How to properly measure OT inductance??

[Alan0354]

Transformers are huge distortion machines. The permeability is small for small signals, increases by a factor of about 2-5 for medium-sized ones and decreases again towards saturation.

The distortion isn't as bad as you'd expect because the reactance of the magnetising inductance is large compared to the load impedance over most of the audio band. Things get interesting in the bottom octave, and that's where much of "transformer sound" comes from.

If you measure one winding with another loaded, you see a combination of the leakage inductance, magnetising inductance and reflected load impedance. Maybe your LCR meter can help you tease those apart.

The classic way to test a transformer is: First an open-circuit test at rated voltage and frequency to measure the magnetising inductance and core loss resistance. In an audio transformer this would be the lowest frequency in its range at its rated output power.

Second a short-circuit test at rated current (and whatever voltage it takes to drive that current) This gives you the leakage inductance and winding resistance (copper losses). For an audio transformer this might be done at a quite high frequency.

A multi-winding transformer can have a different leakage inductance for every coupled pair of windings, so you would repeat the short-circuit test n-1 times. The open-circuit test only needs done once, because there is only one magnetising inductance, it being a property of the core. The different windings just reflect it according to (the square of) their number of turns.

Actually thinking about what you said about u changing with power, Actually the formulas hold. All you are doing is changing the Lo and Ls. You can still find the value from experiment and plug into the formula and predict the Low and high frequency cut off.

[Alan0354]

I apologize if I come out this way, I just got too use to Physics Forum, we proof everything, everything has to be explained, we don't say "it just is". We site books and challenge them. Through debate, people learn.

Case in point, I think in this case, the inductance change due to u change with the loading. So this is not like the OT change which invalidate the equations. You separate and compartment the problem into the change of u and the basic transformer theory. So now you can concentrate on finding the u vs different loading. Then you can easily plug the u into the formulas to get the OT low and high cutoff characteristics.

I don't know how much you guys characterize OT, but if we can agree the u is the key, then you do open and short circuit at different drive to find inductance like in this thread, then back calculate the u or just use the inductance found to do calculation of the OT. Point is, there is a will, there is a way to explain and simplify the problem.

[R.G.]

Think about inequality "equations".

[Gregg]

I also have a DMM that measures inductance (100Hz/1V) and from my experience I can tell you that the readings I get are within 5 to 10% of those derived from the formulas and calculations.
Sometimes you'll see in datasheets huge inductances listed but most probably these are measured at AC line level. The small signal measurement will give you the values close to the calculation. A 100W OT for example will usually measure 3-5H and that's what you'll get from the formulas at -3dB (50-70Hz).
Measuring chokes under load is more complicated. Steve Bench has a very good article with schematics on his site:

Steve's Tube Pages

[Bruce / Mission Amps]

What does any of this nit picking have to do with making music?
I guess for some, that part isn't very important or escapes them in terms of why we do what we do.

Stick a 16 ohm 50 watt resistor on the secondary. Plug the outside leads of the primary (not the center tap) into a 75vac to 100vac source, like a variac at 50Hz to 60Hz, and read the secondary AC voltage, then do your math over again.
I will be surprised if that OT isn't with 5%-10% of dead on what you were looking for to be and you can use a $15.00 DMM for this.

[Alan0354]

This is Theory section, of cause I expect to be nit picking here!!! You do nothing like this in the other section of this forum. That's the reason why I hang around here and seldom venture out.

There are some big brains here, If I don't nit pick them, where?

[Jazz P Bass]

Maybe the RCA book is incorrect.
It happens.
Hah.
I found gross errors in the Boyscout electronics book.
That was, oh, 4 years ago.
Contacted numerous individuals.
They are still printing it with the errors.

[jpfamps]

But a lot of time, it is true, I do find mistakes, you blame me in challenging the articles or books. You study something and you really take all at face value? If you don't challenge, how are you going to learn?

I think you should re-read my last post properly.

I never "blame" you for finding errors in books.

What I was pointing out was that many of your posts contain quite lengthy expositions about your academic credentials that are completely irrelevant to the subject of the thread, and could be regarded as disrespectful to the many well-qualified contributors to this forum who have taken the trouble to answer your questions.

To highlight further my previous point, I used one example from a post you made in the this thread about "correcting" a book; very laudable I'm sure, but totally irrelevant to any content in the thread. I certainly wasn't criticizing you for "correcting" a published book, but was questioning why you feel the need to include this information in your posts.

I could have chosen several other examples where you spuriously mention your patent applications, how you design SMPS, your experiences with Laplace Transforms et cetera (by the way anyone who understands the maths behind the Laplace Transform shouldn't need to seek advice on an internet forum regardling the validity of the maths in RHD4). Indeed a significant proportion of your posts are about youself, rather than the topic of the thread.

In light of this I have to question your motivation for engaging in this forum; you claim you are keen to learn, but then challenge most of the information you are given; in fact you seem more interested in informing us how much you know.

And finally, whilst I'm sure that your advice on "how to learn" is well-intentioned, I'm fairly confident in my understanding of the process of scientific research and publication.....................

[Alan0354]

I apologize to everyone and I'll watch out.

[Bruce / Mission Amps]

This is Theory section, of cause I expect to be nit picking here!!! You do nothing like this in the other section of this forum. That's the reason why I hang around here and seldom venture out.

There are some big brains here, If I don't nit pick them, where?

Oh yeah I forgot... I'm a real musician and actually get paid to play out, to build and work on amps too.

[Wombaticus]

Thank you Steve, RG, etc. for your thoughtful replies, they are much appreciated. On the other hand, those of you more interested squabbling about who's got the bigger dick, can you please stop crapping up the S:N ratio of my thread? Pretty please?

