Transformer Overload Question

[Frosty_47]

Dear Komrades,

I have a Torodial step-down transformer. It's got a 39V - 0 - 39V secondary and is ratted for 6.7Amps. I want to attach a capacitive load to the secondary (after the bridge and current limiting resistor). The initial current in the capacitor exceeds the ratted current of the transformer for the first 3 seconds. Is it appropriate to overload a transformer of this type? Below is the current load on the transformer:

Initial = 8.4A

after 3s = 6.64A

after 5s = 5.653A

after 8s = 4.43A

after 10s = 3.77A

Thanks

 

[ecerfoglio]

Dear Komrades,

I have a Torodial step-down transformer. It's got a 39V - 0 - 39V secondary and is ratted for 6.7Amps. I want to attach a capacitive load to the secondary (after the bridge and current limiting resistor). The initial current in the capacitor exceeds the ratted current of the transformer for the first 3 seconds. Is it appropriate to overload a transformer of this type? Below is the current load on the transformer:

Initial = 8.4A

after 3s = 6.64A

after 5s = 5.653A

after 8s = 4.43A

after 10s = 3.77A

Thanks

You max overload is (8.4 / 6.7) - 1 = 25 % , but it's only for an instant, the total overload time is just 3 seconds ans then the load keeps dropping, so i'd say that there should be no problem.

BUT (there is allwaya a BUT) How had you measured the current? It's not constant (DC) and not a sine wave, you should use a true RMS meter or else your readings may be off by more that 25%

Also, you didn't state the duty cycle that the transformer will have - that is how many 3 s overloads per hour there will be.

Your best bet is to try the transformer and meassure i's temperature afer each capacitor carge. If you can touch it and don't get burned, then it should be OK.

But take your ice man costume off before yu do it

 

[Hero999]

That's normal, in fact the initial spike is probably much higher than that, it could be of the order of 100A. Transformers are designed to handle surges so it shouldn'd be a problem.

 

[Frosty_47]

But take your ice man costume off before yu do it

Lol!

But only after I have "collected" all the candy from the neighborhood kids.

 

[Frosty_47]

That's normal, in fact the initial spike is probably much higher than that, it could be of the order of 100A. Transformers are designed to handle surges so it shouldn'd be a problem.

Thank you!

That saves me the trouble of adding another transformer in parallel.

 

[Frosty_47]

BUT (there is allwaya a BUT) How had you measured the current? It's not constant (DC) and not a sine wave, you should use a true RMS meter or else your readings may be off by more that 25%

Also, you didn't state the duty cycle that the transformer will have - that is how many 3 s overloads per hour there will be.

I based my current calculations under worst-case scenario that is based on peak voltage not rms. There is actually step-up microwave oven transformer connected to the secondary of this step-down. There is a bridge rectifier at the output of the microwave oven transformer followed by a 3750 Ohm resistor. The total capacitance connected after the resistor is 3300uF. Which means the time constant is RC = 12.375S. The turn’s ration on microwave oven transformer are:

2000V/120V = 16.7

That makes the output voltage: 78 * 16.7 = 1303Vrms. Or 1842Vpk (I used 1900Vpk in my calculations). So the initial current on the secondary of microwave oven transformer is: 1900V/3750Ohm = 0.507A. Multiply that by 16.7 and you get 8.4613Amps on the secondary of the step down transformer. At the time I forgot that this is peak current so the RMS current would be roughly 6 Amps.

I apologies for not realizing this earlier.

 

[Diver300]

The larger the transformer, the longer the time constant so the longer it can stand a large overload.

The sort of overload time you are talking about is insignificant for a 500W transformer. It's thermal time constant will be about half an hour.

The rms current (when averaged over the first 10s) is only about what the rating is, so as long as you don't start up more often than about every 10 seconds you will be OK.

 

[Willbe]

You need the IsquaredT rating for the transformer. A cooperative Application Engineer at any xformer manu. may help you with this.
The bigger the thermal time constant, the longer it can stand I^2.

For a 1N400x diode it is ~7 amps-squared-seconds, so you are looking for way bigger numbers.

 

[Nigel Goodwin]

The transformer itself isn't a concern, it's blowing the mains fuse that is - so for high power PA amplifiers (using toroidal transformers), anything over 300W+300W RMS often uses a softstart circuit to limit the initial current. With a normal 'square' transformer the in-rush current is far smaller, and softstart isn't normallly used until much higher powers.

 

[Willbe]

Putting 100kW into an 8 pound transformer will raise the temp. by 80°C in one second.
If the impedance is 200Ω, the IsquaredT rating is 500 Amps-squared-seconds, assuming it fails when the windings reach ~100°C.

 

[audioguru]

A transformer winding is a dead short when the current is so high that the core saturates. Make sure that the current surge will not cause the core to saturate.

 

[Hero999]

IS core saturation such a bad thing so long as it doesn't last for very long.

For example if it has a DC resistance of 5Ω then the peak current would be 34A at 120V which won't do any harm if it's just for tens of ms.

 

[Diver300]

IS core saturation such a bad thing so long as it doesn't last for very long.

For example if it has a DC resistance of 5Ω then the peak current would be 34A at 120V which won't do any harm if it's just for tens of ms.

Core saturation will happen on any transformer if it is turned on as the mains input voltage is near zero in the AC cycle. The problem is that the primary of a 500W 120V input transformer will be far less than 5 Ω. The efficiency of a transformer of that size is about 90% so the total loss is about 50 W. If half of that is in the primary winding, the resistance will be 1.44 Ω, and I guess that in practice it will be less, and the peak current will be over 100 A.

The problem is worse with toroidal transformers which saturate the whole core at the same time. Square core transformers saturate some parts before others.