Frequency inverters and interference on the power supply side, sizing alternators

[MartinLe]

I'm not an electrical engineer. Recently I learned that gas powered cogeneration plants (common on wastewater treatment sites) are poor replacement for a Diesel engine for an emergency power supply, and about another problem:

Frequency inverters produce high frequency signals on the side of the power input. This is relevant when designing an emergency power supply for a site with lots of VFD, apparantly it is advised to install far bigger alternators so the alternators don't get damaged by these high frequency voltaes (So for 100 kW worth of VFDs you would install a 200kVA generator on a 100kW engine). These high frequencies are explained by how an FI works:

The "switches" switch with the output frequency or far faster, to create the desired frequency and output voltage via pulse width modulation. This rectangular wave form of course creates high frequency signals.

Now here are my questions:

The DC Bus is mostly a capacitor, so it should essentially short circuit high frequency signals. Why, then, do we have high frequency signals on the input side?

Should it not be relatively easy to either separate the FI from the power supply with a band pass (50 or 60Hz) to block high frequencies, or protect the alternator in a similar fashion?

 

[davidbeach]

Are you mixing up the noise on the motor side, due to the PWM, with the noise on the input side due to the rectification?

I’ll see your silver lining and raise you two black clouds. - Protection Operations

 

[waross]

Your first graph shows the input three phase voltages.
The second graph could be the input current.
You can see the current clipping. Each phase only contributes for about 60 electrical degrees per half cycle rather that 180 degrees.
This clipping, or the abrupt transition from zero current to high current and 60 degrees later. back to zero current causes steep wave-fronts that generate harmonics in the alternator.



--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[dpc]

This issue has been around for decades - since the first AFDs became common. There were some issues with older engine-generators, in particular the old voltage regulators. Local generators have been successfully applied at WWTPs with heavy AFD loads for quite awhile. I suspect an internet search will turn up quite a few papers. I'd recommend talking with reputable engine-generator suppliers to see if they have any specific recommendations. It is always advisable to specify the percentage of load that will be non-linear, and put the burden on the generator supplier.

 

[MartinLe]

davidbeach and waross, you are right and I misunderstood where the problem is coming from.

I found this:

https://www.trane.com/content/dam/T...lectrical-considerations/admapn019en_0106.pdf

I don't understand quite *why* a bandpass (Or low pass) is not used, if I understood the paper above correctly it is because they would change the power factor of the VFD?

 

[jraef]

All “non-linear” loads (the term for what waross described) cause this distortion. You can add inductors (line reactors) on the input to help mitigate that, as well as a DC bus inductor (choke), or you can use “Low Harmonic” versions of the VFDs, and/or you can apply Active Harmonic Filters or Passive Harmonic Filters ahead of them. But all of these things take up space and increase the cost of the system, so are often unfortunately designed out to save money.


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

 

[TugboatEng]

Look into PMG excitation if you are driving inverter loads with a generator.