TVSS on UPS output

[txg84]

On of the building on the campus has been having issues with it's UPS. I decided to install a Dranetz on the output of the UPS to monitor it. The UPS is 225kVA, 480V Input 208Y/120V output. The load on the UPS is about 200A but there have been current spikes at diffrent times.

The first spike was during normal operations on a single phase. It was 800A and lasted about 2 cycles. The second spike happened two days later during an outage. The spike was about 2000A on all three phases for about the same amount of time. I have not been able to determine what is causing these spikes.

The distribution panel fed by the UPS has a TVSS unit. Might this be causing the spikes?

Thanks

 

[jghrist]

Was there a voltage transient at the same time as the current spike? If the UPS output increased above the clamping voltage of the TVSS, the MOVs in the TVSS might have conducted briefly, causing a current spike.

 

[txg84]

There was no voltage transient shown in the data. The UPS feeds servers and computers. None of the breakers tripped and they did not have any damage to equipment.

I do not have any idea what might be causing the current spikes. Has anyone had experience with TVSS units "operating" under normal conditions? Could the spikes be caused by the UPS?

Thanks

 

[alehman]

It seems very strange that there would be such a current spike with no associated voltage disturbance. I suppose it's possible the TVSS did it's job and clipped a voltage pulse below the Dranetz trigger point.

I've never heard of a TVSS causing a disturbance on it's own. Also it seems unlikely the UPS initiated anything like that, but I suppose stranger things have happened.

Are there any static transfer switches fed from this UPS?

Alan
“The engineer's first problem in any design situation is to discover what the problem really is.” Unk.

 

[txg84]

There are no static transfer switches fed by the UPS.

The UPS should have transfered to bypass or shown some type of overload condition when these spikes occured, but there is nothing on the UPS log. This is why I am leaning towards the spikes being caused by the UPS.

 

[ghostbuster7]

1.Is the U.P.S. a full on-line unit or is a off-line(line interactive unit)?

2.Have you checked the U.P.S. logs during that time frame?

 

[txg84]

Ghostbuster7: If by on-line you mean that the output power is normally in-line with the rectifier and inverter, then yes the unit is a full on-line unit.

Also, I have checked the log file during the time frame that the Drantez recorded the current spikes and have not seen anything. I was reading the UPS manual and it stated that:
"Momentary overloads from 155% to 1000% of rated current is supplied by pulsing the solid state static bypass switch for 40 milliseconds. If an overload condition exceeds 40 milliseconds, the UPS module continues to supply power to the critical load if it remains within the current-versus-time curve of overload capacity. If the system overload capacity is exceeded, the UPS control logic completes an automatic transfer to the bypass"

Since the current spike have been short duration, usually around 2-cycles, the UPS might have been able to withstand them without going to bypass. That might be the reason there is nothing on the log file.

 

[sibeen]

txg84, you've stated that you're having issues with the UPS but haven't stated what they are, other than to say that you are measuring some current spikes.

What prompted you to put a dranetz on the output of the UPS in the first place?

Knowing this may give posters here a better indication of what may be the potential problem and further advice could flow from that.

 

[txg84]

Sibeen, the UPS at this building failed a couple months back. The SCRs for static switch failed while the UPS saw an overload condition. The UPS could not transfer power to bypass and shutdown. The SCRs were replaced but the cause of the failure was never determined.

 

[sibeen]

I would also be monitoring the bypass line to check to see whether an output current spike is seen by the bypass. If so, this would indicate that the bypass thyristors are firing to help clear the downstream overload.

I'd say the likelihood of it being the downstream TVSS causing the problem would be extremely, extremely low; especially as you have detected an over-current on all three phases. I'm assuming here that the MOVs are connected Line to Earth, and not some weird delta arrangement.

When you state you have checked the UPS log, is this just the alarms shown on the screen or outputed to some other monitoring system. I have seen UPS systems which will not generate any alarms, visible to the user, when the bypass has been initiated to clear a downstream fault. The event will have been logged down in the software that is only available to the UPS service tech and not the end user.

Where exactly do you have the dranetz connected on the output?