3PH transformer blows breaker on power up

[ptonsparky]

NM. Found a paper on it. (Insert crossed eyes emoji)

 

[winnie]

How long does it take for the residual flux to disappear?

That will be heavily dependent on the undesirable characteristics of the core materials.

An ideal transformer has zero residual flux.

I don't know how much it would be for a real transformer.

Jon

 

[winnie]

NM. Found a paper on it. (Insert crossed eyes emoji)

Please share

Jon

 

[FOG1]

Transformer inrush current is affected by the timing of applying power and by residual flux in the core. The residual flux is set by the previous operation of the transformer.

My out on a limb guess is that the previous operation of the machine is somehow setting up the transformer for a bad startup the next time around.

The only way I could see this happening is if the machine were somehow causing a DC imbalance, and even this is a pretty weak theory.

But you do have a history of installing multiple transformers, having them appear to work, and then having them unable to start because of excessive inrush.

Jon

The machine on the secondary side is not being turned on.

All of this is occuring when turning on the transformer only.

This is the only transformer situation in our shop that has gave us trouble.

After installing #3 51kva last Friday and having at least 12 on off cycles without tripping breaker the machine tool was never turned on.

Saturday morning turned on disconnect and tripped breaker instantly, again machine tool was not turned on.

So from install to removal the transformer never saw a load on secondary side.

 

[FOG1]

220328-2007 EDT

FOG1:

I can not access your link. But based on what winnie posted I suspect that you do not have good information on peak current in the first half cycle.

You can get a good reading on peak current for this application with a suitable current probe, and a Fluke 87 in peak mode. The 87 in peak mode is faster than 1 millisec, a 60 Hz half cycle is 8,3 milliseconds.

When using a shunt for fast measurements it is important to minimize a 1 turn coil in the connection of wires to the scope from the shunt. One time I made a special shunt to minimize this problem. This was related to measuring thousands of amperes to a 15 A circuit breaker.

In this motor experiment we need to see the first half cycle current, and you will not necessarily get it with a single random measurement. The largest peak value will occur for an unloaded transformer when the applied sine wave starts on a voltage zero crossing, with the correct slope for the residual flux state, and the transformer left in its highest residual flux level from its last turn off.

You and most electricians on this forum lack the capability to make the measurement. Further most electricians here have no more knowledge of this problem, or how to measure it than you do.

.

The link still worked when I clicked on it Gar???

 

[gar]

220329-1422 EDT

All ferromagnetic materials have a characteristic curve called the hysteresis loop. This is a curve of magnetic core flux level vs magnetizing force.

If you apply an AC magnetizing force of some level this curve will approach a steady state hysteresis loop. When you remove excitation and magnetizing current drops to zero the residual flux density in the core does not drop to zero. This generally does not drop away from this residual level. However, some materials may drop off more easily than others from various external causes. Alnico is very resistant to demagnetizing.

You can demagnetize a magnetized material by applying an AC field of which you gradual reduce the intensity.

See chapter 13 of 'Electric and Magnetic Fields" by Stephen S. Attwood, 3 rd Edition, 1949, John Wiley and Sons, Inc.

FOG1 the problem is my computer.

.

 

[winnie]

FOG1: if the machine was never connected to the system then I guess my idea is a non-starter. But I did say it was an 'out on the limb guess'.

Gar: If you can do a zoom call I can show you the files.

-Jon

 

[FOG1]

FOG1: if the machine was never connected to the system then I guess my idea is a non-starter. But I did say it was an 'out on the limb guess'.

Gar: If you can do a zoom call I can show you the files.

-Jon

Well the machine was connected but it’s disconnected was never turn on so no secondary load had been applied.

 

[gar]

I posted a reply and it was destroyed and non recoverable. It was a description of how to make a shunt and use a fluke 87 to measure peak inrush.

.

 

[winnie]

FOG1:

1) On the service with the 75 kVA unit (the one that doesn't have tripping issues), are there any other devices connected?

-Jon

 

[FOG1]

FOG1:

1) On the service with the 75 kVA unit (the one that doesn't have tripping issues), are there any other devices connected?

-Jon

No other devices are connected to that 75kva trans , Each transformer runs one machine only. Each machine has it's own built in safety disconnect switch to turn it on when in use. There is another machine on the primary side of the service as its a 240 volt machine.

 

[winnie]

No other devices are connected to that 75kva trans , Each transformer runs one machine only. Each machine has it's own built in safety disconnect switch to turn it on when in use. There is another machine on the primary side of the service as its a 240 volt machine.

