UPS Nuisantly Transfers to Bypass 5

[nightfox1925]

I am presently working on a problem with one of the UPS systems in the plant. There are two Single UPS systems (480-208/120) feeding two UPS buses (A and B) separately. The two separate bus circuits feeds a couple of 208VAC-24Vdc rectifiers which in turn individually feed as redundant power supplies to downstream PLCs, SIS, etc.

Of the 24Vdc rectifiers, the 24Vdc, 400A rectifier is the largest. The UPS systems always shifts to bypass everytime the UPS output MCCB is closed or this Rectifier is energized. The UPS alarm shows "Output Overload". We suspect that it was the downstream in-rush which is causing the UPS to nuisantly transfer. An idea by one of the contractor was to put a reactor on the input of the rectifier. A 5% impedance reactor was connected in series but still the UPS keep to transfering to bypass.

It is observed that the UPS output transformer secondary is wye and the downstream rectifier isolation transformer primary is delta...is it possible that the problem is also caused by phase mismatch between the UPS and downstream charger?

is there a possibility that the UPS settings may be too sensitive?

I am gathering all the data so I can post them here soon. But any help or guidance will be very much appreciated. Thank you.

 

[davidbeach]

Wye feeding delta is not an issue. Does the UPS come off bypass with the rectifier on line? You will find that your UPS can be very finicky about what loads it serves. What has the UPS vendor said about the situation?

 

[nightfox1925]

Well, the design/construction contractor tapped on another contractor and they bought these stuff from Mitsubishi while they also bought the rectifiers from Staticon. Coordination is really hard since each one of them tries to cover up their butts.

Well, actually everytime the staticon charger is energized, the UPS transfers to bypass and indicates an "output overload" and after the staticon charger is started, it shifts back to the inverter.

First, they tried to patch holes by using a reactor but did not do any good.

Now they the contractor is suggesting phase shifting isolation transformer between the UPS output and the downstream charger input.

The contractor later on suspected that the downstream charger is in resonance with the UPS meaning, the UPS sees a zero impedance and erroneously see as a fault.

So far, everything was just suggestion and everybody points one finger to another which makes the situation worse for the client.

Do you have or does anyone had experience with this problem. I appreciate any guidance I could get.

 

[davidbeach]

So, it is in bypass momentarily for what should be a rare event; is that actually a problem? You may well find that any "cure" is worse than the "disease". The best "cure" would be a bigger UPS.

 

[nightfox1925]

The concern that the plant is having is when there is a plant wide power failure and the UPS may not transfer the system to the batteries since it is sensing a system fault at it's output.

 

[nightfox1925]

Is there really an issue or disdvantage if a UPS output is wye and feeds a downstream rectifier with its isolation transformer in delta...The contractor is claiming that the phase shift mismatch is causing the problem..but I am doubting (unfortunately, I can't directly explain why it isn't so).

 

[davidbeach]

Nope, no issue at all between a wye source and a delta load; millions of installations around the world have a transformer wye winding feeding the delta winding of another transformer.

 

[nightfox1925]

We are planning to rent a high speed recorder and do the energization again hoping to measure the amount of in-rush. They showed me a Fluke measurement of 20A..but I think the fluke meter is not fast enough to catch a 0.1 second in-rush. The Contractor's consultant is insisting for an isolation transformer between the UPS output and rectifier input to match the phase shifts but we have held this up for further investigaton.

If the UPS is recording "output overload" this is definitely due to the rectifier in-rush during energization or because of swithing transients right? harmonics is not an issue due to the rectifier's isolation transformer. is there any other factors that I am not aware of?

 

[waross]

Have you tried a simulated power failure? If the UPS will handle a power failure you should be okay. I would follow Davids suggestions. If the UPS hangs and won't pick up the load during a power failure you must find a solution.
The thing to do now is to throw the main switch and see if the UPS can handle a power failure. That should be done as part of the acceptance testing.
Who's paying for the tries in the dark? Renting the proper test equipment may be cheaper and quicker than guessing what power equipment to buy next.
In the meantime and to keep the contractor diverted for a day or so, you may also try energizing the transformer with the rectifiers disconnected.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[nightfox1925]

Hmmm thanks waross and david, I'll arrange for that. I will keep you guys updated with this.

Any more suggestions will be appreciated.

 

[rcw retired EE]

Based on my UPS installation and maintenance experience, I think the inrush current to the charger is too high for the UPS. High inrush pulls down the output voltage and the UPS does what it is supposed to do - transfer to bypass to handle the inrush current (or the fault), then pop back to normal configuration when all is well.

