Regular occurrence of Drive3 fault

[danieldejager]

Hi, i'm hoping that someone can shed some light on an occasional issue I keep seeing with my setup.
Sometimes when starting up, I get a Drive3 fault, and the only way to solve it is to just shut everything down and restart.
I also have an issue where if I use the ESTOP, I get into a similar state.
Toggling the VCP RESET seems to clear it, but the moment I try to get the drives to power up, the system halt with a Drive 3 fault.
I have replaced cables, and IIRC I think I've also swapped the order of the drives, and it's always DRIVE3. I can look at the drive's logs with Motion Studio and nothing shows up.
This is the last thing to button up before calling this build complete.
 
Components used:

• Hickory
• WMPG-6
• 3x Leadshine EL7-750EC Drives
• 3x Leadshine ELM2H-0750LA80F Axis Motors
• LENOVO_MT_10MY_BU_Think_FM_ThinkCentre M910x
  • Intel(R) Core(TM) i5-6600 CPU @ 3.30GHz 3.31 GHz
  • 16GB Ram
  • 512GB SATA SSD
  • Win 10
  • CNC12 5.10

Build Post: viewtopic.php?f=57&t=10526

[centroid467]

Hello,
That looks like a nice build! Thank you for the schematics and report. I believe we have run into this before and came up with a fix but it is not in TB323 yet.

Please upgrade to the latest release version of CNC12 and set CNC12 parameter 365 to 250. This will set a delay to make the system start up more reliably.

[danieldejager]

I am already on 5.10 and it seems to be the latest version available here: https://www.centroidcnc.com/centroid_di ... loads.html
Is this the right place to be looking?

Parameter 365 was already set to 250, so I bumped it up to 750.
The problem still shows up, though maybe less frequent.

[centroid467]

I am already on 5.10 and it seems to be the latest version available here: https://www.centroidcnc.com/centroid_di ... loads.html
Is this the right place to be looking?

Parameter 365 was already set to 250, so I bumped it up to 750.
The problem still shows up, though maybe less frequent.

Hello Daniel,

I believe your issue will be resolved by some changes to Hickory firmware that were made in CNC12 V5.20 which we are aiming to release very soon. You can also use the latest RC build as posted by Keith in this thread viewtopic.php?f=67&t=10491.

The firmware changes fix some communication timing issues that appeared with certain drives when they are in the third position in the chain.

[cnckeith]

I am already on 5.10 and it seems to be the latest version available here: https://www.centroidcnc.com/centroid_di ... loads.html
Is this the right place to be looking?

Parameter 365 was already set to 250, so I bumped it up to 750.
The problem still shows up, though maybe less frequent.

Hello Daniel,

I believe your issue will be resolved by some changes to Hickory firmware that were made in CNC12 V5.20 which we are aiming to release very soon. You can also use the latest RC build as posted by Keith in this thread viewtopic.php?f=67&t=10491.

The firmware changes fix some communication timing issues that appeared with certain drives when they are in the third position in the chain.

v5.20 has been released.
https://www.centroidcnc.com/centroid_di ... loads.html

[danieldejager]

I've tried both of the updates since the last post here, to no avail.

I started thinking about it, and I was wondering if there's some weird race condition between the PLC and the drives' STO state that is getting things in a tizzy, so I bypassed the STO system, and things seem to be working correctly now.
Though, as I'm writing this, I am also recalling that I am running a custom PLC program, so I am questioning if the software updates actually applied as expected.

[centroid467]

I've tried both of the updates since the last post here, to no avail.

I started thinking about it, and I was wondering if there's some weird race condition between the PLC and the drives' STO state that is getting things in a tizzy, so I bypassed the STO system, and things seem to be working correctly now.
Though, as I'm writing this, I am also recalling that I am running a custom PLC program, so I am questioning if the software updates actually applied as expected.

At first I thought a custom PLC would have some effect on the enable delay parameter (P365) but then I remembered that delay is handled in the DSP itself rather than the PLC. So, I do not believe your custom PLC is the issue unless something was written very wrong.

A weird race condition sounds possible - the delay was an attempt to account for the time it takes for the drives to change state from off due to STO to active and ready to run. The 5.20 update should have brought in a fix that improved the communications synchronization specifically with the third drive in the chain.

You can verify that it took by looking at the date on the file "MPU12.hex" in your CNCM directory and comparing it with when you installed that software version.

[eng199]

For Leadshine, p981 delay is needed.

Check for a line like this in the PLC:
IF SV_MACHINE_PARAMETER_981 > 0 THEN MasterEnableDelay_T = SV_MACHINE_PARAMETER_981

If this has been dropped from the custom PLC, copy over the MasterEnableDelay_T related changes from the default HICKORY PLC.