Inductance Problems

[cwarner]

I have come across a small transformer that was surprisingly sensitive to frequency, and there is nothing in my reference books to explain this. It was found in a camping lantern, and was used to step up rectified 12 Volts DC to something like 70 Volts AC for operating flourescent tubes. At first I thought there was something wrong with it, because I could not get the expected ouput voltage, although my LCR meter indicated that it was good. I accidentially discovered that the operating frequency was soemthing like 125 kHz; when I input a signal at this frequency, I was able to get the output ratios I expected. The output ratio fell off rapidly as the frequency varied around the central frequency (to about 50% at +/- 25 kHz from the central frequency). This transformer was wound on a ferrite core. Now, I know that the impedance of the coils is a function of frequency, but the ratio of input to output is supposed to be strictly a function of turns ratio. Why do I find that this transformer is so sensitive to frequency?
Transformers and inductors never seem to have any sort of identifying marks or ratings that can be used to determine if they are functioning as intended. I have tried a number of methods to quantify inductors and transformers, but I get inconsistent results. I use an LCR meter now, but when I compare it to my old method of testing in a resonant circuit, I get results that vary considerably between the test methods (even when I try to set the resonant circuit to the same frequency as the LCR meter). Also, I have built a Hayes bridge, using a decade box for variable capacitance, and this method also gives significantly divergent results when compared to the other two methods. What am I missing about measuring inductors?

cw

 

[Warpspeed]

Difficult to say really, but there could be many factors at work here.

The first thing is that the magnetic coupling between primary and secondary may be deliberately very loose. This is going to give very high leakage inductance, so things like turns ratio really have no meaning.

Another factor may be deliberate resonance of one or other of the windings due to the combination of inductance and internal capacitance between turns. This is going to foil any attempt at measuring the inductance at any frequency close to resonance.

Also some of these transformers that look like transformers are actually a pair of coupled chokes that run in flyback mode. These almost always have an air gap in the core and operate quite differently to a conventional transformer.

No doubt someone has gone to a lot of trouble to design an electromagnetic device which may not actually function how you might normally expect a conventional transformer to function. It probably operates with waveforms that are far from symmetrical sine-waves as well.

 

[STUONTIPS]

All transformers are designed to work over a certain frequency range their output is not just a function of turns ratio other factors will determine the frequency range of the device, core material, air gap, winding style are just three. The device that you have is likely to form part of an oscillator/switched mode PSU and is designed to work over a range in this case 125Khz, These things very rarely go wrong and when they do its normally easy to see, if you have continuity on both primary and secondary then its most likely OK,there should be a large Transistor check this out first followed by the elecrolytic caps in the cct,if you have an output but its low check out the smaller non electrolytics

 

[BrianG]

In additions to the excellent comments already made by Warpspeed and Stuontips, a further point to note is that these sort of ferrite materials have a huge range of permeability characteristics. Often transformers for such inverter systems are designed to allow a large d.c. current component in the waveform. When testing the transformer on the bench this has to be taken into consideration, and often simulated with a d.c. power supply, to get the correct inductance reading.

 

[buzzp]

If the input is DC then it is not transformer and is likely what warpspeed has suggested. So typical transformer tests will not produce any usable results.

 

[BrianG]

Buzzp: please read posts properly before commenting so dismissively. Clearly transformers cannot work with d.c. - what I said was it may have a d.c.component in the waveform - in other words a bias current or standing current along with the switching waveform.

 

[Warpspeed]

BrianG is quite right. I have spent many years designing switch-mode power supplies for mass produced commercial equipment.

Consider the forward boost topology, that uses a choke and a rectifier diode to convert a low d.c. voltage, to a higher d.c. voltage of the same polarity. This choke has a constant d.c. current with a usually small triangular ripple component superimposed.

This "choke" may also have tertiary windings to generate additional isolated supply voltages of any voltage, depending on turns ratio.

Now this is not a transformer, it will have an air gap, and d.c. current flowing through the primary that never reverses. But it may have several windings and look like a transformer..........

There are some queer beasts lurking around out there disguised as transformers, so beware.

A tesla coil is pretty strange. And so is a constant current neon sign transformer. Or how about a constant voltage ferro-resonant transformer ?

Any attempt to analyze any of these as a conventional transformer will defeat you.

 

[cwarner]

Thanks all for the response. The transformer was used in a power supply for a battery powered flourescent lantern. There was a large transistor that originally tested good, but subsequently failed- I assume the original problem was that it was weakened. The new transistor output a high-frequency square wave, as was suggested. Now, where does one go to find theoretical information that can help one understand this magical thing called inductance? Searching the Net results in an infinite repetition of the same, mostly useless information...

 

[buzzp]

BrianG,
Geez, mellow out and read the posts yourself. My comment was not for you it was for the original poster.
He stated it takes a DC in and AC out. I was merely pointing out that it is not a transformer in the traditional sense. If I am directing comments to someone other than the original poster, I will identify that person as I have with this post. So in your case, "please read posts properly before commenting so dismissively."