How Can You Test a Coil for Shorts Without Disassembly?

[thetexan]

Let's say a radio power transformer or other electronic transformer (audio for example) has a short between two of its coils (for example, the enamel coating has burnt off). Is there any way to test it to find out without taking it apart?

As I see it the short will be either between adjacent coils or in the coil that is in the row just above or just below. In either case I would think the change in impedance or inductance would be minimal. Plus to check for a change in values would require that you know the original value.

A short to the casing would be easy to check for continuity. But the short between coils I think would be impossible.

Any ideas,

tex

 

[berkeman]

Let's say a radio power transformer or other electronic transformer (audio for example) has a short between two of its coils (for example, the enamel coating has burnt off). Is there any way to test it to find out without taking it apart?

As I see it the short will be either between adjacent coils or in the coil that is in the row just above or just below. In either case I would think the change in impedance or inductance would be minimal. Plus to check for a change in values would require that you know the original value.

A short to the casing would be easy to check for continuity. But the short between coils I think would be impossible.

Any ideas,

tex

You can measure either resistance or inductance. Inductance would be a more sensitive measure, since inductance is proportional to turns^2, and resistance is proportional to the number of turns. But as you say, you need to know the initial value to make a good comparison. Otherwise, you need to know the specification value from the datasheet, as well as the tolerance.

If it's just a single adjacent-winding short, you probably will not be able to detect that (and it shouldn't affect the performance of the transformer much).

 

[jim hardy]

One can measure how much no-load current it draws. But he needs to know what it should read.

It shouldn't get hot with no load.

Shorts between adjacent turns of a winding will heat those turns and the damage will spread.
Shorts between adjacent layers of a winding will heat more turns and the damage will spread faster.
A winding to winding short will show by continuity checks. I've had transformers that wouldn't work in circuit but on the bench gave correct voltage on each winding...sneaky...

So check by continuity for winding to winding shorts ,
then plug it in and leave it overnight... on something that won't burn.

If you notice that upon removing your ohmmeter lead you get a shock from the transformer's inductive kick, it's probably okay.

 

[berkeman]

If you notice that upon removing your ohmmeter lead you get a shock from the transformer's inductive kick, it's probably okay.

LOL

 

[Averagesupernova]

Doing a ring test can indicate some things. A tank circuit is formed with a capacitor in parallel with the coil and the network is pulsed. The circuit is observed on a scope for a decaying oscillation or something that counts the number of decaying oscillations. It is somewhat subjective but if you have known good parts to compare to or a spec you can usually find problems.

 

[Windadct]

Since many coil shorts are intermittent or not a dead / solid short these can be difficult. IMO the best test is a full voltage test with no load - this is easiest usually be energizing the low voltage side with a variable AC source, and an ammeter in series with to measure the current. If there is a short ( or damaged insulation between winding) - the that winding will conduct at some voltage - and the current will increase dramatically.
Some transformer datasheets will list the DC resistance of the windings - it may help, but if the short - is not a great connection you will probably not see the resistance vary much from the datasheet value.