5kVA 480V to 120V single phase transformer blowing fuses, and shouldn't be.

[sparkie1]

What I'm ultimately asking myself here, is if I should replace this transformer.

It has been in service for a long time. The transformer has primary and secondary protection, though only primary protection is required. It is a 5kVA 480-120 single phase transformer that is feeding a set of 480W DC power supplies. We had a power outage yesterday, and when this cabinet had power restored, the primary fuses opened up. They are Cooper Bussman type CF 15A fuses. The primary is also protected via a 15A circuit breaker. Given that the circuit breaker held, I went ahead and put in a couple convenient 30A fuses and the system held. So I shut it back off, went and got a couple 15's, disconnected the secondary conductors, powered on the transformer, and the blew right away. So now I know that the 15A fuses blew with and without a load on the secondary.

I checked the calculation, and 250% is allowed with primary and secondary protection, but only primary is required here, so for primary-only, 10A would be the fuse, and 15A with primary and secondary. The secondary protection is the same style of fuse, but at 30A. It's sufficient for our load.

Ultimately, I went and grabbed a couple 20's to get it going, tried a few power cycles unloaded, and they held fine. I turned the load on, and everything was good. I'm a bit perplexed but the mechanism here, as this system has had numerous power cycles with these 15A fuses and has been fine. I do have that breaker protecting the transformer primary, so I'm okay with the 20A fuses in the panel, but I'm curious what would cause this transformer to suddenly be opening 15A fuses that have been in there for years.

 

[tom baker]

Did you Meg the transformer?
How critical is the load compared to the cost of a new transformer?

 

[kwired]

Changes to the source could allow more starting current to be available.

Other load conditions could allow more starting current to be available.

The original fuse maybe was near blowing from the beginning and one of the above mentioned items now allows just enough extra current to flow to take that fuse out?

If you had say a turn to turn short in windings, it would lessen impedance but also would impact output voltage.

Is incoming voltage same as it always been with similar loading? If not POCO could have tap changer stuck on a high setting and that could also contribute to blowing this fuse if it was on the border of blowing to begin with.

 

[ptonsparky]

A 15 has always been marginal, IMO.

 

[kwired]

A 15 has always been marginal, IMO.

I mostly agree on that.

Would be allowed to protect the secondary if the secondary is two wire system, but may or may not hold when energizing.

 

[winnie]

IMHO the mechanism is the randomness of transformer inrush current.

Depending upon the timing of energizing the transformer and the residual core flux, the initial current flowing into an unloaded transformer can be anywhere from steady state magnetizing current to several multiples of full load current.

Using a 15A fuse on a transformer with 10.5A FLC? Could reasonably work a bunch of times, then fail twice, then work again for a bunch of times.

Jon

 

[sparkie1]

Thanks everyone. As an update, as I was walking the plant and going through the panels, I found another set of blown fuses, and a couple cabinets with bad fuses in the bottom. Seems like this is a common issue.

The thing is that they don't even seem to be needed. I think I'm just going to replace the fuse holders with terminal blocks and be done with it. They literally have a breaker in the bucket right next to them.

 

[winnie]

All over the plant you have step down transformers protected by both breakers and fuses?

IMHO worth trying to figure out _why_ before changing anything.

Jon

 

[ptonsparky]

A little research.
A typical 5 kva single phase 480v transformer inrush is about 260 amps.
Your 15 amp CF has a time current curve of about .1 second at that level. You may get better resolution if you blow up the chart.
You will need to do a bit of research as how inrush is affected. It is above my pay grade to explain.

 

[kwired]

Thanks everyone. As an update, as I was walking the plant and going through the panels, I found another set of blown fuses, and a couple cabinets with bad fuses in the bottom. Seems like this is a common issue.

The thing is that they don't even seem to be needed. I think I'm just going to replace the fuse holders with terminal blocks and be done with it. They literally have a breaker in the bucket right next to them.

Must be trip curve of the fuse? If you had a RK5 fuse in there I would think in most cases the breaker would trip before the fuse would. This based on selection for motor circuits, where you would generally go with 175% of full load current for the fuse and 250% for an inverse time breaker for maximum settings, but I think most any inductive load would have similar initial "inrush".

