Help solve a debate between my twin brother and I (both electrical engineers) read message below for details

[PE (always learning)]

Let me see if I understand the argument. He said, "The sky is blue." You replied, "No, you are wrong. The grass is green." Do I have that right?

1. I agree that 125 amps is industry standard. I've never seen any other size used for this application.

2. I also agree that a 100 amp would "likely" be successful. I say "likely" because we, as engineers, generally have very limited control over the exact makes and models of components that the contractor purchases based on our drawings and specifications.

3. I have nothing to offer on the question of relative costs.

4. If you called for a 90 amp breaker, you would be getting the blame for what will almost certainly be occasional (if not downright numerous) unnecessary trips.

5. A PE's job includes protecting the client's interests. That does include costs, but it also includes convenience (e.g., future operational and maintenance considerations).

Look at the Siemens absolute maximum and practical inrush chart provided in my pdf. You can clearly see that the inrush is half as much as what is assumed and that is at absolute maximum! I've seen the same results with other manufacturer's data as well. If you see my time current curve charts you will also see that the 90 amp breakers are well above the inrush current and again this is 90 amps!

 

[ptonsparky]

Small scale testing with a 1 ph transformer in my shop. Testing to catch the peak with a Fluke 43B. It would trip often enough with the breaker I was using that I doubt peak sign wave was it entirely. I was trying to see if an added load would cause added inrush. It did not.

 

[JoeStillman]

Not only that, but the sign wave would have to be at the peak when they energize for them to have 578 amps, which again is well below what the 90 amp breakers show on their curves.

Finally, let's say that they did trip the primary breaker upon energizing with some other manufacturer's transformer somehow and it was caused by inrush. Couldn't they just reset it and hope that it doesn't energize when the sign wave is at it's peak.

I think the point Charlie was making @#13 is that, as engineers, we owe it to our clients to see that they never have to reset breakers under normal operations, like retransfer or utility restore.

 

[JoeStillman]

Gentlemen,

Please reference the following:

1) Siemens Series J Typical Performance Data for an Aluminum 3 phase dry type K1 without electrostatic shield transformer at 150 degree temp rise. They have produced charts with the absolute peak inrush and the practical max inrush in amps. I have clouded them for reference. SKM typically assumes that the inrush is 12 times the FLA of the primary current which would put you at 90.317 amps x 12 = 1083.815 amps of inrush current. Now compare that to the absolute peak inrush of the 75 kVA transformer produced by Siemens, which clearly shows 578 amps of maximum. This is half as much as what you would normally assume for inrush. The practical inrush is even less at 193 amps.

2) Reference a time current curve plot I've created using Siemens absolute peak inrush vs 90A breakers of several different manufacturer's.
You will see that they all clear the inrush with no issue. Again, this is at 90 amps!

Just food for thought....

I wonder why SKM doesn't have that chart in their default library? They only have a few Square D models in there and the inrush factor is 12.0 for all of them.

I'm thinking I need to add that chart to the library.

 

[charlie b]

Couldn't they just reset it and hope that it doesn't energize when the sign wave is at it's peak.

That notion (particularly the "hope" part) should NEVER be taken into consideration in any project undertaken by a PE! We owe our clients the best we have to offer.

 

[charlie b]

Look at the Siemens absolute maximum and practical inrush chart provided in . . . .

If you see my time current curve charts you will also see . . . .

I don't need to look or to see. I will concede that you are right. We will call this an academic discussion and you have won your point.

That said, I strongly recommend that you don't attempt to sell this argument to a client, be it an architecture firm, an engineering firm, a contractor, or anyone else. Nobody is going to want to read your explanation, let alone want to think their way through it. If your drawings were to show a 90 amp breaker in this application, you are going to get questions. It will not be worth your time or the client's money for you to deal with such questions.

 

[PE (always learning)]

That notion (particularly the "hope" part) should NEVER be taken into consideration in any project undertaken by a PE! We owe our clients the best we have to offer.

Hope was not the right wording, I just mean that the worst case scenario is that they would trip a 90 amp breaker upon energizing and then reset it and probably never have the issue again unless the whole facility went down, in which case you have bigger problems...

I agree that I don't deal with hope when I do my work, that's why I wanted to back up my statements with the siemens inrush charts and the breaker trip curves.

I also never argued for using a 90 amp breaker to protect a 75 kVA transformer, I just wanted to show you that even with a 90 amp breaker you're well above documented inrush currents. I would most likely use a 100 amp breaker if going up to 125 amps forced me into a sub feed or if one of my contractors wanted to reuse an existing 100 amp breaker that was available, but yea 90% of the time 125 amps would probably be what I'd use.

