3Ph 5KVA Transformer Troubleshooting with Anomolies

[jkcowboy]

Hi All,
I'm trying to figure out what happened with some transformers and hope someone has seen something similar before:

Problem: All three Xfmr (5KVA each) Bank will blow a fuse when hooked up to 480VAC three phase. The customer reported that when power was applied (input fused switch closed) that two of the three line fuses would blow immediately. These are Acme 240/120 general purpose xfmrs. Primary: Ph1 to H1,H5 and Ph2 to H2,H6. Secondary: X1(Hot), X4(Neutral), X2/X3 connected. I haven't gotten far enough to see if these connections actually produce the desired Voltages or not. In summary, two of 6 primary windings are energized with a full secondary.

Background: Transformer bank was built in Alaska years ago to step down 480VAC to two outputs: 380VAC Delta and 220VAC WYE to run some IEC Euro Equip. Rumor (no personal observation) has it that it ran good once. Not anymore. I think their thinking on how to transfer euro equip to run in the US is faulted, and i've given them some better suggestions, but I was paid to make the transformers produced the Euro voltages at 60Hz. That part sounded easy to me on something that has worked before.

Ok, so what did I do:
I lifted leads on the primaries and secondaries and supplied 480VAC to input terminals, no xfmrs on line, to check the input pwr cord and input pwr fused switch, no problems noted. Meggared all three primaries and got a little less than an Ohm on each. Meggared the secondaries and got a little less than an ohm on each as well. Pri and Sec to ground was greater than 40 MOhm. Pri to the sec was greater than 40 MOhm. So far so good. I kept the secondary leads lifted and taped and one by one applied power to the primaries. No go. Each time the input Fuse (one of the two powering the transformer) (FRS-r-10A) blew immediately. All three transformers had the same readings, indications and all blew the fuse immediately. I just can't believe that 3 xfmrs would all go bad at the same time.

What is confusing me is that I am sure that I have short somewhere in the equipment but cannot find it with test equipment. Also, the probability that a transformer could go bad is always there but three transformers with exactly the same indications? The odds on that one are small.

I'm going back tomorrow morning. My plan is to open the leads at each tranformer, pwr the primaries directly with a new wire straight from the generator -> fused switch -> new wire directly to the xfmr and measure volt across the secondaries (x1 to x4). If it blows after I've removed all equipment and existing wiring... it has to be bad. I would just hate to tell the customer he has 3 bad xfmrs when I cannot show them as bad with test equipment or explain how three went bad at the same time.

I'm positive I'm missing something and just don't know what. Help or suggestions would be appreciated?

John King

 

[Smart $]

Hmmm... 10A primary fuses on essentially a 15kVA 480V 3? xfmr.

15,000VA ? 480 ? √3 = >>>18A<<<

Inrush could be killing 'em :slaphead:

 

[jkcowboy]

Sorry, not to mention it but the customer was using 15A when the problem developed. I started troubleshooting with 10A fuses figuring that a time delay 10A would be enough while I trouble shot the fault. I did not want to damage anything until I understood the problem better. Tomorrow, I'll use 15A (50% increase in amperage) fuses to make sure I'm not introducing another unneeded variable (inrush effects) into the troubleshooting. Thanks, I'd appreciate any other advise others might offer.

Hmmm... 10A primary fuses on essentially a 15kVA 480V 3? xfmr.

15,000VA ? 480 ? √3 = >>>18A<<<

Inrush could be killing 'em :slaphead:

 

[kwired]

I'm having trouble deciphering just how you have this connected. Make sure you have it connected properly.

If I understand correctly you have three single phase transformers and are combining them to make one three phase transformer bank.

If that is correct you can connect each single phase unit separately and check for proper output, if that is alright you probably are connecting something wrong when attempting to make a three phase bank.

 

[Volta]

You state that the transformers are 240/120, yet you (they) were applying 480 volts. Is that true?

Depending on the way the coils are wound, the inrush can be great, but I'd think that your step-down application would be more likely to hold than reversed, if they are general-purpose. But you could try to apply a low voltage to that winding and show that they do hold and produce the proper high voltage.

 

[kwired]

Is this something that was working and now does not? I first had impression that it was moved and you are setting it up in a new location.

