Dynomotion

Group: DynoMotion Message: 11523 From: scubadiverdab Date: 5/14/2015
Subject: Problems with KSTEP (two bad channels)

First I would like to say, I have really enjoyed working with the kflop and kstep, they are nice pieces of hardware.  With that said, recently I have had two kstep channels stop working and need some guidance.  Here is what happened.  


I had my system up and running spread out on the table and decided to make an enclosure to protect my kflop and kstep.  When I hooked everything back up in the enclosure all the motors powered up like usual. 

When I tried to jog the Z axis, it appeared to step once and then nothing else no noise or movement.  I had no increase in current or burning smell or odd noises accompany this.


I was using a different power supply, so I thought maybe that was the problem.  When I switched back to the old power supply I still didn't have anything from the Z.


I switched the Z motor over to the Y channel and had the same one step then nothing happen.


So now I have a kstep with two channels that are not working.  I brought the boards back up to my lab to try to figure out what was going on and here is what I have done:


First I checked the wire/plug going from the board to the Z axis stepper.  There were no shorts and I see the coil resistance I expect between the different pins.


I have probed and tested all 4 step outputs from the kflop and they are sending pulses like expected.

  

I have also checked for the step signal going into the fet drivers on pin 26.  I see the pulses coming in on the two working axes, but just see a solid high signal on the two that are not working.


I see no traces of burned components or blow traces or debris on the board.


Looking for ideas on where I should go from here.


Thanks,


-Andrew
Group: DynoMotion Message: 11524 From: Tom Kerekes Date: 5/14/2015
Subject: Re: Problems with KSTEP (two bad channels)
Hi Andrew,

Sorry to hear that.

Was the Y Axis working before moving the Z Cable to the Y Output?

Did you check for shorts in the cable to the shield?

When you say "Motors powered up" do you mean they have proper holding current and holding torque?

Did you check the cable between KFLOP and KStep for being fully plugged in or shorts?

The Step signals for each channel go through a 100ohm resistor R71 then to opto isolators.   As shown for channel3 here:

Inline image

If there is input current on pin 1 the the StepIso3 pin should drive low.  If that doesn't occur this device could be bad or there is a short in the driver chip forcing StepIso3 high.

HTH
Regards
TK


Group: DynoMotion Message: 11525 From: scubadiverdab Date: 5/14/2015
Subject: Re: Problems with KSTEP (two bad channels)
Yes, the Y-axis was working before I swapped plugs.

There were no shorts to the shield I could find with the multimeter.

When I powered up the motors starting hissing like usual.  I say they were working because I saw the same amp draw on the supply I usually get when they are powered on but not moving.  ~0.5A per motor.

The cable was seated firmly in both boards.

I am going to probe at the optoisolators now.

Thanks for the quick reply.
Group: DynoMotion Message: 11526 From: scubadiverdab Date: 5/14/2015
Subject: Re: Problems with KSTEP (two bad channels)
Ok, Scratch that.  The optoisolators are working.  I can trace the steps going all the way into the driver.  Man those pins are small!  I was on the wrong trace probing the other two earlier.  The signals are there now that I am on the right traces.  So the problem appears to be the driver or the fets.  On the working channel I see the nice square wave going into the fets from the drive and on the output to the motor.  I do not see this on the channels that aren't working.  The inputs to the fets are low or high, never changing.  The outputs are doing the same thing.  Does that point to bad drivers?
Group: DynoMotion Message: 11527 From: Tom Kerekes Date: 5/14/2015
Subject: Re: Problems with KSTEP (two bad channels)
Hi Andrew,

That would indicate a bad Driver chip or the FETS.  For each channel there are 5 chips involved (4 dual FETS and the Driver).  To repair it is pretty much necessary to replace all 5 at once otherwise one remaining bad can re-blow the others or worse.  So with two bad channels that would be 10 chips.  I can give you the P#s if you feel like attempting it yourself.  All the copper heat sinking makes it extra difficult.

Did you check all combination of pins on your cable for shorts with the motor connected? ie between coils?  (I'm sure you did but can't help from asking as everything points to the cable).

Do you have a large standoff causing a short to a trace or via?  Too big of a screw through a hole cutting into the internal layers?  Grasping at straws....

Regards
TK


Group: DynoMotion Message: 11528 From: Andrew Bryant Date: 5/14/2015
Subject: Re: Problems with KSTEP (two bad channels)
Tom,

Ok, just found the issue.  If I move the connector on the Z motor a certain way, one of the leads shorts to the metal plug body, which gets grounded through the box it is mounted in.  Could that still have taken out the driver chip? It should only be isolated to one coils worth of Fets.  I guess I can replace them all just to be safe.  I've got a hot air station, I should be able to float the chips off without too much trouble.

Are these the correct chips?
Driver Chip: http://www.digikey.com/product-detail/en/A4989SLDTR-T/620-1389-1-ND/2533654
Fets: http://www.digikey.com/product-detail/en/IRF7351PBF/IRF7351PBF-ND/2206033

Thanks,

-Andrew

On Thu, May 14, 2015 at 6:03 PM, Tom Kerekes tk@... [DynoMotion] <DynoMotion@yahoogroups.com> wrote:
 

Hi Andrew,

That would indicate a bad Driver chip or the FETS.  For each channel there are 5 chips involved (4 dual FETS and the Driver).  To repair it is pretty much necessary to replace all 5 at once otherwise one remaining bad can re-blow the others or worse.  So with two bad channels that would be 10 chips.  I can give you the P#s if you feel like attempting it yourself.  All the copper heat sinking makes it extra difficult.

Did you check all combination of pins on your cable for shorts with the motor connected? ie between coils?  (I'm sure you did but can't help from asking as everything points to the cable).

Do you have a large standoff causing a short to a trace or via?  Too big of a screw through a hole cutting into the internal layers?  Grasping at straws....

Regards
TK


Group: DynoMotion Message: 11529 From: Tom Kerekes Date: 5/14/2015
Subject: Re: Problems with KSTEP (two bad channels)
Hi Andrew,

Well at least you found something that can be fixed to avoid future problems.  A shorted FET is very possible to take out the driver chip as the Motor Supply can short through to the gate driver.

Yes those are the correct P#s.

Good luck
Regards
TK

Group: DynoMotion Message: 11559 From: scubadiverdab Date: 5/24/2015
Subject: Re: Problems with KSTEP (two bad channels)
Tom,

Just wanted to update.  Repaired my plug and got both channels up and running again with $20 worth of new chips.  Thanks for the help!

-Andrew
Group: DynoMotion Message: 11561 From: Tom Kerekes Date: 5/24/2015
Subject: Re: Problems with KSTEP (two bad channels)
Hi Andrew,

I'm impressed!

TK