I apologize if I'm coming off as pretentious, I've just done enough of these studies to see what the real values are vs the codes values (which are always overly conservative).

I don't think there is anything wrong with upsizing to 125 amps, there are multiple ways to skin a cat and I just think that a lot of times people fall into the trap of going by what they call "industry standard" when there are perfectly fine solutions that involve less conservative answers as well.

Anyway, please don't take my feed back the the wrong way, I was never arguing with your points, I just wanted to bring up something I thought was interesting and share.

 

[PE (always learning)]

I wonder why SKM doesn't have that chart in their default library? They only have a few Square D models in there and the inrush factor is 12.0 for all of them.

I'm thinking I need to add that chart to the library.

Yea, SKM always defaults to the 12 times inrush factor, which lead me to a lot of problems when publishing my studies for other clients / engineer's designs. I kept having to tell them to upsize their breakers and got sick of it, so I asked Siemens and other manufacturer's to give me the real data so that I wouldn't have to unnecessarily force my clients into upsizing breakers after they were already bought and on site.

 

[Tulsa Electrician]

, I just wanted to bring up something I thought was interesting and share.

Thank you.

 

[PE (always learning)]

I wonder why SKM doesn't have that chart in their default library? They only have a few Square D models in there and the inrush factor is 12.0 for all of them.

I'm thinking I need to add that chart to the library.

Attached is a pdf to the full chart for your future reference.

 

[PE (always learning)]

When it comes to larger transformers, I whole heartedly agree that I will stick with sizing the overcurrent protection higher as the larger transformers seem to be more in line with the inrush factor being 12.

 

[Knightryder12]

I think you are off base on this. I am pretty sure every manufacturer has a 125A three pole panelboard breaker. For siemens its an NGB or a 3VA41. For a square D NF panel its an EDB. For GE its a TEY Not real familiar with eaton but Im sure its the same for them. I didnt price anything out, maybe there is a big price jump but since its the same frame I wouldnt think it would be that much more.

I think Sq D is limited to a 110A, 3 pole in a normal breaker space.

 

[MD Automation]

...Couldn't they just reset it and hope that it doesn't energize when the sign wave is at it's peak...

Pardon me if I am reading any of this thread incorrectly, but I believe some are thinking the worst case scenario for inrush is switching on at the peak sine wave of the Mains AC?

The science behind transformer inrush fascinates me (nerd engineer!!) and I wanted to say that I believe the worst case scenario for max inrush current is, counter intuitively, NOT at peak sine wave turn on, but rather at the zero crossing on the Mains.

Please somebody correct me if I'm wrong. I have been wrong before and surely will be again!!

Powering up a transformer is different than powering up a pure resistive load. Obviously, the inrush for the resistor will be highest if you switch on at the Mains peak.

But... the transformer is an odd duck - you are looking to re-establish a rotating magnetic field in the core. But... there is also a residual magnetic flux left in the core from when you last turned it off. This flux has a "direction" or polarity. The magnitude and direction of this leftover flux is totally random, depending on when the Mains was switched off. And it's hard to remember when the transformer is cold and dark - but - there is something leftover and waiting in it magnetically. And it can help or hurt you at next power up.

Also, the voltage applied to the primary windings is 90 degrees out of phase with this rotating flux. One way to think about it is the whole positive (or negative) half of the Mains sine wave input is responsible for building up the flux in one particular direction / polarity. And so when the voltage crosses zero, you have just finished doing all you can to build up the max flux you can in that direction / polarity. And at the Mains zero crossing you are going to start moving that peak flux in the other "direction". So they are (voltage vs flux) 90 degrees out of phase.

But suppose you froze everything at that zero crossing - with the flux at a maximum - and then instead of the sine wave crossing zero and heading in the "other direction" - suppose it went back up in the direction it was already coming from. Meaning it went back positive if it was at 180 degrees - or went back negative if it was at 360 degrees. Then you will start adding to the flux in the core where it was already at its peak - and trying to push it higher still.

So the absolute worst thing to have happen then is to have a leftover residual flux built up in one particular polarity - and then - switch on the Mains at the zero crossing where the next half sine wave will try to build up the flux in that same direction. This will, naturally, drive the core deep into saturation. With the maximum inrush current resulting.

Of course it's a double crap shoot - meaning where the flux is when you turned if off - and where the Mains are when you switch it on. But the worst case scenario will happen when the Mains are at the zero crossing and getting ready to drive the magnetic flux in a direction it's already at.

Hope this makes sense. It's the way I was taught to think about this topic - and I always thought it correct.

 

[electrofelon]

I think Sq D is limited to a 110A, 3 pole in a normal breaker space.

Look at a EDB34125 , that is for their NF panelboards.

