Siemens variable frequency drives and RF noise


Thread Starter

Dan Kalla

Looking through your e-mails I found something about IEEE having information available on RF noise. We're having problems with Seimens Variable Frequency Drives emitting noise and confusing our RS-485 signal to & from the I/O racks in a few pieces of equipment we currently produce. I would like to find material explaining what/how/where and what to do about it (including calculations). Can anyone help? D. Kalla Electrical Detailer PDQ Manufacturing, Inc (920)983-8333 ext 3537 [email protected]

Scott Whitlock

> I would like to find material explaining > what/how/where and what to do about it > (including calculations). Can anyone help? RF noise problems are tough. All I can give you is a starting point: The road to RF suppression is long, and you must walk much of it on your own. :) Truly, it's a black art. But if you're lucky, an easy fix is to just put some ferrite beads on the VFD's leads... Good luck, Scott
IEEE Std 518-1982 is entitled: IEEE Guide for the Installation of Electrical Equipment to Minimize Electrical Noise Inputs to Controllers From External Sources. Noise is a complicated problem and there aren't easy fixes. I usually suggest that you suppress the source of the noise. In this case, the motor. Tom
Dan, Noise issues with VF drives of any manufacture require a disciplined approach to system design and installation. Taking time to do the following with regard to the PWM voltage waveform sources will go a long way to solving this very common problem. PWM Motor leads should exit from the drive and be routed at least 10 inches from any non PWM control or power conductors. Best scenario would be to put them into a grounded ferrous metal trough or conduit. Optimally, conductors should be twisted with at least 1 turn per inch. PWM Braking resistor leads should be treated the same as PWM motor leads. Motor ground should be continuous from motor location to drive frame. Drive frame should be grounded at single point ground in cabinet with low impedance earth ground. Motor leads should be run in a dedicated conduit that is not shared with any other conductors. A ferrous metal shield between the drive chassis and any control related hardware often solves many noise induced problems. RS485 communications should be run in a daisy chain topology rather than a star topology with terminating resistors at the ends of the network per manufacturers recommendations. We have found that taking these basic steps prevent 99% of the noise related problems with PWM waveforms generated by inverter based drive systems. Ken Brown Applied Motion Systems, Inc.

Hakan Ozevin

I can advise: 1. Decrease the switching frequency of the VSD's using P076 2. Decrease the baud rate of your communication system 3. Use RFI filters for the mains connection of the VSD's. 1 and 2 are easy solutions with no cost, but 3 has much effect with some cost. If you cannot afford it please let me know. I can advice more, but I have to know details about your hardware configuration. Good luck Hakan Ozevin
Also, you do not want to run AC in parrallel with DC (e.g. do run RS422, RS485, or DC control lines with AC). When DC crosses AC it is best to cross at a 90 degree angle. [email protected]