Welcome! Log In Create A New Profile

Advanced

Duet Board failure?

Posted by smoothboy 
Duet Board failure?
July 08, 2015 07:59AM
After moved Y axe back and forth for a while due to a breakage and replacement of the right arm without disconnecting the Y stepper motor, I noticed a strange behaviour: it seems a steps loss in the Y direction, that gets worst if I set the Y current at 1000 mA. 800 mA gets better results but doesn't fix the trouble completely.
It is not at all a mechanical problem, so, it could be possible that the current generated by the Y stepper motor caused any Duet Board faults in its components? It could pe possible some kind of checking or fixing?
I tried out restoring firmware or reducing acceleration, but no way!
Thank you all.

Rick.
Attachments:
open | download - evidence.jpg (93.8 KB)
Re: Duet Board failure?
July 08, 2015 08:13AM
Hi Rick

This is almost certainly a mechanical issue. Most likely is either the belt is slipping, or the belt pulley is loose on the motor shaft (check the grub screw is tight). Increasing/decreasing the current has an effect, as the motor moves are more/less powerful, increasing/decreasing the amount of slip. For other possible causes, see: [reprappro.com]

Ian
RepRapPro tech support
Re: Duet Board failure?
July 08, 2015 08:22AM
Besides a board fault, a bad connection in one of the motor wires can result in the motor misbehaving in the way you describe, so maybe check your cables and connections.

One way to check if it is a Duet fault (stepper driver) would be to temporarily swap the X and Y motor connections. You could try swapping at both the Duet end and the motor end to further diagnose a driver or cable fault. Ensure the Duet is powered off before unplugging or plugging in the motors. After powering up following the swap, you should manually home X & Y by positioning the nozzle to the center of the bed by hand and sending a "G92 X100 Y100" command. You can then move X and Y using the web interface or Pronterface buttons, and see if the missing steps stays with the Y axis or shifts to the X axis. (It is also possible to print with the X & Y motors swapped so long as you don't have any bed compensation and the G-code file does not contain any X or Y homing commands).

I've noticed that moving the motors by hand with the Ormerod switched off generates enough voltage to make the PSU fan start spinning, but have not heard of it causing any damage - and I have moved X & Y rapidly by hand many times after lubricating the shafts with no ill-effects (touch wood).

Dave
Re: Duet Board failure?
July 09, 2015 02:37AM
Thank you for the answer. I tried out the X Y stepper motor swapping test and it conduced to an object with the error in the X direction! So, unfortunately, as I suspected, it is a Duet board failure due to a movement of the Y by hand. Is it possible some kind of fix on the board or I have to change it with a new one? I do have an Omerod 1; Is the Duet Board currently available compatible?
Pay attention moving the stepper motors by hand because it can result in a Duet board damage! :-(

Rick.
Re: Duet Board failure?
July 09, 2015 08:19AM
Quote
smoothboy
Thank you for the answer. I tried out the X Y stepper motor swapping test and it conduced to an object with the error in the X direction! So, unfortunately, as I suspected, it is a Duet board failure due to a movement of the Y by hand. Is it possible some kind of fix on the board or I have to change it with a new one? I do have an Omerod 1; Is the Duet Board currently available compatible?
Pay attention moving the stepper motors by hand because it can result in a Duet board damage! :-(

Rick.

It would surprise me if hand-movement can blow a stepper driver, and I suspect the damage was caused by something else. Are you certain that the cable is making good connection at both ends? If you swapped the cables as well as the motors (i.e. swapped at the motor end rather than the Duet end) a faulty cable will give the same result.

The usual cause of stepper drivers blowing is if a motor is connected or disconnected while powered up - the disconnection could of course be due to a bad connection rather than a deliberate unplugging of the motor, so after you repair/replace the Duet I would advise you to check or replace the motor cable just in case.

Dave
Re: Duet Board failure?
July 09, 2015 09:37AM
Quote
smoothboy
Pay attention moving the stepper motors by hand because it can result in a Duet board damage! :-(
Rick.

Hi Rick,
I have always done some movement by hand... M84 and then all I need to move.
Fortunately? I have never broken....
do I need to worry?

Dario


Ormerod 187
Firmware Electronics: Duet 0.6
Firmware Version:1.18.1 (2017-04-07)
Web Interface Version:1.15a
Slic3r 1.2.9a and Simplify3D 4.0.0
[www.dropbox.com]
Re: Duet Board failure?
July 09, 2015 10:21AM
I don't think this is a blown stepper driver. The print is far too consistent. When a stepper driver fails it moves very erratically, or not at all - it wouldn't print a part like your picture. What it looks like is a move in Z causes a move in the other axis. Are you using bed and/or orthogonal compensation (M556 S75 X0 Y0 Z0 in config.g)? If so, what are the numbers you are using? Post your whole config.g here, if possible. And what firmware version you are using (send M115 to the Duet).

It could also be a fault in the gcode. If you can ZIP the gcode file and attach it to your reply, I can check that too. What happens when you try to print the supplied gcode (Whistle.g, Snowman.g or CoatHook.g)?

