I have a fairly large (several pounds) old wire-wound transformer, which steps down 220V to 110V (or, I suppose any other suitable voltages, but that's the intended use).

I don't know the rating of this transformer. It is not indicated anywhere on the unit. Based purely on size and weight comparison to "modern" transformers1, I would estimate the rating at around 1000 VA, but I can't be sure.

I'd like to run an ~800 VA load on this transformer, but monitor for failure.

What would be the typical failure mode of a transformer me if I've exceeded the load? Overheating? Voltage drop on the load side?

1 "modern" is in scare quotes here because I don't think this tech has changed much in decades.

  • 1
    \$\begingroup\$ You are very correct on the “modern” part. \$\endgroup\$
    – winny
    Mar 27 '20 at 7:13
  • \$\begingroup\$ (1) A photo will be good here. (2) If the transformer has other windings except 110/220 it is good to know if the 110v winding is the most powerful. It may pretty well not be and the transformer may be, say, 800W @ 24V and 50W @ 110V, depending on the original intended use. \$\endgroup\$
    – fraxinus
    Mar 27 '20 at 10:03
  • \$\begingroup\$ One more: If the transformer is intended to power something at selectable 110/220V then you can pretty much abuse it as an autotransformer at about twice the original rating. \$\endgroup\$
    – fraxinus
    Mar 27 '20 at 10:06
  • \$\begingroup\$ @fraxinus - I believe it is intended for 220 -> 110 step down as the only labeling on the transformer indicate 220 and 110 and the polarity of the plugs imply that direction (i.e., the 220V end is male, at the end of a short extension, and the 110V is female built into the unit). I'll work on getting a picture, a bit tough as I'm using it at the moment and I'll have to remove the metal casing to see anything interesting. \$\endgroup\$
    – BeeOnRope
    Mar 27 '20 at 21:25
  • \$\begingroup\$ @fraxinus - it is not selectable. \$\endgroup\$
    – BeeOnRope
    Mar 27 '20 at 21:25

A transformer has several failure modes, all to do with the insulation.

1) It can punch through promptly, due to overvoltage. At these low operating voltages, this is only likely to happen if there's a very large mains spike, and there's not a lot you can do about it. Transformers designed for mains connection have to be designed to survive 1500 V spikes as a minimum.

2) It can degrade over time. Keeping the operating temperature down to something you can put your hand on will generally mean this process is slow enough not to be an issue in your lifetime, and is rarely an issue in low voltage transformers like this.

3) It can be damaged promptly with heat, aka smoking or bursting into flames. This is the usual way we damage transformers, when trying to push them beyond their design ratings.

It's not the current, or the current density that kills the transformer, but the temperature. As the temperature can take seconds or minutes to rise, this means you can safely draw excessive current from transformers for short periods of time, as long as the maximum insulation temperature is not exceeded. It also means that if you're going to do a soak test to estimate the maximum safe current, you need to run it for an hour or two, to let it reach its final temperature.

While it's fairly easy to measure temperature rise, it's not possible to know what grade of insulation is used in the transformer, premium transformers may use higher temperature capable insulation. Fortunately heat goes as current squared, so you don't lose too much possible power by making a guess. I'd be comfortable with a 50°C rise above a 25°C ambient, but if you took a chance and went for 75°C rise, you'd only be looking at sqrt(3/2) or roughly 20% more power.

Measure the resistance of a winding before you start. The tempco of copper is around 0.4% per C, or roughly 10% increase in resistance for 25°C rise in temperature. Run at power for a few minutes, disconnect and measure the resistance. Repeat the cycle until you decide the temperature has stopped changing, or you stop the test at that power because it's going to get too hot. This will allow you to determine the power rating of the transformer.

It has another rating, the regulation. On load, the output voltage of a transformer will sag by a few percent. You have to be happy that this sag is not excessive for your application, a typical figure is around 5% or so. This is rather quicker to measure. Even if you're happy with a large sag, you still have to operate the transformer within its power rating.

I've not discussed voltage overload. Don't. Unlike current overload, there is no margin for going a bit higher on the rated voltage. Don't.


The failure mode of a transformer is insulation failure. Insulation failure is the combined result of temperature and time. Ultimately the life of a transformer is determined by the number of total hours that it has been used and the temperature of the winding during that time. There is some level of current at which a transformer will last for many decades. There is another level at which it will fail in less than an hour.

Insulation failure would most likely result first in a short circuit between one tour and the next. Some shorted turns will result in additional heating and more shorted turns. Ultimately, the primary could short to the secondary or to the core.

If you don't know the transformer rating, you need to monitor the temperature. If you can keep your hand on the transformer comfortably it is probably ok. Someone else may be able to offer a better way of estimating.

Normally, the voltage will drop on the low side perhaps 3 to 5 percent from no load to rated load.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.