What does surge current look through large 30a transformer?

[CopperCone]

So if you have a 30a transformer meant for mains isolation with a 1:1 winding ratio, what does the surge current through it look like when shorted?
I know the outlet is capable of like 5kA when shorted out before the breaker trips. How much current limiting will such a transformer provide?

[Benta]

It's practically only limited by the primary and secondary resistance, so it will be very large. The primary to secondary magnetic coupling is high on such a large transformer, so leakage induction will not help you.

[CopperCone]

Does it make any difference during a transient? I don't have any feeling for such things.

Might it saturate and help somewhat, or is the magnetic geometry of such a large device such that it makes little difference?

I.e. I can see a benefit towards using higher voltage MOV/GDT infront of the transformer, to protect the insulation, but to use lower voltage MOV/GDT after the transformer, to protect after isolation, but at reduced peak currents to extend their lives, as a kind of voltage selective passive protection network.



My rational is this:

500V transient might stress a transformer winding insulation, so a 500V* MOV should be placed infront of the transformer, exposed to full mains impedance, where it is less likely to go off, but say a 200V* mov placed after the transformer, to act as the primary circuit protector, but with the impedance of the transformer.

Any idea how much benefit I can get doing this? *arbitrary numbers require some calculations

[T3sl4co1l]

What kind of transformer?

Not all are created equally.  Reactance makes as much difference as resistance.

Tim

[CopperCone]

I dont know i guess a industrial isolation transformer that uses a e core or whatever is common i never saw one larger then a mot before

[T3sl4co1l]

A shell type transformer design should have less reactance than resistance, and a short circuit current around 20 times rated.

Toroids can be more.  Bank wound transformers (typical of smaller and cheaper transformers) have more reactance (maybe S/C 5x).

Tim

[CopperCone]

Thats pretty significant in terms of surge. A 30A transformer will only surge 600A, vs the 4+ kA from a shorted outlet protected only by circuit breaker. 85+%  reduction. I would imagine this to be much less explosion.

[SG-1]

One other consideration is the distance to the mains & the wire size.  The wire length can add a good bit of impedance.  UL did a study here in the US.  They found in houses larger than 3000 sq. ft. that less than 75 amps of fault current was available on the outlets farthest from the service.  This is was concern because the breaker could never get into the instantaneous trip portion of its trip curve.

[SG-1]

One correction on my last post.  You need to total the circuit impedance all the way back to the distribution transformer.  5K amps of fault current seems high.  I am thinking a 15kva distribution transformer is only capable of 2 or 3 thousand amperes.

[CopperCone]

I guess i need to make. Better rogawski coil to measure the fault current.
The spread i got from my research is between 75 amps to 5000amps. Kinda fustrating.

[Benta]

Just put an ohmmeter on the primary and the secondary to measure the resistances. This will give you the maximum possible current.

Important!! Short the winding not being measured!! Otherwise you and your ohmmeter might get a whack. (Think ignition coil).

[CopperCone]

I guess that depends on how insignificant leakage inductance is.

But yea i need to measure the outlet short circuit current or find good data on pole pigs and home wiring impedance to see if a transformer can be justified.

[CopperCone]

Well I looked online for some distribution transformer stuff, it really depends on what you have in your street:

https://iaeimagazine.org/magazine/2015/07/07/calculating-short-circuit-current/

You can have between like 500-5000kVa on the pole it seems near your house.

http://www.ecmweb.com/content/beware-simplistic-fault-current-calculations

for that beast, 60,000A or so

I am starting to think that its a pretty good idea, if money can be spared, to transformer isolate anything you build..

Relying on the wiring being basically high Z is kinda a design fallacy I think, its winging it.

Not too happy with my plasma cutter now, despite how compact it is.


In my home I have one outlet basically right near the circuit panel, where the washing machines and shit are, vs the garage which is far away (probably protects my precious plasma cutter a bit)

I can see things like crowbar protectors greatly benefiting from a transformer. I believe the benefit of a crowbar vs a open switch is that it won't make a inductive kick from a high current.

will a power company disclose what they put on the pole near a house? Still kinda risky, even if you know, because they might change it out based on requirements. Is there legislation in place for them to provide reasonable current limiting?

[T3sl4co1l]

will a power company disclose what they put on the pole near a house? Still kinda risky, even if you know, because they might change it out based on requirements. Is there legislation in place for them to provide reasonable current limiting?

We have a lot of pole pigs around here with
"50
4800 / 14400"
on them.

Fault current at 240V isn't going to be over 10kA, that's what breakers are rated for.  Presumably they are required to meet that at the drop.  If they don't... you should probably look into that.

I think normal in-house fault current is in the ~2kA range.

