Testing a transformer for shorted turns

[Merlin]

Split from this thread:-

https://www.vintage-radio.net/forum/...ad.php?t=13072

Graham. Station X, Forum Moderator.

--------------------------------------------------------------------------------------------------------------

Originally Posted by ppppenguin

Unfortunately you cannot test the valves or the transformers (for shorted turns) with a multimeter.

But here's a simple way to test for shorted turns on a transformer using a neon lamp and a battery:

[ppppenguin]

Nice simple idea.

I'm a little worried about what might happen if you used it on a low voltage secondary of a transformer. If there's enough voltage to light the neon (60 to 80V) then the voltage across the primary will be excessive.

Also don't connect it to a very low power transformer such as an intervalve one. The current from the battery may be sufficient to burn out a winding.

[oldeurope]

Originally Posted by ppppenguin

I'm a little worried about what might happen...............

The tester should be connected at the primary (230VAC) input only!
Any ideas why there is a 47K resistor?

Darius

[ppppenguin]

Originally Posted by oldeurope

The tester should be connected at the primary (230VAC) input only!

Agreed.

Originally Posted by oldeurope

Any ideas why there is a 47K resistor?

Presumably to avoid a high peak current through the neon. With a big transformer the stored energy in the transformer (0.5*L*I^2) might be enough to destroy the neon. I know that not all the energy will reach the neon because the voltage will quickly drop below the maintaining voltage of the neon but the total energy might still be destructive.

[oldeurope]

Good morning,
I think the neon current is V batt divided by R coil. The resistor makes the "on time" of the neon shorter. If you want to limit the current of the neon, a resistor (470 Ohms for example) should be placed in series with the battery. This makes the lightning more independent in the size of the transformer.

But in real live I use my VARIAC and an amperemeter to test transformers.

Kind regards,
Darius

Lenz told something like: " A coil tries to keep the last current flow alive untill the stored magnetic energie is empty."

[oldeurope]

Good morning,
I tested it and it works fine. 470 Ohms at 6VDC or 1K at 12VDC are good values. I tested a small and a big transformer. Secondary shorted, no flashing.
Connect the neon without series resistor, it makes the flashing time short.
Thanks again for this nice idea!
Kind regards,
Darius

[Merlin]

How does it work anyway? I assume it's due to some flyback effect, but I still don't understand what flyback actually is!

[ppppenguin]

Any inductor will try to oppose a change in current flowing through it. If you suddenly force the current to zero, for example by opening the switch, the stored energy in the magnetic field will induce a voltage that will attempt to make the curent flow again. This voltage can be very high, limited largely by stray capacitance. This is why you get a spark at the contacts when you switch off an inductive load. Also why you need a diode across a relay coil driven by a transistor. The high voltage will easily light a neon.

This is sometimes called "flyback" because it's seen in in TV sets during flyback when the current in the scan coils has to change rapidly.

A transformer winding is simply an inductor. Unless there are shorted turns or a load on another winding which will severely reduce the inductance.

In mathematical terms, for an inductor:

V = L * dI/dt

In simple terms: The voltage across an inductor is equal to the inductance multiplied by the rate of change of current.

[See_Mos]

I use the same method as Darius, but it's not portable.

I also sometimes put a 250 or 500 watt halogen lamp in series when testing primaries or even repairing switch mode supplies.

They make a great dynamic current limiter. If the current is low the element stays cool and therefore the resistance is low so there is little voltage drop. If there is a problem the lamp lights and there is no damage to the unit under test.

It's also a great replacement for a fuse if you are having problems tracing the cause of persistant fuse blowing but but cannot find any obvious shorts.

S.M

[oldeurope]

Good morning,
be carefull with switched mode supplies, they often don't like undervoltage!!!
500W means more than 2A and a high rush current. A 15W lamp should not light up with a 75VA mains transformer, open secondaries of course.

Darius

[See_Mos]

Yep, you are quite right.

I replace all the 'lytics under 47uf and use an ESR tester on the rest. The one published in Elektor a few years back has been worth it's weight in gold!

I then put the lamp in series with the primary supply with a small load on the output to keep them happy. The resistance of the fillament stays low unless there is a problem and helps to prevent the switching transistor failing if there is.

It's not a 100% safeguard but then what is with SMPS

SM

[bob_rich]

Hi Folks.

I have tested numerous transformers for a shorted turn (or other maladies) by simply feeding mains power to the mains winding via a 100W to 200W tungsten filament bulb in series with the supply to the transformer primary. A cold bulb has a very low resistance so the transformer magnetising and loss current will not usually light the bulb and increase its resistance so a dim bulb suggest the transformer may be OK. With a totally shorted turn the impedance of the transformer under test will be low and the bulb will light up --thus liming the current . If the bulb does not light then the transformer should be OK and its ratio can be checked by using a voltmeter on the primary and secondary winding. For very small transformers use a lower power bulbs.

Idea can work on audio output transformers in this case feed the speaker winding from a low voltage transformer say 12V rms with a car stop light bulb as the series limiter (21W/6W) most audio transformers are usually OK at 50Hz.

The only problem I find with this idea is getting filament bulbs!!! not sure what would happen with low energy ones!

hope this helps

Bob

[Boom]

I'm not sure of the exact procedure as I haven't done it for years but you can test a transformer for shorted turns simply by using an Avo on a low Ohms scale.

If the transformer has good windings the inductance will make the meter slowly climb over. If it has shorted turns the meter will go over to FSD much more quickly.

Ed Dinning will probably be able to say if I have this correct.

Dave

[Station X]

Two off topic posts deleted.

[cannissolis]

I also use variac + ammeter and also listen for unnatural strained humming/vibration and rapid heating.

[Wendymott]

Why has no one suggested the "Ringing" Method ? With TV line output transformers, yes I know they work at 16Khz, but the theory should still work. Take a capacitor (say 1uf) for experimental purposes. from the "Sweep Sawtooth output" from a scope to the "Top end" of the primary of the transformer, also attach the scope probe to this point. The ground of the scope to the other end of the primary winding. Use a GOOD known transformer to set the parameters. Set the sweep to about 2ms/cm, and adjust the display for a pulse on the left of screen and there should be a decaying sinewave, displayed. If you short any secondary the ringing will be severely damped. Now transfer this to the suspect transformer. If I have time tomorrow I will try this method and post the scope settings and cap value.
Regards
Wendy Mott G8BZY

[Eddy_Stone]

I would be very careful using this method. If you hold the momentary switch on for more than a moment you could burn out the transformer winding. The max duration of the moment is dependent on the winding resistance, wire gauge and how stiff the battery voltage is , one second could be enough to burn out the winding . I would use a sig gen and voltmeter instead.

Eddy