 

[topgone]

Look at Table 450.3(B). Transformers 1000V or less and 9 amperes or more says 250% fusing on the primary.

 

[petersonra]

If you want to keep a primary fuse maybe try a fuse that is designed for transformer inrush like an FNQ.

 

[TwoBlocked]

...

If you had say a turn to turn short in windings, it would lessen impedance but also would impact output voltage.
...

In my experience, an internal fault with any sort of AC coil (transformer, motor, solenoid, etc.) will quickly result in the magic smoke coming out. I look at it as an autotransformer with a shorted secondary.

To me, this is in the category of nuisance tripping and calls for upping the fuse after taking an amp reading.

 

[ptonsparky]

In my experience, an internal fault with any sort of AC coil (transformer, motor, solenoid, etc.) will quickly result in the magic smoke coming out. I look at it as an autotransformer with a shorted secondary.

To me, this is in the category of nuisance tripping and calls for upping the fuse after taking an amp reading.

Guessing that most of us do not have the meters capable of capturing that inrush.

Fusing too small and/or the wrong type stands out.

 

[gar]

230402-0824 EDT

A transformer core has a residual flux level after turn off that results from the last turn off timing and load.

With relatively good core materials this can be a fairly high flux level. When power is next applied its phasing will determine whether the flux is driven further toward saturation or not. Going into saturation produces a very high magnetizing current. See my scope plot at .http://beta-a2.com/EE-photos.html Photo P6.

In this plot the core is driven well into saturation. The peak current is 40 A and this compares to a steady state RMS current for a fully loaded transformer of 1.46 A. A ratio of 27.4, and if the 40 is changed to RMS, then the ratio is 19.4 .

.

 

[kwired]

Look at Table 450.3(B). Transformers 1000V or less and 9 amperes or more says 250% fusing on the primary.

That is maximum permitted. If you are trying to protect secondary side with primary device, that is too high. If you can come up with a fuse type that will allow the inrush current you can protect the secondary with the primary where this is allowed though.

 

[sparkie1]

230402-0824 EDT

A transformer core has a residual flux level after turn off that results from the last turn off timing and load.

With relatively good core materials this can be a fairly high flux level. When power is next applied its phasing will determine whether the flux is driven further toward saturation or not. Going into saturation produces a very high magnetizing current. See my scope plot at .http://beta-a2.com/EE-photos.html Photo P6.

In this plot the core is driven well into saturation. The peak current is 40 A and this compares to a steady state RMS current for a fully loaded transformer of 1.46 A. A ratio of 27.4, and if the 40 is changed to RMS, then the ratio is 19.4 .

.

I had a chance to look at your plots. Thank you for this, as it does likely explain what I'm seeing here. Both panels that were tripping did have a significant load on them from the power supplies.

As far as what these are doing and/or changing them, the bottom line is that I can either change out the fuses to a larger size (code violation), change out the fuses to a circuit breaker with something like a D trip curve, change out the fuses to a different time delay fuse (the fuses in there are time delay and should be able to handle these), or just use the 15A circuit breaker in the bucket next to it. Because I have secondary protection, the 15A circuit breaker should be just fine for the task.

 

[winnie]

1) have you determined what the reason was for both fuses and circuit breakers in the first place?

2) if the 15A circuit breaker remains, what is the code violation of using a larger fuse?

Jon

 

[sparkie1]

1) have you determined what the reason was for both fuses and circuit breakers in the first place?

2) if the 15A circuit breaker remains, what is the code violation of using a larger fuse?

Jon

1) Some engineer sat there and designed this particular setup with the transformer and the power supplies and put the fuse in there so it can realistically be fed with whatever size of circuit breaker went into the bucket.

2)Using a larger fuse exceeds the 250% allowed for a transformer with both secondary and primary protection.

I see what you are saying on #2. I'm removing it to get the fuses out of there. Just something to blow and cost the company more money and cause down time with no real reason to be there.

 

[winnie]

1) If the purpose of the fuse is to permit a system that works with any size breaker, then in the case where the breaker is suitably sized, then I agree it is fine to eliminate the fuse.

2) The FLA of a 5 kVA 480V primary is about 10A. 250% is 25A unless I am missing something.

Jon