Phase loss on input to vfd

[philly]

We have an application where we are pulling in a starter to supply power to a vfd. The starter has an MCP breaker and solid state overload device. (Prob not code compliant). When we want to start the motor we pull in the starter which then poweres up the vfd and in turn starts to motor up to a set requency.

When we pull in the starter however the solid state overload device is taking it down on a "phase loss". We disabled the phase loss detection and everything works fine. Current is seen on all three legs.

Any ideas what could be causing this phase loss on the relay ahead of the drive? The only thing I can guess is that maybe the drive was set up for single phase.

The only other thing I'd guess would be the internal drive capacitance or harmonics?

 

[winnie]

I don't know the details of the phase loss protection, so I am not even guessing here, just throwing out a datapoint:

If you look at the input current to a three phase rectifier, the current flow is always between the two phases that have the greatest voltage difference between them, with no current flowing from the third phase. Which two phases are at the greatest voltage difference will change 360x per second, so over time each phase participates in supplying the load.

This is completely analogous to the fact that for a normal linear load each phase will have its current drop to zero twice per cycle, the perhaps significant difference being that the duration of 'zero current' on each phase will be a couple of milliseconds.

-Jon

 

[Jraef]

I don't know the details of the phase loss protection, so I am not even guessing here, just throwing out a datapoint:

If you look at the input current to a three phase rectifier, the current flow is always between the two phases that have the greatest voltage difference between them, with no current flowing from the third phase. Which two phases are at the greatest voltage difference will change 360x per second, so over time each phase participates in supplying the load. ...
-Jon

Well stated, and I think this is likely the basis of your problems. It all depends on the technology used inside your SSOL to detect phase loss. Most "self powered" SSOLs, meaning those that do not have a separate voltage input, look for a severe current imbalance to determine a phase loss. A typical design criteria is one phase having <20% of the current of either of the other two. Then there is an issue of sampling rate and RMS calculations; the sampling rate might be low and/or the RMS algorithm assumes a clean sine wave. Add to this the current harmonics and non-linearity of VFD input current draw mentioned by Jon, and it's easy to see why you might get false tripping. By the way, I know of NO SSOLs that say they can be used on either side of a VFD.

PS: Yes, you are correct, the MCP / SSOL feeding a VFD is not code compliant. The SSOL is also redundant. Change the MCP to a standard TM breaker sized for 125% of the VFD input current and jettison the SSOL. Also double check your conductor sizes.

 

[Cold Fusion]

... Yes, you are correct, the MCP / SSOL feeding a VFD is not code compliant. ...

What code section(s) would this be?

I looked through (2008) 430 part III, IV, V, and X. A mag-only CB and an overload wouldn't be my first choice. But I didn't see anything that would preclude it.

cf

 

[Cold Fusion]

...Most "self powered" SSOLs, meaning those that do not have a separate voltage input, look for a severe current imbalance to determine a phase loss. ...

philly -
My experience is all with AB E3+. These have a separate phase imbalance from current imbalance. The phase imbalance is watching for voltage imbalance and the current imbalance is watching for .... you get it The phase imbalance part is not watching for current imbalance.

There is a setting for the amount of imbalance and a timedelay setting. The only use I have ever figured out for a phase imbalance is if the supply is using fuses for
OCP. If you don't have fuses upstream, it's pretty useless. I'd do what you did and shut it off. However, I likely would shut off all the overload protections except the over-current. The Drive likely has all the protective functions it needs. Set them up right and let the drive take care of its self.

If you feel you need some of the overload protective functions (besides overload) then get out the book and check the programming. None of us here can help with that. We don't know the overload brand or model, nor the drive brand or model. Anything any of us would say, is pure conjecture - or flat wrong.

That being said, if the drive has capacitive front end, maybe the inrush is getting to the phase imbalance relay. If your customer thinks you have to have that one in service, I'd suggest programming in some delay.

cf

 

[philly]

philly -
My experience is all with AB E3+. These have a separate phase imbalance from current imbalance. The phase imbalance is watching for voltage imbalance and the current imbalance is watching for .... you get it The phase imbalance part is not watching for current imbalance.

There is a setting for the amount of imbalance and a timedelay setting. The only use I have ever figured out for a phase imbalance is if the supply is using fuses for
OCP. If you don't have fuses upstream, it's pretty useless. I'd do what you did and shut it off. However, I likely would shut off all the overload protections except the over-current. The Drive likely has all the protective functions it needs. Set them up right and let the drive take care of its self.

If you feel you need some of the overload protective functions (besides overload) then get out the book and check the programming. None of us here can help with that. We don't know the overload brand or model, nor the drive brand or model. Anything any of us would say, is pure conjecture - or flat wrong.

That being said, if the drive has capacitive front end, maybe the inrush is getting to the phase imbalance relay. If your customer thinks you have to have that one in service, I'd suggest programming in some delay.

cf

The SS overload device I am referring to is indeed an E3+ relay. It has a 24V power supply and I believe you are correct with the "Phase Loss" feature looking at voltage imbalance.

I'll have to read up on non linear loads as a result of full wave bridge rectifiers to see how this could cause this unbalance to be seen as other have suggested.

I belive the reason this setup would not be code compliant would be due to the fact that an MCP/Overload combination in a starter is only rated and UL listed for use with a motor only application. But Ill let the code experts chime in on this.

 

[ptonsparky]

Look into the installation instructions and I believe you will find that solid state relays should not be used with VFDs. When used overlaods should be the old bimetal or melting alloy. Just take the overload out, you don't need it anyway.

 

[Cold Fusion]

... Just take the overload out, you don't need it anyway.

It's a listed combination starter - overload goes, it all goes.

cf

 

[ptonsparky]

I agree, get rid of it all. Don't need it.

 

[Cow]

Sometimes when a customer decides to change from a starter to a VFD, we've simply pulled the line/load leads off the starter and wire nutted them through straight to the VFD. Then we'll wire the VFD's run signal through the starter. Starter pulls in---VFD starts. Supposedly, it's better for the VFD also, rather than killing the input power to it.