Meaning that the 400A service has a 200A breaker for the 75 kVA transformer, a 200A breaker for the 51 kVA transformer, and another breaker for the third machine?

Is the any correlation between the operation of this third machine and the operation/failure to operate of the 51 kVA transformer?

-Jon

 

[FOG1]

Meaning that the 400A service has a 200A breaker for the 75 kVA transformer, a 200A breaker for the 51 kVA transformer, and another breaker for the third machine?

Is the any correlation between the operation of this third machine and the operation/failure to operate of the 51 kVA transformer?

-Jon

The 200 amp breaker feeds into a service disconnect inside building. That disconnect splits and feeds 75kva trans and to a fused disconnect feeding the other machines 240v 3ph.

I dont see how it would have any effect on it because it's not been turned on at anytime that we have been energizing the 51 kva trans.

 

[winnie]

If the other machine was never on any time while attempting to energize the 51 kVA transformer, then it has no effect. But if sometimes it was on and sometimes off, that _could_ have an effect, which is why I was asking.

This thread is an example of why we are not permitted to help 'DIY' work on this forum. Someone who isn't an electrician might not even know to ask about something that is _critical_ to the installation. I believe the moderators permitted this thread because the actual installation work was being done by electricians, and you (as the owner) are asking for input to better direct what the electricans are doing. I'm happy to continue, but just want to note that potentially relevant details keep getting added to the picture.

Where things stand right now: you have a transformer that is probably functioning normally, with an installation that is probably electrically correct, but its inrush current is sufficient to trip upstream OCPD. I say _probably_ because 3 separate transformers have had similar problems in the exact same installation, but it is still possible that there is something about the install which is causing the problems. Additionally a larger transformer in a similar install (same building, same power source, same size supply breaker) is having no problems starting. The other transformer's working is suggestive but not a smoking gun.

If the problem is some (as yet unidentified) error in the installation, you of course want to fix it.

If the problem is the normal operating inrush current, then you want to come up with a way to reduce this inrush current.

And of course you don't want to simply keep trying things and banging your head on them.

Right now my thoughts are: 1) Could you try the existing 75 kVA transformer in place of the 51 kVA unit. If this works, then you've essentially determined that some characteristic of the 51 kVA transformer is the problem. 2) Can you have your electrician install some sort of inrush current limiting components. 3) Can you change the 200A OCPD feeding the transformer to something with a slower trip characteristic, or even increase the size of this OCPD. (Note: there are lots of details here which your electrician will need to explore. Eg. the transformer might be kosher with a 300A primary OCPD, but your 200A panel won't be!)

-Jon

 

[gar]

220330-1200 EDT

winnie:

I do not want to load it.


An unloaded transformer when turned off is probably left in a more saturated state than when turned off with a maximum resistive load on the secondary. I have not run experiments on this or even given any close theoretical look.

My guess is that 100 ft of #10 copper wire will have resistance of about 0.1 ohm. If such a roll can be purchased, and a roll was left rolled up it also has inductance, and then use three rolls, one in each hot leg, then see if this produces enough impedance to get power to the transformer.

.

 

[FOG1]

If the other machine was never on any time while attempting to energize the 51 kVA transformer, then it has no effect. But if sometimes it was on and sometimes off, that _could_ have an effect, which is why I was asking.

This thread is an example of why we are not permitted to help 'DIY' work on this forum. Someone who isn't an electrician might not even know to ask about something that is _critical_ to the installation. I believe the moderators permitted this thread because the actual installation work was being done by electricians, and you (as the owner) are asking for input to better direct what the electricans are doing. I'm happy to continue, but just want to note that potentially relevant details keep getting added to the picture.

Where things stand right now: you have a transformer that is probably functioning normally, with an installation that is probably electrically correct, but its inrush current is sufficient to trip upstream OCPD. I say _probably_ because 3 separate transformers have had similar problems in the exact same installation, but it is still possible that there is something about the install which is causing the problems. Additionally a larger transformer in a similar install (same building, same power source, same size supply breaker) is having no problems starting. The other transformer's working is suggestive but not a smoking gun.

If the problem is some (as yet unidentified) error in the installation, you of course want to fix it.

If the problem is the normal operating inrush current, then you want to come up with a way to reduce this inrush current.

And of course you don't want to simply keep trying things and banging your head on them.

Right now my thoughts are: 1) Could you try the existing 75 kVA transformer in place of the 51 kVA unit. If this works, then you've essentially determined that some characteristic of the 51 kVA transformer is the problem. 2) Can you have your electrician install some sort of inrush current limiting components. 3) Can you change the 200A OCPD feeding the transformer to something with a slower trip characteristic, or even increase the size of this OCPD. (Note: there are lots of details here which your electrician will need to explore. Eg. the transformer might be kosher with a 300A primary OCPD, but your 200A panel won't be!)

