help, my printer is posessed

[neorpheus0]

Oh great citizens of the internet, I humbly throw myself upon your mercy. me/my roommate picked up a Rostock Max several months ago, back when the bad Rambo boot loaders were going around. we sent in our rambo and got it flashed and then continued with the build process as normal, everything seemed to be going fine until about half way through the calibration steps. there is a section of the PDF that instructs you to use a piece of paper and adjust the set screws on the skates that contact the limit switches. this part seemed to work fine. all the motors behaved as they should have and that part of the calibration went off without a hitch. the next part was to correct for any concavity or convexity in the travel of the print head, this involved opening up the firmware in the arduino IDE and changing a value that would cause the firmware to ignore the eeprom and then changing another value to basically hard code in a preset amount of concavity/convexity. this is where all hell broke loose. from the moment that the modified version of the firmware was uploaded to the rambo i have been unable to correct the issue im having even after attempting to reflash back to a bone stock firmware. an in depth description of the actual issue follows:
1:turn on printer
2:fire up fire up software on pc (sorry for the vagueness but i cant remember the name of the software we were originally using but it should be noted that the same results were seen when using matter control)
3:connect to printer (from within software on pc)
4:issue a G-code to printer (i usually used G28 (home) for testing purposes but did try a few others as well with more or less the same results)
5:printer starts doing the harlem shake(all motors have a strange jitteryness to their movements unlike the smooth movement seen during earlier calibration steps, and one or more motors are going the wrong direction. number,speed and direction of motors changes with every reset)
6:scramble for power switch
7:reflash stock firmware, reset values to their original statuses and reupload the now "de-modified" firmware, repeatedly place forehead firmly against wall at high speed, check stepper motor connections even though they worked just fine prior to firmware modification, try shifting all 3 axis steppers left or right, check motor wiring again, tinker with firmware some more, repeat the forehead/wall thing, get advised by someone on irc to check motor wires again, go into cursing fit while checking motor wires (by the way, the wires were exactly as they were the last time i had looked at them. surprising eh?),
8:get extremely frustrated and leave printer alone for several months (work, life, etc)
9:roommate decides its time that the very expensive and unwieldy paperweight fulfill its destiny of making lots of plastic knick knacks
10:i come to thee, oh great knowers of things in search for thine assistance in this great quest

at this point im totally lost on how to proceed. as stated i know the steppers are wired right, i know that flashing a stock firmware doesnt fix it and in fact im not even entirely sure that the flash is taking. any ideas? at all? that dont involve checking the darned stepper wiring again?

[guanu]

when you are reflashing the firmware are you also wiping the eeprom? if not, in arduino go to File > Examples > EEPROM > eeprom_clear and upload that.. after you upload that, the board is wiped clean, and then re-upload the stock firmware from our github... see if that changes anything.

Guanu

[neorpheus0]

that is something i hadnt considered, let me give it a try

[neorpheus0]

well, it turns out that somewhere during the intervening months since i last attacked this project the usb cable has vanished from my work area. I will see about getting a new one today and update this thread later tonight or in the morning. thanks for the help so far though!

[neorpheus0]

well i got my hands on a usb cable this morning and tried wiping the eeprom and then flashing the stock firmware from the github but there didnt seem to be any change in behaviour, any other ideas?

[neorpheus0]

*Continues staring intently into the monitor chanting* hear my cry great oracles, bestow upon me the guidance i desire.
seriously though, anything else i can try since the eeprom clear didn't seem to help at all

[astroboy907]

Intermittent shorts in the motor wiring perhaps?

[neorpheus0]

i wont rule out the possibility but i find it highly improbable because as i said i checked the wiring about 10e53475734 times. also the way that the motors are moving doesn't seem like a short/open issue to me. i will shoot a quick video of the printers behavior later on tonight to hopefully better explain the situation

[Polygonhell]

If it's jitteriness and not moving in a consistent direction, it almost has to be motor wiring.
I guess it could be the current settings in the firmware if you've changed them, or a bad RAMBO board.
The most common error is not getting the crimped wires properly seated in the connectors, a close up picture of the motor connector should let us see if that is the case.
What's a little weird is that you describe it as all 3 motors behaving the same way.

[neorpheus0]

yeah thats the thing. it is all 3 motors behaving strangely and all three motors were acting very smooth with no issues at all until the first time i tried to move them after setting that ignore eeprom value in the firmware and flashing it. from that point on this has been an issue. a firmware change could not possibly have effected the wiring, i guess its not outside the realm of possibility that it could somehow have damaged the rambo but i doubt that at this point

[Flateric]

Could also be movement, accel values set to high and the ol rambo is struggling to compute and holding things up with stutters until it does. Running out of cpu power for all the fast moves and accelerations...just a thought.

[neorpheus0]

that actually makes sense but im not sure how bad accel values would have persisted through multiple attempts to wipe the eeprom/flash stock firmware/ etc

[DavidF]

What happens if you disconnect it from the computer and try it from the lcd??

