Preventing inrush current on a variable transformer

[Polesch]

I just recently bought a 750W rated variable transformer, it goes to 250 VAC and can supply 3 A. The toroid weigh about 6-8 kg. But I'm having a bit of a problem, the toroid trips the circuit breaker whenever I plug it into the wall socket. My question is this, which is the best or simplest way to avoid this? Do they sell "soft start" socket adapters? What about a lightbulb in series, with a switch that shorts it? Could I solder a thermistor on the input?

Would be thankful for any advice!

[SeanB]

NTC inrush current limiter, easy to find in any old PC power supply. Just look for the black disc near the input bridge rectifier and place in series with the incoming mains.

[Polesch]

NTC inrush current limiter, easy to find in any old PC power supply. Just look for the black disc near the input bridge rectifier and place in series with the incoming mains.

I've looked through my bins of salvaged components, and I've found a couple of black discs which appear to behave just like thermistors. But they have no markings, so I can't find their specifications. They both measure around 8 ohm at 15 C room temp, which drops quite steadily to about 5.5 ohm with my body heat, and down to 3 ohm when pressed against a warm cup of coffee. Doubt I'll try to use them on the mains though.

What do you think about this one? http://www.tme.eu/en/details/b57237s0709m/protection-ntc-thermistors/epcos/#

Maybe two of them in series?

[SeanB]

Those are the right ones, with  a dull black finish and the resistance is about right.  You can use one only, as they will not have any other effect in series other than wasted heat. You can use the EPCOS device, but it will likely look and act exactly the same as the pulls. Just leave the leads long and remember that these do get very hot in normal use, so allow them to have air flow around and do keep the disc body away from anything that will melt at under 180C. They can reach that in fault conditions before they blow apart.

[TechJunkie]

Interesting... I have a 5KVA/20A Volteq and it has never caused a breaker to trip. I do get a bit of dimming for a brief few milliseconds, but that's about all. The NTC inrush should do the trick for you. If you don't mind, what is the make of the unit? Just curious!

[SeanB]

Probably due to the higher source impedance in the USA allowing the wire and transformer resistance to limit current to below the level required to trip the breaker in under 5 cycles.

In my case the supply is 238V nominal, never below 230V, and the 250kVA substation is visible from outside my window, across the road. Cabling is 40mm copper all the way to the meter, and from there 16mm to the distribution unit. From there 2.5mm cable, so the supply impedance is very low, short circuit current is somewhere between 10 to 15kA, so you need a large breaker for it to be disconnected safely. My variac has a DC resistance of around 1R on the whole coil, and thus the large inrush current.

[TechJunkie]

Probably due to the higher source impedance in the USA allowing the wire and transformer resistance to limit current to below the level required to trip the breaker in under 5 cycles.

Good point, thanks SeanB!

Just out of curiosity, I have a clamp meter that does inrush. I think a little test is in order. 

[TechJunkie]

Just checked my Volteq inrush this morning. 55.2 Amps! 

[SeanB]

Over how many cycles though. A breaker may be rated for tripping at rated current over a period from 1 minute to an hour, but at 2x it will trip within 60s, and at 5x within 10s for many types. Often you find that at 10x load it might take around 30 cycles to trip.

http://static.schneider-electric.us/docs/Circuit%20Protection/Molded%20Case%20Circuit%20Breakers/0100-400%20A%20Frame%20FA-LA/FA-FC-FH/0600DB0105.pdf

Has a set of typical curves in it. You can see that to trip in under 1 cycle you need 40x the rated load. Some might trip at rated current if it is held there for a few minutes though, but others will not trip until you have a prolonged current of 1.2 times the rated current.

You also get an idea of how cascaded breakers react to startup currents and to short circuit conditions. You might already have seen a short circuit that tripped the main breaker, but that none of the lower current ones tripped as well.

[TechJunkie]

Over how many cycles though.

Good info, thanks for the link too! I don't think it's more than 5 or 6 cycles to be honest. I can verify that, be curious anyway.