VFD to PLC via CANopen issues

[PegLeg03]

Hi there,

I'm new to this forum, and am relatively inexperienced with VFD's, so you'll have to excuse my (likely) ignorance.

The long and the short of the issue is as follows:
- I am integrating a machine that uses 2x 400V AC VFD's (TechTop) to drive two motors that do the 'heavy lifting' of the machine, with a drive system that uses 2x DC brushless motors (Curtis Instruments) for differential steering. Both are controlled centrally from a PLC (TTControl) via CANbus.
- When the VFD's are disconnected from the CAN network and are driven off the terminals manually, the rest of the CANbus nodes have no issue. When I plug the VFD's into the CANbus, very soon after the motors driven by the VFD's are run, it trips out the bus.
- I have operated the VFD's in isolation off a PCAN USB adapter. This tripping out behaviour is consistent when the CAN_GND from the VFD RS232 terminal isn't connected to the CAN_GND pin on the adapter. It seems to be consistent when it is connected, which leads me to think the GND's haven't been wired correctly on the machine.
- The PLC doesn't have a CAN_GND designated pin, unlike the VFD's RS232 port and the Curtis' Molex plug, but I am of the belief it gets it's reference from BAT-.
- The power system seems complicated -> the PLC and Curtis motor controllers are powered from a DC battery, which is charged by an AC converter, whilst the VFD's are powered straight from the 3 phase mains.
- Our company doesn't have any experience integrating high and low voltage systems (we don't usually touch AC systems), so this issue provides somewhat of a challenge to us. We have tried contacting a technician/engineer from TechTop, but no one seems to know what they're doing.
- When we put a scope on the bus, there was around a 60V peak-to-peak voltage/noise when the VFD's were running, which leads me to think there is also an issue with the VFD shielding. No one is an expert on this either, and likewise haven't got anything from TechTop to help.

Does anyone have any experience integrating such components? How do I go about tackling GND references? Can/do I connect DC BAT- to chassis (earth) or could this cause issues? Any help would be appreciated : )

Lucas

 

[jraef]

First idea that jumps out at me is that under CANbus control, the reference command values are changing too fast for the drive to keep up with, so by constantly increasing and decreasing commanded speeds too fast, you end up with decel energy backing up into the DC bus. This would be something typically handled by buffering of the command values. If someone "rolled their own" CANbus interface, they may have not factored that into the design. Just a guess though...


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden

 

[LionelHutz]

Try using isolators.

 

[Heaviside1925]

OP,
Your mix of acronyms made my head spin a little. I first want to get my head straight on some terminology.
When you say trips out the bus, is this a local bus, DC bus?
When you refer to high and low voltage systems, please clarify what you are considering high and low voltages.
Who designed the electrical and comms system? Can you share schematics?
You mention no one at your company are experts on VFDs nor integrating what you call low and high voltage systems, so not to be condescending but why would you try to build something using technology that nobody knows how to install?
All that said, based on your troubleshooting it sounds like your grounding is an issue throughout the system but that goes back to having a proper grounded and shielded system. Could your company contract an experienced EE to come assist you in design and troubleshooting?