[neorpheus0]

i wasn't aware that you could control the printer directly from the LCD. i will be trying this and editing this post within the hour

Edit: it turns out that when attempting to home the towers from the LCD i get the same jitteryness as before. this may just be wishful thinking but they did seem to be slightly smoother but the same issue is there where the 3 towers seem to randomly pick the direction they wish to travel on each restart

[DavidF]

i wasn't aware that you could control the printer directly from the LCD. i will be trying this and editing this post within the hour

Edit: it turns out that when attempting to home the towers from the LCD i get the same jitteryness as before. this may just be wishful thinking but they did seem to be slightly smoother but the same issue is there where the 3 towers seem to randomly pick the direction they wish to travel on each restart

You said something about a video?? Lets see this evil spirt

[neorpheus0]

sorry about sound/video quality, this was shot in a rush on my way out the door for work the ohter day. just now getting time to post links
https://www.youtube.com/watch?v=XcDRhAePedU

[dtgriscom]

Maybe it's the music...

[Flateric]

Unfortunately the video does not show any movement at all.

Thought occured to me that perhaps your steppers are wired incorrectly.

This would cause very erratic and choppy movement.

[neorpheus0]

haha yeah the music. the printer was dancing or something. As for the misbehavior being tied to incorrectly wired steppers, well ill just say read my first post and ill leave it at that.

[Tinyhead]

Just wondering if you could check things systematically by say, unplugging steppers Y, and Z and see if X goes down smoothly when you adjust the Z height. Then plug in a different one and try again. Might just be able to verify some things that way. As was mentioned though, the video doesn't let us see what's happening. Try and throw up another video if you can.

[teoman]

sorry about sound/video quality, this was shot in a rush on my way out the door for work the ohter day. just now getting time to post links
https://www.youtube.com/watch?v=XcDRhAePedU

Psychedelic kindergarten music!

[neorpheus0]

Sorry about taking so long to get this second upload up. I didn't realize until long after I posted the link here that it was the wrong video. But now hopefully this will shed some light on the situation.

https://www.youtube.com/watch?v=vDgIiio7bDQ

[dtgriscom]

Looks like your X tower is doing the right thing, but your Y and Z towers are messed up. My strong guess is that it's a cabling issue, such as:

• Two of the stepper wires swapped within a cable
• One of the stepper wires not connected
• Stepper connector(s) in wrong RAMBo jacks

Try some of these things:

• Look at the four stepper motor cables as they connect to the RAMBo. Are the wire colors the same, and in the same order, on each stepper? If not, stop and report back to us
• For each tower's stepper, disconnect the other three steppers and try the following:
  • Move the corresponding Cheapskate to the middle of its tower, home, and repeat several times. I'm guessing your X tower will work fine, the others won't.
  • Confirm that the Cheapskate you thought should move is in fact actually moving. When looking at the LCD, the X is your near left tower, Y is your near right tower, and Z is your far tower.
  • Even if the Cheapskate goes the wrong direction, or does the watutsi, try stopping it by hitting the end stop at the top of its tower (you may have to hit the end stop twice to keep it stopped). If the same tower's end stop doesn't stop the Cheapskate, try the other end stops.
  • If the Cheapskate's motion is inconsistent, try applying some resistance and see if that changes what happens.
• On the connector for each stepper motor, measure the resistance between all possible pairs of pins. You should see a low resistance (10-40 ohms) between pins 1 and 2, and the same resistance between pins 3 and 4. All other pairs should be an open circuit.

And, in case it isn't a cabling issue:

• Move each Cheapskate through its travel range. Motion should be smooth, with no catches or bumps.
• Check the belt tension. If you hold the stepper still, there should be no play in the vertical motion of the Cheapskate, and it should take a little bit of force (but not a lot; yeah... this is ambiguous...) to pull the belt away from the back of the tower.
• Check around the stepper motors for belt debris from the belt slipping

Let us know how it goes,
Dan

[Aurora900]

I'm not sure if it was mentioned, but did you make sure you have no stepper motor wires running in parallel with the end stop wires? Make sure the point where the stepper and end stop wire crosses on the top of the printer is wired exactly as it shows in the manual, and also make sure the wires for the end stops and steppers are separated from eachother down where the rambo is.

[neorpheus0]

I have spare time today so I will go ahead and humor you all by checking on these wiring suggestions. but I would like to point out yet again that the towers behaved EXACTLY as they should up until the moment that a firmware change was made. there were NO hardware changes made at all. Unless gremlins snuck into my house and rewired my steppers during the 2-5 minutes between me editing a value and uploading the new firmware then i am pretty sure the motors are still wired correctly and plugged into the right holes. also i dont get what you mean about the stepper and end stop wires being in close proximity to one another near the top of the printer as the steppers and all associated wiring is in the base of the printer and the end stops near the top of the masts. how exactly are they supposed to be near each other in the first place?