Tim

[Berni]

Yeah fault currents depend on a lot of things. Where i live they wouldn't be so high since we have a practice of having one huge transformer supplying a few streets, but we have 230/400V 3 phase mains here so it also means thinner wires. But i happen to be towards the end of the line tho so there is a lot of wire to provide resistance. The guy living next door to the transformer tho likely has a pretty huge fault current since these transformers are huge and the wire running down the street is designed to handle the load of the entire street where they simply tap it off.

Not sure how they deal with the fault currents in those cases, but we do also have big main fuses where the feed comes into the building. These are big sand filled fuses the size of a fist so they likely handle pretty significant fault currents.

From what you are explaining it sounds like you are trying to use a 1:1 transformer as a surge protector to save expensive gear from lightning and similar. This is actually a good way of doing it. Insulation on mains transformers can easily handle 1kV as they are designed with a fair bit of safety margin but obviously has its limits so a MOV is not a bad idea. The output however might not need a MOV at all but wont hurt. Mains transformers are designed to operate fairly close to core saturation (Like 80% of the way there) because that way they can use less iron and copper to make a transformer of a given power rating. If you try to put significantly more voltage into a mains transformer it quickly causes the core to saturate and this makes it act more like a air core transformer. Due to this the inductance of the primary falls of a cliff, causing the transformer to draw massive amounts of current on the primary, likely enough to cause its power source to droop down in voltage somewhat. At the same time the coupling between the coils and the secondary inductance plummets as well, this hampers the transformer ability to transfer the power out to the secondary terminals and reduces the output voltage. Additionally very fast transients have a hard time making it trough mains transformers because they are designed for 50Hz so the high frequency content mostly ends up inductively heating the iron core.

So yes a transformer is actually an excellent surge protector, but it is indeed wise to protect its input with a MOV and fuse, perhaps also a gas discharge sparkgap for very fast transients.

[CopperCone]

This is kind of interesting.

This was all really spurred by an induction heater i started to work on.. But i feel i am really only protecting the capacitor and diode bridge, since the output short circuit current of the filter capacitor is likely worse then the damn mains transformer.

I dont know if there is some way to protect a igbt h bridge from the transformer short circuit current save for using something even beefier like a SiC that costs even more then the igbt to drain and isolate the capacitor.

[Berni]

Well with the sort of high power electronics its difficult to really protect everything. When things go wrong they tend to go wrong rather spectacularly. Best you can do is run things at lower power and with current limiting resistors in place to make sure it works before letting it go full wack at the designed power levels where things can literally explode.

But if the point of this was to limit the massive inrush current of such a big mains rectification circuit then a 1:1 transformer is a horrendously inefficient way to do it (But can be done as you can design a transformer in a way to have built in safe current limiting functionality). A common way this is solved in switchmode powersuplies is by connecting the whole thing to the mains with some sort of ballast in series. That way the ballast is limiting the input current to a sensible level to gradually charge up the capacitors and after a short delay a switching device activates that bypass the ballast and applies full mains power to the already full capacitors, at that point the power supply then begins its usual startup procedure.

The ballast can be anything that limits current, at low powers a power resistor could be used, at high powers a inductor is better suited (Much like the ones used to run fluorescent lamps). The switching element is often just a simple relay, but could also be some sort of solid state switch such as a triac.

[T3sl4co1l]

I dont know if there is some way to protect a igbt h bridge from the transformer short circuit current save for using something even beefier like a SiC that costs even more then the igbt to drain and isolate the capacitor.

Desat protection.

There is absolutely nothing you can do upstream, unless it's equivalent (semiconductor based protection).  Even then, upstream limiting doesn't account for local bypass caps, which may contain enough energy to push already-hot transistors into the magic smoke range.

Tim

[capt bullshot]

I dont know if there is some way to protect a igbt h bridge from the transformer short circuit current save for using something even beefier like a SiC that costs even more then the igbt to drain and isolate the capacitor.

No way. The brigde can be protected from overload / short applied to its output, using desat or current sensing. Works if properly done and has a lot of caveats. If the brigde itself fails, nothing can protect it, the upstream fuse does nothing but to limit the collateral damage. This is a standard test (failure of one switch in the bridge) in your lab if your company develops VFDs, Servo drives and such stuff. Ends up always with smoke, but total damage must be limited, e.g. the device goes a loud boom or bang and smoke, but won't catch fire, nor are shrapnels flying through the room. You'll destroy a good handful of your devices (not only the semiconductors, but the whole inverter or whatever it is) under development due to this testing process.

[CopperCone]

Why can't you use more switches to crowbar or open the bridge up if you start getting weird currents and a lack of positive response from the de-sat or other protections?

[Wolfram]

Because it is more expensive and complicated, and less effective? Any switching device between the local decoupling and the main bridge transistors will add losses and stray inductance. If your protections don't work properly, they are badly designed. This is best fixed by designing them properly in the first place, not by adding more (possibly badly designed) protection circuits.