 

[Finite10]

"All three Xfmr (5KVA each) Bank will blow a fuse when hooked up to 480VAC three phase. The customer reported that when power was applied (input fused switch closed) that two of the three line fuses would blow immediately. These are Acme 240/120 general purpose xfmrs."

I stopped reading here. Need to know xfmr pri and sec V, then applied V from source and required utilization V.

 

[jkcowboy]

I tried to include the details accurately. A few corrections. The transformers are single phase 480/240 transformers. I wasn't sure how the electricians made the drop from 480 to 380 and called TEMCO who provided a drawing but of a different transformer series. The builders wired the transformers using 2 of the 6 primary windings and the full secondary. I don't know what the output will be but for now I just want to find out what is specifically wrong that keep blowing the fuses. Transformer Pri (480/240), Secondary (240/120), Source (440-480), Utilization 400-500 Volt input Power to DC Power supplies for a 30 VAC/90Hz synchro driven motor. These pwr supplies are Euro and originally built to run off of 380/50. They have been running them at 380/60 with the use of this transformer bank. There might be a couple of 220 volt single phase and 3ph machines as well. They are getting the info on that from a different division in Holland (i.e. the machines are not local).

As for connecting the XFMR one at a time. Done. All blew fuses.

Do you guys see any fault with my checks so far or my plan for tomorrow? Anyone ever have a bad XFMR that could not be detected by meggar?

John

 

[jkcowboy]

I saw one more post I needed to reply to: this machine was moved a couple of years ago from AK to Texas. The last time it was used, it blew 2 of 3 line fuses and could not be fixed at an industrial location. They rented a Generator, dialed the Voltage down to 400VAC/60Hz and ran the machinery without the XFMR bank. They are now thinking of doing some more work and remembered that the XFMR were not running so well last time. Presto, enter me into the situation. My job is not to re-engineer the arrangement but to make it work again, starting with identify what exactly is wrong. I haven't made it to a loaded XFMR yet, just tring to get them to turn on.

John

 

[Volta]

Can you post a link to the TEMCO drawing for the winding connections?

 

[jkcowboy]

Sorry,
I have a pdf of the drawing and could not figure out how to share it. Here is some text from the drawing:

Input: 480 Volts Delta, 3 Phase, 3 or 4 wire, 60 Hz
Output: 394 Volts Wye 228, 3 Phase, 3 or 4 wire, 60 Hz
Refer to individual transformer labeling for voltage tap adjustments.
Input and Output may be reversed.
KVA remains the total KVA of the three transformers used.
This diagram covers ACME single phase transformers used: T-2-53515-3S, T-2-53516-3S,
T-2-53517-3S,T-2-53518-3S

As a note the drawing has 10 H connections, 4 X connections and the nameplate # of the drawing and the H connections do not match the actual existing configuration or wire-up.

John

 

[Smart $]

...Do you guys see any fault with my checks so far or my plan for tomorrow? Anyone ever have a bad XFMR that could not be detected by meggar?

To measure resistance of a shorted winding requires a precision ohmmeter. Just showing 0 ohms on a megger is no indication of a good or bad (i.e. shorted) winding. Only thing it rules out is an open winding... which would not blow a fuse :roll:

 

[Smart $]

Sorry,
I have a pdf of the drawing and could not figure out how to share it.

Depending on its size, you can add it to a post as an attachment. Yet the attachment feature of this forum leaves a bit to be desired. Just the upload process can be frustrating. Haven't tried lately, but in the past I've tried to attach pdf's that were way under the size cap and they would not upload. Can't attach if they don't upload :blink:

If you have a google account, upload to Google Documents, set sharing to anyone that has the link, and include the link in a post here. Can also be done with a Windows Live account on SkyDrive, but the link will likely be changed at some point in time without your knowledge.

There are many other avenues for sharing...

I am going to PM you my email address. If you don't want to mess around with uploading the pdf, just email it to me and I'll upload and link to it here in the thread.

 

[straynote6]

It's late at night and I didn't read the entire post - so, I'm sorry if someone already brought this up.

With Delta connected transformers, you must make sure that the polarities of the coils are correctly phased. With a 3 phase delta transformer, the plus from one coil must be connected to the negative of the next... i.e., If the positive of phase "A" is connected to the positive of phase "B", you will certainly create a short circuit. The positive of phase "A" must be connected to the negative of phase "B". Likewise, the positive of phase "B" must be connected to the negative of phase "C" and so on. Hope this helps.

