What am I looking for is perhaps an Electrical Engineer

[Van G.]

Because of job cuts we don't have a real Electrical engineer here anymore.

Don't really know if anyone can answer this so thought I'd try here in Mike Holts forums. Excuse me I'm an Industrial Electrician for many years with an elevated position of Elect Tech since I know a bit more about computers and programs and actual programming.

Many job changes from our engineering led me into position of VB programmer on some things since I have been doing that quite awhile and everyone trying to get out cheap and let other people do 10 job's, you know what I mean.

Has been 30 years since I've been to school so I am forgetting some elements of that if you don't use you lose it somewhere in there.

Problem:
We have a test station where we check for Load Losses on the units, IE Copper Loss as some may know. I have a 12 KVA power supply feeding a 5 KVA 7 step-up multiple tap transformer testing all KVA ranges we have from 5 to 75 KVA. When this system was built all we was going to do was 25 KVA and below.

It has no problems with 25 kva and below. When it does the larger units is where we have problems. I cannot think of what is happening here. It does some with no problems but there are case where it failed 74 out of 75 of the larger range.

I don't think that it's capacitive reactance because I don't think that a transformer would have that in my mind but I'm not a transformer designer. Maybe IZ or Impedance but reactance is sticking in my head.

We do have a backup test system with a lot more HP that they easily pass back there but that's manual and takes time.

Any thoughts, suggestions?

 

[sgunsel]

Can you provide more information on the test procedure(s) used?

 

[kingpb]

Agreed, need more info.

It is not clear to me what your testing for, i.e. load loss test is different then short circuit test to determine xfmr impedance.

 

[Van G.]

Can you provide more information on the test procedure(s) used?

Under Load Loss we have a preset calculated current level it will try to push target xfmr to by feeding into Primary side which may be anything you see on distribution lines 14400, 7200. The secondary is shorted by a shunt to make it look like a load. The computer tries to put current level as close to pre calculated value as possible and usually obtains that through a looping in the program. However the voltage doesn't always hit what was pre calculated.

Once current level is obtained on Primary it reads and calculates all the losses such as the shorting shunt loss and deducts from final output which is Total Loss Watts. Usually this is what fails by Total exceeding what was expected to be design value.

I believe that some designs of these larger units may be causing a reflective action of some sort or it's not saturating completely. I have tried extending time but that doesn't help. The 5 kva step-up cannot overcome the larger target and I am at a Total Loss myself to understand exactly what is happening.

Have to get product out the door so we fail them and send down to another station to check them which as I mentioned has a lot more HP. It has the capacity to test them directly without a step up xfmr. Sorta like a Volkswagon power supply compared to first station dual suitcase size.

 

[kingpb]

So, it sounds like your doing a short circuit test to determine/verify impedance.

During a short circuit test, the testing voltage is kept very low, thereby making the exciting current negligible. However, if your set-up is causing the voltage input to be too high, then you could also be measuring a portion or all of the core losses.

Another thing to remember is that if this a three phase transformer, you may be measuring total 3-phase, power line volts and line amps. The impedances must be calculated on a phase basis, so conversion would be required.

Hope that makes sense.