-Jon

Thanks Jon,
The swapping out 75 kva to 51 kva is a no go 75kva is mounted up 10ft off the ground.

 

[gadfly56]

If the other machine was never on any time while attempting to energize the 51 kVA transformer, then it has no effect. But if sometimes it was on and sometimes off, that _could_ have an effect, which is why I was asking.

This thread is an example of why we are not permitted to help 'DIY' work on this forum. Someone who isn't an electrician might not even know to ask about something that is _critical_ to the installation. I believe the moderators permitted this thread because the actual installation work was being done by electricians, and you (as the owner) are asking for input to better direct what the electricans are doing. I'm happy to continue, but just want to note that potentially relevant details keep getting added to the picture.

Where things stand right now: you have a transformer that is probably functioning normally, with an installation that is probably electrically correct, but its inrush current is sufficient to trip upstream OCPD. I say _probably_ because 3 separate transformers have had similar problems in the exact same installation, but it is still possible that there is something about the install which is causing the problems. Additionally a larger transformer in a similar install (same building, same power source, same size supply breaker) is having no problems starting. The other transformer's working is suggestive but not a smoking gun.

If the problem is some (as yet unidentified) error in the installation, you of course want to fix it.

If the problem is the normal operating inrush current, then you want to come up with a way to reduce this inrush current.

And of course you don't want to simply keep trying things and banging your head on them.

Right now my thoughts are: 1) Could you try the existing 75 kVA transformer in place of the 51 kVA unit. If this works, then you've essentially determined that some characteristic of the 51 kVA transformer is the problem. 2) Can you have your electrician install some sort of inrush current limiting components. 3) Can you change the 200A OCPD feeding the transformer to something with a slower trip characteristic, or even increase the size of this OCPD. (Note: there are lots of details here which your electrician will need to explore. Eg. the transformer might be kosher with a 300A primary OCPD, but your 200A panel won't be!)

-Jon

Winnie, keep in mind that the breaker trip is happening when the transformer is energized, not when the load is energized. The OP is keeping the load disconnected when the transformer is energized. If it's inrush current that's the problem, where the heck is it going other than magnetizing the core?

 

[winnie]

Winnie, keep in mind that the breaker trip is happening when the transformer is energized, not when the load is energized. The OP is keeping the load disconnected when the transformer is energized. If it's inrush current that's the problem, where the heck is it going other than magnetizing the core?

I absolutely agree.

I'm trying to come up with things that could potentially affect the core saturation.

If there was a strong DC magnetic field in the area, it could predispose the core to satuation. That is why I asked about magnets. But that wasn't the issue (I was imagining FOG1 saying 'oh, like the MRI shop next door??' )

If the other 'directly connected' machine on the same service involved a large motor, then its 'short circuit contribution' might increase the inrush to the transformer being energized. But FOG1 said that the other machines have always been off.

If the machine connected to the transformer in question has something like a VFD with input rectifier, and that rectifier has a failed doide causing DC current flow, that might affect the core saturation during one operating session, affecting the next startup. But FOG1 said that the machine was never operated.

Right now, best guess, is that this transformer, operating normally, has too high inrush current for the supply OCPD. IMHO the next most plausible issue is some huge wiring error that looks correct. (HMM, this is a 'step up' transformer. The _lower voltage_ terminals are labeled H1, H2, H3 because they are the primary input terminals. I could imagine someone connecting the incoming supply to X1...X3 and everyone reviewing it thinking it was correct because that makex X1..X3 low voltage and H1..H3 high voltage...but this doesn't fit with the power quality traces that showed inrush current and then current dropping to reasonable values.)

-Jon

 

[jim dungar]

Winnie, keep in mind that the breaker trip is happening when the transformer is energized, not when the load is energized. The OP is keeping the load disconnected when the transformer is energized. If it's inrush current that's the problem, where the heck is it going other than magnetizing the core?

My anecdotal experience is loads are more likely to reduce inrush current rather than magnify it. I believe the impedance on the secondary somewhat impacts the transformer equivalent circuit.

 

[ptonsparky]

My anecdotal experience is loads are more likely to reduce inrush current rather than magnify it. I believe the impedance on the secondary somewhat impacts the transformer equivalent circuit.

Nice to know because years ago I tried to capture the 'inrush' of a 5kva single phase transformer with and without a secondary load. IIRC it was lower but took longer drop Fluke 43B.