The alternative, as mentioned, is a bigger UPS, or a ferro-resonant style which is a little more tolerant of inrush. (But that design has other issues.)

I can't see where the phase difference has anything to do with yoru problem. The downstream equipment has no way of telling what phase it is getting. There is a remote, but unlikely, possibility that the parallel rectifiers have insufficient filtering on the output causing a "dip" or gap in the DC waveform that coincides with a peak of the subject rectifier/s output waveform. So the rectifier supplies more than its share of the current. Usually, the DC power supply has a "walk-in" circuit to ramp up to full voltage output which would make this theory invalid. Also, the rectifier would be overloaded all of the time.

What is the size of your UPS, its current limit setting and the rating of the DC power supply that is causing the problem?

Another thought - if the UPS is transferring on inrush now, it will probably transfer when it tries to energize the interposing transformer also. Putting a transformer between the UPS and the charger may just change the location of the inrush and not solve the problem.

 

[nightfox1925]

Thank you rcwilson. I am also going to check if the UPS was also properly sized. Normally, the UPS size should be calculated to consider both normal loading and during transient peak load conditions right?

I am in the process of preparing some field request to obtain all information needed to evaluate the system. I will verify if the UPS was functionally tested on mains supply outage. I will also going to get the UPS settings, operations, manual and vendor drawings to see what kind of design we are dealing here.

I will keep the forum updated with this.

For the meantime I appreciate any technical guidance I could get. Thank you for all the support.

 

[nightfox1925]

I also would like to request any helpful links about related to our subject matter to have a quick grasp of the technical aspect of our subject matter. My deepest thanks for the assistance. I will keep everybody updated.

 

[nightfox1925]

is the in-rush possibly mitigated using series reactors? I know that series reactors are used to bring the short circuit current, but will it help for this application?

As I know, the series reactor will again be seen as an additional load by the UPS during starting and the starting voltage drop Ist*Z will even pull down the voltage at the UPS output and it will also consume power in form of a loss Ist^2 *Z

I am thinking that this will aggravate the problem rather thah mitigating it.

 

[nightfox1925]

here is the SLD sketch:

 

[waross]

The series reactor increases the impedance of the circuit. The increased impedance reduces the current drawn. Additionally, it acts in phase with the magnetizing current and has the most effect reducing the magnetizing component of the inrush. The reactance and the reactive voltage drop is at aphase angle of 90 deg to the real current and may not have much effect on load inrush. Hence the suggestion to try to energize the rectifier transformer with the rectifiers disconnected. If you can energize without the rectifiers connected then you have a load inrush rather than (or in addition to) transformer inrush. First identify the problem, then look for a solution.
If you are able to perform this test and get back to us, we can help with a solution.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[nightfox1925]

Thank you waross. I will definitely do that.

 

[sibeen]

nightfox, a few questions

Do both of the UPSs have the same problem?

What is the size of the UPSs and who is the manufacturer and model number? This would help as we could then look up the specs on the unit.

As to the rectifiers, do you have a single line of the units. Knowing the topology of the units will help out.

To me, the suggestion of a phase shifting transformer just seems ridiculous. Someone is definitely grasping at straws.

I'd be getting a reasonably fast oscilloscope with a decent 'roll mode' and examain the level of the inrush current.

 

[nightfox1925]

Yes, its odd to look at the phase shifts as a culprit. The phase shift matching is significant if you are paralleling two transformers.

I will be able to get the data soon and keep this forum posted. I have requested all vendor info and site info as well. Thank you for all the support.

 

[waross]

Yes,there is a phase shift of 30 degrees between wye and delta on the same system. But, this is an issue when paralleling transformers. How would the invertor or the rectifiers sense a phase shift? The phase shift will be inherent in a rectifier designed to be delta connected to a wye network. Not an issue.
Feeding a delta load from a wye inverter may be a problem if the inverter has a problem with line to line loads but apparently the load is not an issue, just getting energized is the issue.
Another issue that we haven't considered is the original reactor. We have assumed that because a reactor was installed that it was properly sized. Possibly a reactor is needed but the original reactor may be wildly inappropriate.
It happens when the "Hope, Faith, and Charity" method of engineering is used.
(I Hope that this will work, but whatever, I have Faith that through the Charity of the owner I will be paid.)
Nonetheless, please continue testing. If you can get the specs on the reactor it may be helpful.

Bill
--------------------
"Why not the best?"
Jimmy Carter