to jogged the extruder manually to the left, right, forward and backwards correctly. But, jogging Z up and Z down are reversed when using the Toolhead controls. I am worry if I change it in the printer.cfg then wont the beacon calibrate be going the wrong direction and crash into the bed.
That is two different questions and probably should open a new thread at this point to keep each thread focused.
@TheTik I changed the wording to clear things up.
I chose to invert the up and down arrows in mainsail because logically to me the "up" arrow should make the bed go away from the toolhead but by default it is the other way around
Really? I press up and the bed moves away, and I didn't think i changed anything.
@blacksmithforlife yes, if I press up the bed goes down, and if press down the bed goes up. But, "BEACON_RATOS_CALIBRATE" works how I would think it should. Moving the bed out of the way when untill needed to be checked for Z. then it bring the bed up and gets it reading then stops.
maybe you inverted the controls a long time ago?
@blacksmithforlife everything is default except changing the exturded direction. I removed the "!"
no, this is a mainsail interface setting
what you did was change the actual behavior of the steppers
@blacksmithforlife wut?
@blacksmithforlife
@blacksmithforlife never changed anything but, the extruder direction.
if everything is working correctly except for the z interface in mainsail, just change your preference
@blacksmithforlife are you sure that if I invert the Z's that it will not nose dive into my new bed when homing or running the beacon config. I understand that you say its mainsail and it very well might be, but I am to new to all of this to desern which is really mailsail vs klipper vs ratOS and which is going to effect the others.
again, you should only invert z in mainsail interface after you have verified everything is working. Don't make your life harder
@blacksmithforlife yep, understood. That's why I was asking for help. There is a lot of info on Z and motors going the wrong direction and how to fix it. But, there isnt alot on RatOS, Beacon, mainsail with properly running "BEACON_RATOS_CALIBRATE" but, an inverted Z.
That's the default mainsail behavior π
up means "increase z"
@miklschmidt yeah, I was reading that in "https://os.ratrig.com/docs/configuration/#initial-configuration"
Configuring RatOS | RatOS
RatOS only needs minimal configuration to work, but comes with a lot of optional functionality that you can enable manually.
Yeah so if you click the up button the bed should move down on a corexy printer with a moving bed like the V-Core.
If it doesn't do that your z-steppers are inverted (but sounds like you fixed that)
Beacon, mainsail with properly running "BEACON_RATOS_CALIBRATE" but, an inverted Z.You cannot run
BEACON_RATOS_CALIBRATE with an inverted z, it would throw errors.
As the bed would move away from the beacon when it's trying to probe it.
So if the calibration went fine your Z is not inverted.@miklschmidt errors like "6:14 PM Probe samples exceed sample_tolerance"
That's a mechanical problem, not a configuration problem
Yeah I hate that, it drove me bonkers
That means your Z repeatability is bad
Yep but it's the only thing that makes sense on eg. a minion
It depends on the printer kinematics which mainsail knows jack squat about.
@miklschmidt I dont think it likes my wham bam flex sheet
I wouldn't know, never used one. But i doubt that's the issue. It would have to be quite soft.
Ohh yeah agreed, not saying it should default the other way. Just was pointing out there's a way to fix it
Pretty sure it's just a mechanical problem
It always is
ie. assembly issue
fortunately! π i think i have an issue on my list to make RatOS write the right direction to the mainsail namespace in the moonraker db.
@miklschmidt It has a PEX clear coat on the sheet and think there is to much room to move when doing the heated touch off part of the test.
You can try an do it on a bare metal and see if that helps (ie, flip the sheet around)
@miklschmidt yep, trying a non-PEX sheet now.
@miklschmidt hmm. I had it pass once before when it was just the aluminum sheety without the magnetic sheet and PEX bed wasnt installed.
@miklschmidt yep, z must be off. "Probe samples exceed sample_tolerance"
@miklschmidt home x:250.00 y:250.00 z:15.00. Using paper test. x:250.00 y:250.00 z:0.20 the paper has some resistance.
@miklschmidt yep. I think I am done for tonight. new error:Klipper reports: SHUTDOWN MCU 'mcu' shutdown: Timer too close
This often indicates the host computer is overloaded. Check
for other processes consuming excessive CPU time, high swap
usage, disk errors, overheating, unstable voltage, or
similar system problems on the host computer.
Once the underlying issue is corrected, use the
"FIRMWARE_RESTART" command to reset the firmware, reload the
config, and restart the host software.
Printer is shutdown"