DHambley
Electrical
- Dec 7, 2006
- 246
I have an EMI problem to solve. Our 3-ph motor cables have a sawtooth waveform of ~80A pk-pk at 20kHz and ~300V pk-pk square wave (ignoring the low frequency sinusoidal component). The motor is 2m from the driver, with the three cables ~2m long and 5cm spacing each (three side by side). This cable bundle is ~20cm average above chassis. Before coming up with solutions, I need to model the present system and predict what the E-field in dBuV/m will be at 10m distance. This is to show a predicted before-and-after. The radiated fields will be from several sources: the voltage square waves, the current triangle waves and the common-mode current to the motor capacitance. The large currents in this loop have me concerned. The closest equation which I can find in my old textbook is:
E = 120*pi^2*I*A/(wavelength*D), A is cm^2, I is Amps, D is distance to measuring antenna, and the author doesn't say what the units of E is. For a quick estimate I would like to just get a few values at say, 100kHz, 2MHz, and 30MHz assuming the harmonics from the original triangle wave with a fundamental of 20kHz. I'm also assuming we'll be in the ballpark by using 750cm^2 as the average area between each cable and the other two.
Question: Any better formula than the one I've shown above which is in units of dBuV/m?
We need to meet FCC part 47 class-A, industrial location. I have used various solutions in the past but, with much lower current levels. The advice in the thread, < from Mark's comment has been useful. The solution will probably be a combination of several of the following methods:
- Shielded cable for each of the three cables (at both ends)
- shield around the three cable bundle
- Cap + snubber from battery return to chassis at the driver
- large ferrite around cable bundle
- cables pulled closer together and closer to chassis
The formula to estimate the current radiated signal strength is what I'm trying to pin down first though.
thanks for the help
E = 120*pi^2*I*A/(wavelength*D), A is cm^2, I is Amps, D is distance to measuring antenna, and the author doesn't say what the units of E is. For a quick estimate I would like to just get a few values at say, 100kHz, 2MHz, and 30MHz assuming the harmonics from the original triangle wave with a fundamental of 20kHz. I'm also assuming we'll be in the ballpark by using 750cm^2 as the average area between each cable and the other two.
Question: Any better formula than the one I've shown above which is in units of dBuV/m?
We need to meet FCC part 47 class-A, industrial location. I have used various solutions in the past but, with much lower current levels. The advice in the thread, < from Mark's comment has been useful. The solution will probably be a combination of several of the following methods:
- Shielded cable for each of the three cables (at both ends)
- shield around the three cable bundle
- Cap + snubber from battery return to chassis at the driver
- large ferrite around cable bundle
- cables pulled closer together and closer to chassis
The formula to estimate the current radiated signal strength is what I'm trying to pin down first though.
thanks for the help