[CopperCone]

crowbar tube won't really add much of anything like I mentioned above, other then requiring an expensive firing circuit

i think it depends , layered protection will depend on the application, not everything is about losses or cutting a bit off EMI margins

I am still interested though, I want to see where this is going (not only what some industry decided to do)

The local capacitors are a problem though, especially if they are bolted onto the IGBT terminals. Even for a tube. You would probably need to design something real schmick to put a bypass on the bypass capacitor. Sounds kind of fun to see what results you can get.

Is there joule estimations on what can fuck up a fresh IGBT? Possible to derate on heat and put slightly inadequate bypass (I assume you mean average junction temperature and not a joke)?

I did not see the device curves go into that kind of instantaneous current territory but I will think about it and look again.

[rstofer]

You can get the potential fault current on the secondary if you know the percent impedance of the transformer.

https://www.quora.com/Why-is-the-transformer-impedance-given-in-percentage

Short the secondary winding with an inline ammeter.  With a variac, increase the primary voltage until you get full nominal current flowing through the secondary.  It won't be much, probably nominal primary voltage multiplied by 0.06 or 7.2V for a 120V transformer.

This measure voltage divided by the nominal voltage gives the percent impedance (which I assumed to be 6% when I ginned up that 7.2V number).

Now you can treat the transformer as a resistor.  You just take the nominal full load secondary current and divide by the percent impedance and you have the maximum fault current.

Percent impedance is usually a low number for utility transformers, on the order of 6%.  So a transformer with a nominal secondary full load current of 1000A will have a short circuit current of 1000/0.06 or 16,666 Amps.  This assumes that the source to the primary can supply the reflected primary current.

In the electrical business, we normally assume 'infinite bus' for 12kV -> 277Y480 transformers because the voltage is so high that full primary current is quite low and fairly easy to deliver and 16x current still isn't very much.

I would expect a residential transformer to be more like 75 kVA than 15.  It would depend on how many houses it serves.  A 75 kVA 240V (single phase) transformer has a full load secondary current of 312 amps and, if the transformer has an impedance of 6%, the maximum fault current is 312 / 0.06 or 5220 Amps.  Maybe the transformer is a little smaller (or the impedance a little higher) because most residential breakers around here are rated for 5000AIC (Amp Interrupt Capacity).

[CopperCone]

I thought I saw in home depot the circuit breakers are rated for 10kA breaking.

[T3sl4co1l]

I am still interested though, I want to see where this is going (not only what some industry decided to do)

Industry practice varies.

When I started at Radyne, they were using 20+ year old Inductoheat designs with wide open loop switching.  Literally a TL494 going into drivers, except not using a TL494, but drawn out with discrete logic (gates, timers) on a 10" wide PCB (with tons of free space, but still dozens of chips).  No desat protection: each gate drive was an opto, a linear supply (transformer, rectifier and filter cap), and an H-bridge (using generic MOSFETs) to drive the IGBT module gates. So, +/-15V and a modest number of amps.

When I left, they were using new controls with gate drives using SMPS power, planar transformers, magnetic isolators (ADUM series, it was a few years later that alternatives arrived from other manufacturers), IXDD614CI drivers, and desat protection, acting within a couple microseconds.

In the first week I worked there, a technician had made a wiring mistake inside a cabinet, I think a 300kW model.  He turned on the power, hit the "start" button... KABOOM!

When I left, there was probably a dozen events (that I was aware of) where a unit had been started, and it just... clicked.  Press the "start" button, "tick".  Huh.  Press it again, are you sure it's working?  "Tick."  Hrm.  Maybe it's... did we wire this one right?  Oh... oohhhhh.

That includes the intentional events, where we looped a fine Rogowski coil inbetween the IGBT bus bars and observed the transient.  A nice smooth ramp up to 6kA or so, then off within a couple microseconds.  Just as expected.

Only popped one IGBT module throughout development, and that included a large (~1MVA) research unit.

Is there joule estimations on what can fuck up a fresh IGBT? Possible to derate on heat and put slightly inadequate bypass (I assume you mean average junction temperature and not a joke)?

IGBTs are often rated for short circuit of some duration, usually 5 to 20us.

As in my above example, take 10us * 700V * 6000A = 42J.  For a 1200V 600A IGBT (not a single die, but a few arrayed, I forget how many).  It's basically die heating, same as desat protection on MOSFETs (a slightly less urgent but nonetheless applicable protection method -- we did this also on the smaller power supplies, at up to 400kHz), which are sometimes rated for short circuit durations, and almost always rated for avalanche energy which has a similar limit (die temperature).

In any case, the limiting factor is die temperature, so use the transient thermal curve to find out.

Tim