When nothing else works, it's wise to go back to the beginning. After so many suggestions, recommendations, and disagreements we have not solved this problem since last year.
While grounding could be a major issue, it's not necessarily so in this case. As long as there is star wired ground. Besides ground loops don't always add up, sometimes they subtract too. Of course, we are not at the broken system so troubleshooting is obviously difficult. On 12/26/2015 07:16 AM, John Thornton wrote: > I just checked the ground to housing on the VFD filter, VFD, computer, > and the smaller filter and they are all internally connected to ground. > The VFD is controlled via modbus so I don't know how to check that cable > (which is just a phone cable with a phone jack on one end and a DB9 on > the other end to connect to the computers serial port, it's a Automation > Direct cable not home made. > > Is injection to the mains coming back to the machine somehow? > > JT Troubleshooting was mostly focused on ground and wiring so far. Modbus is where the (noisy) action is so it's the most obvious place to start troubleshooting. Not being familiar with modbus I decided to do some reading on it. https://en.wikipedia.org/wiki/Modbus https://en.wikipedia.org/wiki/RS-232 If I understand it correctly, there are 3 physical means to carry modbus: RS-232, RS-485, and ethernet. RS stands for Recommended Standard, which never became a real standard as far as I'm concerned. I've seen RS-232 implementations for mainframe and PC computer peripherals and ham radio ranging from ±25V to ±12V and what's most common today, ±5V. RS-232 is relatively immune to noise, for speeds we are dealing with here, as long as one side of shielded cable is grounded. We used to have cable runs between mainframe computers and terminals in other buildings with (to a degree) different ground and no major issues. After some search, I ended up reading this document, Modbus For Field Technicians: http://www.modbusbacnet.com/includes/pdf/MODBUS_2010Nov12.pdf see pages 29 and numerous pages dedicated to RS-485 issues. I am assuming that your implementation is based on either RS-232 or RS-485. At this point it might be worthwhile to look into VFD "black box" to see how exactly it connects to the outside world. Time to check: - (HW) drivers, ICs, transistors, local caps, voltages [1], - configuration (parity, etc.), - use RS-232 breakbox and check signals quality with scope - try different RS-232 port on PC or replace it if necessary. - any related jumpers, if any [1] optical connection protects you from different voltage levels but that was not your first try I think so the port on PC side could be damaged if the VFD side uses 12V for example. It depends on how old the VFD is. Unknown: - was VFD/modbus fully functional before rewiring BP? - What says the VFD maintenance manual? - Can you run standalone tests on VFD? How about replacing the DC motors with large resistors and run the tests? One by one or all. Are those brushless motors? What's the noise around motor wires with a motor or with a resistor? -- Rafael ------------------------------------------------------------------------------ _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
