Spread Spectrum Technologies and EMI

Not sure if this is what you are looking for, but spread spectrum magazine has a lot of good links: http://www.sss-mag.com/reftopics.html

One that I found useful was: http://www.sss-mag.com/pdf/1mod_intro.pdf That article goes into the math quite a bit.

Thanks!

Hi,

what are you planning: the different motors not disturbing each other? Hum, noise, vibration, cross-coupling, inter-modulation? Or not to disturb the outside systems?

Any PWM-driven motor will act as a good low-pass filter as the inductance and the inertia will not transmit high frequency components.

But you will need another inductive filter to lower the dU/dt - voltage gradient load - on the insulation of the motor in order to achieve longterm reliability.

This filter should be placed immediately after the power-stage in order to avoid the power lines to emit large disturbances.

Not to disturb the outside: any motor-connecting cables should be shielded, make sure not to introduce earth-loops.

Select a proper frequency and frequency spacing between different motors: if your system is ready for a test then make a measurement how much influence (resonances?) there will be and avoid these by the first 5 to 7 harmonics of your PWM.

Now: look at your sensors in the system, these are most likely to be affected. Test the sensor performance with respect to high frequency response.

Separate any digital system by opto-couplers.

Make sure there is no military test-facility nearby, they will use impulses of some 100 MW in the UHF to microwave region.

Best wishes

RHABE

You it the nail on the head. What I need is the info you just presented in more formal documentation (a book, article, etc.), so I have something more to offer than "I heard it from a guy on the internet".

Well, all I said was common sense, if you're conversing with an electrical engineer with at least a passing knowledge of EMC.

All you need to do is write it up, and if someone questions it, ask them to explain their concern.

I realize it is the culture today, but you don't have to footnote every obvious assertion.

Alternatively, I can write up a learned treatise on the subject, and my name will carry some weight. But I need a purchase order with some hours on it to bill against, for that. :-)

emc_c

Or you could point me to a decent textbook.

No obvious single compact source that would support what I maintain is common sense. The division by "n" is simply common sense. An explanation of the peak detector function may be found in the appendix of MIL-STD-461E/F. But you have to extrapolate from that described performance to get the discussion of the benefit of frequency spreading as being between 10 log n and 20 log n. I performed that explanation for you, so all you need is the -461 appendix discussion (with pictures) which is public domain.

emc_c