I bought this LCR meter hoping that I'd learn something about transformers by satisfying my curiosity and taking some actual measurements, maybe asking a question or two. I see now that I've succeeded in learning a few useful things, just not quite in the way I had anticipated. RG's points about learning the limitations and proper use of the test equipment are well taken.

Let me try and restate the basic issue to see if I've more-or-less understood, at least in lay terms.

First, the approach I took of directly measuring the winding inductances of an OT was problematic largely because of the nature of the core material. If I understand correctly, the permeability (mu) of the iron core material is such that it does not respond uniformly (i.e. is non-linear) in terms of the magnetic flux density that will be elicited in response to a given magnetic field intensity. Moreover, the magnetic field intensity depends in turn on the electrical current through the windings of the transformer. So with an inductance test involving small electrical currents, relatively less magnetic flux is elicited per unit of magnetic field intensity than would have been the case had the test involved currents more typical of those seen during working conditions. Am I correct that if I look at a B-H hysteresis curve, the non-linearity at work here corresponds to the lower left-hand side of the graph, whereas the non-linearity at core saturation corresponds to the upper right-hand side (incidentally, it's remarkable that these two regions should look so symmetrical, which must be telling us something fundamental, I guess)?


I have a couple more questions, the answers to which I think will clear up a couple of further points for me.

First, because I can't seemingly measure iron core inductors accurately (see my results even with the 40mA 9H choke), I have an unresolved doubt about the calibration of the meter. What's a good way to get a feel for if thing is at least calibrated reasonably.

Second, when I was trying to do the leakage inductance test by shorting the secondaries, I got a different result if I soldered together just the common/16ohm wires than if I soldered together all three of the common/8ohm/16ohm. I consistently a *higher* number (~85mH) if I solder just the common/16ohm than I do if I tie together the common/8R/16R (more like 36mH). The question is, even given everything mentioned here before, why would that be?

Third, after leaving the meter measuring the primaries all night, then trying to measure leakage by soldering together the secondaries, I observed that the reading dropped WAY less then expected.. and once again, putting together all three leads lead to the reading barely changing relative to the primary reading with secondaries open... It's like something was getting charged. Again why did this happen? It's reading back to normal small numbers now after leaving the meter unplugged all day.

Cheers,

Paul

[Enzo]

As to the latter, since the meter has to send a small current through the device under test, perhaps over time your meter battery slowly warmed up and dropped a small amount of voltage, affecting readings. Then disconnecting it allowed the meter battery to recover and so on. Just a thought, few phenomena occur in isolation from the rest of the world.

[Steve Conner]

Hi Paul

To check the meter calibration, try measuring some air-cored inductors. Crossover coils from Parts Express or the like.

Like I said before, every winding combination has its own leakage inductance. By shorting multiple secondary wires together, you're paralleling up the leakage inductances, resulting in a lower total leakage inductance.

Finally:

Am I correct that if I look at a B-H hysteresis curve, the non-linearity at work here corresponds to the lower left-hand side of the graph, whereas the non-linearity at core saturation corresponds to the upper right-hand side (incidentally, it's remarkable that these two regions should look so symmetrical, which must be telling us something fundamental, I guess)?

No, it corresponds to the middle of the graph near the origin, the "initial magnetisation curve" as shown here: Hysteresis - Wikipedia, the free encyclopedia The top and bottom are both saturation, just with opposite polarities of flux.

You can also see from this graph that the material has a memory of how it was last magnetised. The slope of the B-H curve (which is permeability, hence inductance) has different values depending if the material is starting from an unmagnetised condition, or if it was previously saturated. However, in good transformer iron the difference should be quite small.

I am puzzled by the changing-overnight results from your meter. I know that when trying to measure the DC resistance of a large transformer winding, the reading can take tens of seconds to stabilise, because of the RL time constant. But the test signal from your meter should be AC, so this shouldn't apply.

[Wombaticus]

Hi Paul

To check the meter calibration, try measuring some air-cored inductors. Crossover coils from Parts Express or the like.

Like I said before, every winding combination has its own leakage inductance. By shorting multiple secondary wires together, you're paralleling up the leakage inductances, resulting in a lower total leakage inductance.

Finally:


No, it corresponds to the middle of the graph near the origin, the "initial magnetisation curve" as shown here: Hysteresis - Wikipedia, the free encyclopedia The top and bottom are both saturation, just with opposite polarities of flux.

You can also see from this graph that the material has a memory of how it was last magnetised. The slope of the B-H curve (which is permeability, hence inductance) has different values depending if the material is starting from an unmagnetised condition, or if it was previously saturated. However, in good transformer iron the difference should be quite small.

I am puzzled by the changing-overnight results from your meter. I know that when trying to measure the DC resistance of a large transformer winding, the reading can take tens of seconds to stabilise, because of the RL time constant. But the test signal from your meter should be AC, so this shouldn't apply.

Steve, this is exceedingly helpful, thank you! The overnight results mystery surrounding the leakge inductance is resolved too, it turns out to be a matter of whether parallel mode or serial mode was used on the meter. The big inductors seem to do better with parallel mode, so I measured overnight this way. But it turns out that parallel mode gives nonsense results with the small leakage inductances. It's necessary to use serial mode for those readings.

Finally, I tried some measurements with a higher-quality transformer with significantly different properties and core material. I tried a Sowter UK 1010A 15K:600ohm line output transformer. I believe the alloy in this core has a significant amount of nickle and mumetal. I'll post more details later, but suffice it to say that inductance measurements of primary and secondary came much closer to predicting the correct 25:1 inductance ratio. Impedance measurements did slightly better, if I recall, predicting a 23.4:1 ratio.

[R.G.]

OK, stop that.

If you learn all the tricks about your meter, there won't be any mystery left...