Ian
RepRapPro tech support
Re: Duet Board failure?
July 09, 2015 11:57AM
I used my Ormerod very very intensive, producing complex mechanical parts (like turbine and turbine propellers for example) without any problem. The Y cable is not guilty. I moved the Y by hand and it was able to give enought power to light my Panel Due for a while! The Stepper Driver is not broken for sure, otherwise the fault will be much more evident than that. This trouble is very sneaky! So I thought in the first instance to a corrupted firmware or something like this. I had a similar problem times ago: suddenly in the middle of a print, the movement of the Y was changing completely with very "long" steps corrupting completely the object printed out. As in this case, I moved Y axis by hand before the issue for some mechanical fixing without disconnecting it.
In that case I fixed it restoring software.
Unfortunately it is not enought to fix the issue this time. I restored the firmware twice! I used DC42 1.09c firmware ad my printer was working perfectly on very complex parts as I told before.
This time suddenly after the first layer or so, I notice changing in the noise of the Y and the layers are shifting on the right. With a 1000 mA current the behaviour is much worst with totally erratic Y movements. May be some components around the motors drivers? I don't know.
Any advice is appreciated.
Thank you!
Re: Duet Board failure?
July 09, 2015 05:23PM
I suggest a visual inspection of the soldering of the Y stepper driver, in case the soldering on one of the legs is bad and the connection intermittent. It could also be caused by debris in the Y motor, so try swapping the Y motor with one of the others.



Large delta printer [miscsolutions.wordpress.com], Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].
Re: Duet Board failure?
July 10, 2015 02:05AM
Thank you. I already tried out swapping X and Y at the Duet side. Printing a simple square object I can observe the same issue now in the X direction, so I suppose the stepper motor is in good healt.
New Duet board is on its way and I hope to receive it soon; I use my printer for my own work so I can't stay whitout it.
If new board will be the evidence of a fault in the old one, I suggest to write down a warning advice in the Ormerod instruction saying not to move the Y stepper without firdt disconnecting from the Duet.
In the meantime I'll inspect for a soldering failure as kindly suggested.
I'll keep you informed.
Thank you.

Rick.
Re: Duet Board failure?
July 18, 2015 08:37AM
News!
I have changed the Duet Board without touching any other mechanic or electrical parts and my Ormerod is now printing perfectly as before. I have set up the Y current at 1000 mA as usually done and all is correct.
The trouble was caused by the hand movements of the bed with the Y stepper connected. In the meantime I received some A4982SLPTR stepper driver and I'll try to exchange the Y one on the old Duet to see if I can hopefully fix the board keeping it as spare part.
Definitely NEVER move the Y carriage without first have disconnected its stepper motor! It can cause a very insidious fault!

Happy 3D printing! smileys with beer
Re: Duet Board failure?
July 18, 2015 10:49AM
Let me know whether you manage to change the chip. I have an SMD rework station and a spare A4982, but I haven't needed to replace a driver chip yet. Getting the old one off shouldn't be too hard, but soldering the new one on may be tricky because of the thermal pad. I suspect it would be better to to heat the board from the underside using hot air, rather than heat the new chip from the top.



Large delta printer [miscsolutions.wordpress.com], Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].
Re: Duet Board failure?
July 19, 2015 08:45AM
Thank you for the advice! Unfortunately I've run out of solder flux now and it is mandatory for a good SMD solder job. I'm waiting to receive it so I'll try out a.s.a.p. I really hope to have my older Duet Board fixed because is rather expensive also grinning smiley I'll keep you informed thumbs up

Edited 1 time(s). Last edit at 07/19/2015 08:46AM by smoothboy.
Re: Duet Board failure?
July 20, 2015 07:33AM
Quote
dc42
Let me know whether you manage to change the chip. I have an SMD rework station and a spare A4982, but I haven't needed to replace a driver chip yet. Getting the old one off shouldn't be too hard, but soldering the new one on may be tricky because of the thermal pad. I suspect it would be better to to heat the board from the underside using hot air, rather than heat the new chip from the top.

I have a big high-wattage plumber's soldering iron that is useful for soldering components to thermal pads and similar - it allows a good joint to be made quickly. The I.C. shouldn't be affected by soldering temperatures even if kept hot for quite a long time, but the PC board can be damaged if heat is applied for too long, which tends to happen if the iron is under-powered. A heat gun is also usable but can end up desoldering nearby small components.

Such irons are not all that expensive, and it may be worthwhile buying one especially for this job (after which it is a useful addition to the toolbox).

Dave
Re: Duet Board failure?
July 20, 2015 09:37AM
Dave,
do you apply heat from the big soldering iron on top of IC to be solder or on the back of the p.c.b.?
Thank you.

Edited 1 time(s). Last edit at 07/20/2015 09:38AM by smoothboy.
Re: Duet Board failure?
July 21, 2015 07:54AM
Quote
smoothboy
Dave,
do you apply heat from the big soldering iron on top of IC to be solder or on the back of the p.c.b.?
Thank you.

I apply to the IC, or IC and pad together if the geometry allows, the same as soldering normally. It is only hot air that it is sometimes better to apply from the other side (which can be awkward to achieve). Make sure the pad is adequately tinned (probably already is from the removal of the duff IC). Pre-tin the IC where it will mate to the pad - not too much. Locate the IC and apply the iron until the solder melts and the joint "sweats" together. If the iron is big enough this will only take a few seconds. Applying a bit of flux to the tinned pad usually helps (but don't use an aggressive flux) - if I don't have flux available I just apply a small amount of solder wire around the edge of the joint as soon as the solder melts, not because I need to add solder, but to get a bit of the flux from the solder wire into the joint. Careful not to move the IC while the solder solidifies.

The longer the pad is heated, the greater the chance that the copper pad will come unstuck from the PCB, which is often irrecoverable damage - that's why a big hot iron is best.

Dave
Sorry, only registered users may post in this forum.

Click here to login