 

[Open Neutral]

I troubleshoot fuse-blowers with a good old lamp incandescent in series. But that's at 120V not 480 and far smaller power levels.


Can you use that approach somehow? Maybe even power them up at 120V to measure the secondaries?

 

[Smart $]

It's late at night and I didn't read the entire post - so, I'm sorry if someone already brought this up.

With Delta connected transformers, you must make sure that the polarities of the coils are correctly phased. With a 3 phase delta transformer, the plus from one coil must be connected to the negative of the next... i.e., If the positive of phase "A" is connected to the positive of phase "B", you will certainly create a short circuit. The positive of phase "A" must be connected to the negative of phase "B". Likewise, the positive of phase "B" must be connected to the negative of phase "C" and so on. Hope this helps.

Not [100%] true, and definitely not true [0%] when connecting three 1? transformers with separate cores and enclosures as a delta configured bank.

 

[jkcowboy]

All, heading out to the customer in about an hour. This "arrangement" has quite a bunch to talk about but I'm just trying to get the Xfmr to run now. The primary and secondary leads are lifted... so there is no configuration to speak of now.

Based on the discussion, here is what I took from the conversation:

(1) My meggar is not good enough to measure the resistance of the windings, even more so if the windings are in parallel (low to even lower resistance). Therefore I could have a short in across a winding and not see it with my testing. I just can't believe that it could happen to all three xfmrs. That is what makes me think that I'm missing something.

(2) I'm going to install 2 wires. The first from the generator bypassing the input pwr cord and going to the input of a fused switch. The second from the fused switch directly to the transformer bypassing internal wiring, terminal boards and conduit runs. I'm going to properly ground the transformer and pwr the primary windings with two phases of a 3 phase source. The secondary leads (X1/X4) will be open and exposed at the xfmr. In my opinion, this is a run - no run situation from which I could dub the xfmr as good or bad. In the switch, I'm going to use FRS 15A fuses. There should be no inrush issues with a single xfmr and proper fuse.

(3) I do not have a good understanding of how the xfmr is wired since it is not according to the wiring diagram or as an autoxfmr, they did something different. I'm not going to address the wiring for now at all for two reasons: it was reported to work before and if the xfmr are identified as bad and they hire me to fix the problem... redesign will be in order. There are other xfmrs avaialble. If possible I would like to get away from the 380V all together as I don't believe it is necessary to run their equipment. They might need a 3 ph 220 or some 1 ph 220 but that is much easier to get.

The thing that bothers me is that when you have 3 isolated components that appear to have gone bad but cannot prove it, you are setting yourself up to later on find out they were fine all along and you missed something in common that should have been caught. This is the mistake that I do no like or intend to make.

Thanks for the comments thus far and I solicit any other to help me troubleshoot the xfmr that will not run.

John King

 

[kwired]

Your megger is good for faults to ground or primary to secondary faults.

A simple ohm meter is good for continuity through windings.

For shorts from one winding turn to another you need a meter that is accurate at low resistance ranges, and you also need to know what kind of reading to expect for the coil you are testing. Length and size of conductor used to make the coil plus temperature will all effect the final result.

If all other tests are good but you have overcurrent with proper voltage applied and no load on the secondary the coil likely has shorted turns someplace.

 

[jghrist]

It's a little difficult to figure out the winding connections without a diagram. You say there are three 1? transformers, but you mention H1 thru H6. You say the secondary is 120/240, but then you say the output is 394/228 wye. You say there is a diagram but the transformers were not connected according to the diagram. Did you make a diagram of how the transformers were connected before you disconnected all the wires? This is all very confusing.

 

[gar]

110817-0841 EDT

I believe you have the primaries connected in parallel, thus for 240 input. You want the two primaries in series for 480. Your secondaries appear to be connected in series as would be required.

See the right hand diagram at http://www.acmepowerdist.com/pdf/Page_147-156.pdf
You should supply 480 across H1 to H8 with H4 and H5 connected together. This is listed as 504 input for an output of 240. Thus, 240 * 480 / 502 = 228. Somewhat closer to your desired 220.

These transformers won't convert 480 delta to 380 delta. But 480 delta primary to 228 Y will provide 395 delta. I believe these voltages will be the full load values, and therefore unloaded would be higher.

With your primaries connected in parallel you will certainly blow fuses.

.