Power Quality Issues

[burntcoil]

We are facing issues with our 90kVA UPS which is more than 25 years old. We connected Fluke 435 to the output of the inverter which recorded multiple events of Dips and Swells. We shifted the Fluke 435 to 480VAC UPS input side and recorded more than 6000 transients events on the fluke within 24 hours. These transients were between phase A and C and their magnitude was 100V above the normal voltage. UPS output was not effected and no event was recorded in the DCS. Before these transients there were several events recorded by DCS for the Static switch transfer although in real there was no transfer and system was operating normal. Please advise what to investigate.

 

[Skogsgurra]

Is there a reason why you connected the 435? Or did you connect "just to see"?

A UPS that old probably doesn't have a pure sinewave output voltage but contains lots of PWM or PWM residues, which the 435 doesn't interpret very well.

If there's no other problem than the 435 showing funny events and results, then just leave it.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[jraef]

Here in the power starved West (USA), we see a lot of "grid switching" transients as the PoCo swaps power hungry sections around the transmission grid to try to keep everyone happy. We have seen upward of 60-80/day, but 6,000 in 24 hours? That's excessive! That comes down to once every 14 seconds or so. I'd be looking for something local to your facility that has a controlled heating process using either a solid state contactor doing phase-angle SCR control or maybe even a mercury displacement contactor, something with a substantial load that is cycling on and off rapidly.


"You measure the size of the accomplishment by the obstacles you had to overcome to reach your goals" -- Booker T. Washington

 

[magoo2]

jraef - I'd be real surprised if our power company caused 60-80 "switching transients" at a location. I don't think we have much more than a dozen operations in a day at a bad location.

More than likely, a significant number of them would originate at the facility itself. That's certainly true with the 6,000 transients in 24 hours, but I think it also applies to the 60-80 range too.

 

[jraef]

I did a PQ study back in 1992 as part of a program designed to help Coscto (big-box buying club here in the US) avoid costly data transmission errors during their nightly satellite up-links. At that time, we recorded MAYBE 2-3 significant grid switch transients per MONTH on average. But the company I work for now bought another small company that makes power loss ride-through modules, and as part of that company, we inherited an installed base of web connected line monitors all across the country. Here in California about 2 years ago, we (and most competitors it turned out) were having a rash of problems with VFD failures, all in a small geographical area (Salinas). I accessed the database and saw the 10-15 transients / hour taking place all over that same area, different user sites, all at the same time, all day long. It might be an anomaly, but out here, we generally suffer from the NIMBY syndrome (Not In My Back yard) big time. Everybody wants power, nobody wants the infrastructure for it to block their views. So when the big Silicon Valley users, who pay for guaranteed power quality contracts, want their power, they get it. Everyone else for 200 miles around them becomes subordinate. The utility will deny it to the bitter end, until confronted with evidence.


"You measure the size of the accomplishment by the obstacles you had to overcome to reach your goals" -- Booker T. Washington

 

[burntcoil]

Skogskurra / jreaf
435 was connected just to see the input voltage because we had swells and dips event recorded by the DCS on the output of the UPS (without any trip, only alarms.

A 480V switchgear is feeding multiple MCCs and one of those MCC in turns feed the UPS. Another MCC on the same switchgear feeds the Central Airconditioning system for the building which is probably around 100 tons. Do you think that can have impact because of its frequent switching because the outside temp. here is nearly 50 deg. Celcius now a days.

 

[Skogsgurra]

It is an easy matter to record switching and mains transients on the same device. Then you can easily see if there is a corelation between the two.

Best is to hook up current clamps on the Air Conditioning system and keep the connection to the 480 V. Then trig on transients and see if they coincide with heavy inrush or starting currents to the AC system.

If there is a corelation, you have to either live with it or find a fix for it.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.