 

[ptonsparky]

Pardon me if I am reading any of this thread incorrectly, but I believe some are thinking the worst case scenario for inrush is switching on at the peak sine wave of the Mains AC?

The science behind transformer inrush fascinates me (nerd engineer!!) and I wanted to say that I believe the worst case scenario for max inrush current is, counter intuitively, NOT at peak sine wave turn on, but rather at the zero crossing on the Mains.

Please somebody correct me if I'm wrong. I have been wrong before and surely will be again!!

Powering up a transformer is different than powering up a pure resistive load. Obviously, the inrush for the resistor will be highest if you switch on at the Mains peak.

But... the transformer is an odd duck - you are looking to re-establish a rotating magnetic field in the core. But... there is also a residual magnetic flux left in the core from when you last turned it off. This flux has a "direction" or polarity. The magnitude and direction of this leftover flux is totally random, depending on when the Mains was switched off. And it's hard to remember when the transformer is cold and dark - but - there is something leftover and waiting in it magnetically. And it can help or hurt you at next power up.

Also, the voltage applied to the primary windings is 90 degrees out of phase with this rotating flux. One way to think about it is the whole positive (or negative) half of the Mains sine wave input is responsible for building up the flux in one particular direction / polarity. And so when the voltage crosses zero, you have just finished doing all you can to build up the max flux you can in that direction / polarity. And at the Mains zero crossing you are going to start moving that peak flux in the other "direction". So they are (voltage vs flux) 90 degrees out of phase.

But suppose you froze everything at that zero crossing - with the flux at a maximum - and then instead of the sine wave crossing zero and heading in the "other direction" - suppose it went back up in the direction it was already coming from. Meaning it went back positive if it was at 180 degrees - or went back negative if it was at 360 degrees. Then you will start adding to the flux in the core where it was already at its peak - and trying to push it higher still.

So the absolute worst thing to have happen then is to have a leftover residual flux built up in one particular polarity - and then - switch on the Mains at the zero crossing where the next half sine wave will try to build up the flux in that same direction. This will, naturally, drive the core deep into saturation. With the maximum inrush current resulting.

Of course it's a double crap shoot - meaning where the flux is when you turned if off - and where the Mains are when you switch it on. But the worst case scenario will happen when the Mains are at the zero crossing and getting ready to drive the magnetic flux in a direction it's already at.

Hope this makes sense. It's the way I was taught to think about this topic - and I always thought it correct.

Gar would have the answers.

 

[GoldDigger]

During the design stage, when I have no idea which of the 3 or 4 manufacturers will be chosen to supply the project with equipment, it is not worth the headache associated with a potential for inrush trips to be tighter than 125% of FLA (I round up from there). Why take the chance with transformer manufacturers trying to design to DOE 2016 efficiencies (and inevitably DOE 2024 - Distribution Transformer Efficiency and Supply Chain Reliability Act of 2024) that they may be higher than 12xFLA or the breaker that is installed is slightly more sensitive than the TCC.

The manufacturers' specs for inrush are based on a step application of voltage at the worst possible point in the waveform.
The step voltage when switching between two full voltage unsynchronized waveforms can be double that.

 

[GoldDigger]

The manufacturers' specs for inrush are based on a step application of voltage at the worst possible point in the waveform.
The step voltage when switching between two full voltage unsynchronized waveforms can be double that.

I did not mean to imply that the worst case was a peak voltage. Indeed for a core with zero or little residual flux it will be at the voltage zero crossing as described.
I did mean that it is likely that the combination of initial flux and applied sine wave can be worse when doing an unsynchronized transfer that would be the case for a "resting" transformer with only residual magnetism.

 

[PE (always learning)]

I think the point Charlie was making @#13 is that, as engineers, we owe it to our clients to see that they never have to reset breakers under normal operations, like retransfer or utility restore.

Totally agree with you that's why I wanted to show you the Siemens charts. Again, I would not recommend using a 90 amp breaker for my typical 75 kVA transformer primary, but I just wanted to let you guys see the reality of the inrush when looking at the actual data. I'd probably go for a 100 amp at my lowest, but if a contractor needed to use a 90 amp breaker I would have no problem using one on a 75 kVA Siemens transformer based off this information.

 

[zooby]

I used to think that way until I saw the actual trip curves and the real inrush points plotted by the manufacturers. The transformer inrush is honestly typically well below a normal 90 amp breaker.

most if not all of the original installs at our campus (1986-87) are 90a to the xfmr.

 

[electrofelon]

Just to expand the conversation, can we discuss fuses vs circuit breakers for an application like this? I always thought a fuse would be better for applications with high inrush, but I don't recall ever actually looking at the corresponding trip curves.