Transformer Replacement 1

[Mstrvb19]

Our power system has four nuclear class 1E safety-related 4160V/480V substation transformers that supply our class 1E 480V system. The regulation of these transformers as we move from low house load to high house load is sub par and requires us to change taps. However, the transformers were retrofilled with silicon several years back (they are about 30 years old) and with the new fill, there is insufficient lubrication to change taps. Unfortunately, they attempted to change taps before anyone recognized this issue, and the taps were broken. We have managed to get ourselves into a position where the taps are set at 2.5% boost but CANNOT be moved, and are looking to replace the transformers. We have one current bid for the replacement that would give us exact duplicates of the ones we have, which were supplied integrally as part of a GE AKD Powermaster unit substation. My thought is, if we are going to replace these things, why not replace them with tansformers that regulate better and allow us to avoid manually changing taps every time we start up or shut down the plant. However, I have had two problems in my search for a replacement and I am looking for any thoughts any of you might have. These transformers are 1MVA, and it appears that, historically, constant voltage transformers/ferroresonant isolimiters are not made this large or at these voltage levels. We have been able to find several munfacturers who make LTC's of this size, but we do not have very much space in our switchgear rooms and I am concerned that the extra space for the tap changer is not available. I have been pursuing getting someone to design a regulating (CVT type or other continuous constant output transformer) and having one manufactured, but I can't seem to get a simple answer to my questions. All I need to know is whether or not a transformer of this nature can be designed and built for a reasonable cost, or if I am looking for a pie-in-the-sky solution that doesn't exist. I have done extensive searching and I have yet to find any manufacturers in the states (don't want to ship these babies a long way) that make a transformer like this, so I guess my question is, is it reasonable to have it designed and built (and tested for class 1E service) or is this a waste of time?

 

[electricpete]

LTC's are a maintenance intensive piece of equipment, even in the non-nuc world.

CVT at that voltage and power level I haven't heard of. Usually used on smaller systems.

My suspicion is that any of these would be a first-of-a-kind application within a nuc plant and would be a major pain to analyse.

I think the first/easiest thing to investigate is whether you can obtain a lower-impedance transformer which will provide better regulation enough to avoid having to switch taps on a routine basis.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

... and of course evaluate the effect on fault current if impedance is changed.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[Mstrvb19]

Yeah, unfortunately our fault analysis shows that we have almost no room for increase in fault current-our 480V bus postulated fault levels are at 21kA with 22kA rated breakers! Overall, the power system here is a poor design, and is very old. I have thought of performing an analysis to determine a lower impedance transformer to get better regulation, but we would have to completely re-organize the power system connections (i.e. what busses have the large motors, what buses are fed from other buses) and even that, which is insane to even think about, might not solve the problem.

 

[rbulsara]

Consider and explore transformer with solid state (SCR based) tap changers. No moving parts.

 

[Mstrvb19]

Any suggested manufacturers that might cut my searching time down rbulsara? Sounds like something I'd like to pursue.

 

[electricpete]

What is it that feeds the 4kv/480 volt transformer?

Is that a safety related transformer? Or perhaps 2 non-class transformers with some auto-transfer?

I would think it might be easier to install LTC on a non-class upstream transformer. Even though it would be higher voltage and/or higher current, at least it might be non-class.

And getting to the issue of voltage control at the transmission level, there certainly may be a wide range of options.




=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[rbulsara]

I do not know off hand..but start out with

http://www.chromalox.com/products/productinfo/en/pdf/7920.pdf

http://www.ustpower.com/pdf_files/evr technology comparison 101102.pdf

They may know someone who makes..lareger units


Also check with leading transformer manufacturers..

1000kVA may be little high for such applicaiton..but I am sure someone will make one for you

 

[Mstrvb19]

The upstream supply for each transformer is a class 1E 4160 bus supplied by a non-class 1e 13.8kv/4.16kV transformer. Yes, we could replace these, but keep in mind that it is the failure of the taps on the 4160 to 480 xfmrs that is driving the change out, and I am trying relieve the supply voltage issue at the same time. The voltage range on the 4.16kV buses is fine for the directly connected loads.
Incidentally I looked into the electronic voltage stabilizers which are controlled by a power conditioner, but I have not found any at this high of voltage.

 

[rbulsara]

Also check out S&C co. They make Meduim voltage UPSs .

http://www.sandc.com/

They may be able to help or offer advice.

Not a cheap thing though...

 

[Mstrvb19]

just out of curiousity---how does everyone keep posting these links? Specifically, I tried to post a zipped document this morning and it didn't work.

Thanks for the link, I'll check it out.

 

[stevenal]

The regulation of these transformers as we move from low house load to high house load is sub par and requires us to change taps..

I'm puzzled. The process you described is called tap changing under load, and can only be performed safely with an LTC. And yet you don't have room for an LTC?

 

[Mstrvb19]

stevenal, these four transformers supply two busses- two xfmrs per bus, so when they change taps they live transfer (not fast, though, manually) to the alternate source.

 

[rbulsara]

Links are pasted by simple cut and paste or you can type them in.

The attachments are a different story. I beleive best way to have your web page. Upload the document to your web page and provide a link to your web page and where link to your document resides..

There may be other method...

 

[BJC]

What kind of voltage swings were you getting?
Has the load changed since the plant was built?
Electricpete is righ qualifying anything new or unique for a nuck will require expensive paper equal to about three times the weight of the transfromers.
Hav eyou looked at a 1.5 for 1.25 MVa transformer. They shouldn't be to much bigger than the 1.0 MVA unit.

What kind fo equipment is connected to the f 480 volt supply.
If you have old 22 kaic breakers. I wouldn't hesitate tho replace them. The buswork may be rated for a higher fault current. You may have to dig out some old catalogs.

 

[Mstrvb19]

quite right, BJC, we are, in fact, in the process of replacing the breakers (they are old GE AK25's). The buswork is rated for something higher.
However, they are not replaced yet, and these nukes change so slowly that I can't afford to install a xfmr that would require them to "speed up" their replacement schedule. The problem is this: we have a large number of loads that must start in an accident. With no tap changes, the voltage at the buses is around 90% prior to motor start. When we get a block start of these motors, the voltage dips so low that we lose contactors and, once the new loads (those with latching contactors)have reached full speed, the new house load brings the voltage so low that we would trip onto EDG's. This is in violation of NRC GDC 17 (self induced trip, taxing of safety equipment unnecessarily). At any rate, the problem isn't the direct 480V loads necessarily, but rather, the 240 and 120 volt loads further downstream. We are also considering replacing those transformers. I just figured that, since we were replacing these we should try to improve our regulation at the same time (it takes so long to get a mod approved that you ought to make the best use of it once it is)
The 480V supply has mostly large industrial stuff-motors, heaters, lighting transformers. However, we do have some logic control modules (indirectly at lower voltage)that have fairly tight tolerances. Loss of these modules would be BAD.
Your point is valid, though. Perhaps I can get them to accelerate just these two 480V 1E buses with the breaker replacements so that I can increase the transformer MVA. I'll have to look into that.

 

[rbulsara]

How about using reduced voltage starters..where possible..if not already...

Also automtically staggering the motor starts..using PC based building automation systems..