VFD Input Fuses Blew, Suggestions?

I always like to go around with a temperature gun or ir screen to check for hot spots while the equipment is running....secondly you can monitor and record the power supply, and possibly the characteristics of the individual incomers....and always check the grounds for true grounding properties....then there is always the possibility of an anomaly that is a random occurrence...then just watching it to see what happens next, if anything, will tell the tale....In any case monitoring the power supply for a while would be routine in case further problems warrant another conversation with the drive manufacturer...eh?

Were your other VFDs running loads at the time? Do you have a line reactor ahead of this drive? If not, how large is the transformer feeding this service? Are there line reactors ahead of the other drives?

The point I'm working toward is that there is a scenario in which this can happen, depending on these issues. I can explain it after you answer them (and I'm at a computer instead of my phone).

Hi, No line reactor there

The transformer feeding this bus is huge. we have a 46Kv service, im not sure of the power on the transformer but it is large.

Could you please explain some more to me ?

Thanks

Those fuses are to protect the rectifier components, they are supposed to clear the circuit before the SCR passes enough current (over time) to damage the component.

If your ac line sensing circuit misses a zero crossing, or (their guess) there was enough noise on the line at that particular drive to trick the circuit into missing the timing, one of your SCRs might have fired directly into one of the others already conducting, creating a line to line short circuit. That's why the drive manufacturer blamed the utility, but it could have been quite local to the drive, or may have tricked one and not tricked the others, depending on phase angles and bridge demand at that particular millisecond.

A shot like this could be 1 in a million, or might indicate the need for anti-line notching reactors as JRaef indicates, quite commonly installed when designers fear the worst, especially with multiple SCRs on a common lineup.

Could you explain a bit more in a dumbed down manor?

Im not a drive expert, im usually a protections guy

So there are no line reactors on this setup.

I think i gather what your saying. if you have time to write again that would be great.

What im hearing so far is we are stupid we dont have line reactors. Fortunatley i didnt do the initial engineering. Im just helping clean up the mess.

the very expensive mess i might add

When an SCR fires, the current to charge the capacitor is like a short circuit, it can pull the voltage down that the line sensing circuits use to decide when to fire the SCRs.

If the circuit sees what it thinks is a zero crossing, it can then fire one SCR while another one is still conducting, making a short circuit line to line, and popping the protection fuses.

One or two drives can be OK, but adding more on the same power system bus could tip the scales towards needing to add some anti-notching reactors.

After your excellent suggestion, i have done some more research.

At this point im thinking you are 100% correct with the cause.

Thank you very much for all the info.

the transformer is definatley much larger then 10 * the Kva of the drive

Since we are on this topic. How does one determine the Kva of a drive if the power factor is unknown?

Are you just assuming a Pf of 1 and assume a worst case scenario like your FLA of the transformer?

say 200A drive *480V = 96kva Drive?

The reason you use kVA is because you DON'T need to worry about power factor going into the drive. This is a rule of thumb, so it's kept simple. A (input) x V (line) x 1.732.

And the input impedance of a VFD is tricky. But the impedance of the RECTIFIER alone is virtually nil, hence the issue of the capacitors attempting to charge themselves at the available fault current level. That's why all VFDs have what's called a "pre-charge circuit" that puts a resistor in series with the caps for the first second when you initially energize it, acting as a current limiting resistor. That is then shorted out of the circuit once the caps are charged, basically 1 second or so. So once the VFD is already powered up, that is no longer there and hence the brief transient dip in voltage allows the caps to pull that high current again, unimpeded. That's where the line reactor plays its part.

I guess the other thing im curious about is, what is the input impedance of a DC inverter like this?

Given the fact that the impedence on the input side is really just the SCR?

Thanks