Audio Transformer Inductance

[abbey road d enfer]

Is it battery-powered?

Yes, nothing fancy but it works http://gsmserver.com/shop/equipment/measuring_equipment/multimeter_mastech/digital_multimeter_mastech_my64.php

The limited frequency range may be a problem though. You may not properly see the resonance peak, which may be a problem (or not).

In the "almost perfect" region, the impedance Z is equal to sqrt[Rdc²+(2pi.F.L)]. At LF, the DCR is dominant, but as frequency increases, the impedance gets much higher than DCR, then you can use L=Z/2pi.F as a very good approximation.
You should draw a graph of Z vs. frequency, with a log scales.
You can see clearly the DCR-controlled part (parallel to the X-axis), the L.omega part (6dB/octave) and the resonance peak.

This part I still can't figure out. maybe I have a problem to visualize the connections and therms. I still don't understand what is Z in this situation :/ any drawing please ?

Pls see attached.
You have to experiment with the value of R. You must make sure that at any frequency you wish to measure the inductor, R is at least 10x larger than the impedance.
Starting with a rough estimate, let's say 1mH, at 20kHz Z is 125 ohms, so you can use 2k, which will give you a measured voltage of .6V at 20kHz, about 30mV at 1kHz, and probably a few millivolts at 20Hz, depending on the DCR of the inductor.

 

[marcus4audio]

Pls see attached.
You have to experiment with the value of R. You must make sure that at any frequency you wish to measure the inductor, R is at least 10x larger than the impedance.
Starting with a rough estimate, let's say 1mH, at 20kHz Z is 125 ohms, so you can use 2k, which will give you a measured voltage of .6V at 20kHz, about 30mV at 1kHz, and probably a few millivolts at 20Hz, depending on the DCR of the inductor.

Thanks for drawing. So, if I understand the pic  I have to change the Z at certain frequency for max voltage reading at the inductor ?(or in this case primary of the transformer)

 

[marcus4audio]

Pls see attached.
You have to experiment with the value of R. You must make sure that at any frequency you wish to measure the inductor, R is at least 10x larger than the impedance.
Starting with a rough estimate, let's say 1mH, at 20kHz Z is 125 ohms, so you can use 2k, which will give you a measured voltage of .6V at 20kHz, about 30mV at 1kHz, and probably a few millivolts at 20Hz, depending on the DCR of the inductor.

Thanks for drawing. So, if I understand the pic  I have to change the Z at certain frequency for max voltage reading at the inductor ?(or in this case primary of the transformer)
edit: sorry I'm totally confused how to measure

 

[CJ]

Calculate current sensing resistor from R=V/I * 1/100.

if you apply 10 volts and the inductor chews 1 ma to get there, then you have
10/.001=10,000 ohms x 1/100 = 100 ohms for R1.

if you need 0.1 ma to get there as f rises, , than R1 becomes 10/.0001 = 100,000 x 1/100 = 1k.

the voltage applied to your inductor should be very large compared to the sense resistor,
we want a voltage source such that V-inductor>>V-sense resistor,

this is why i like to use the ammeter, R-sense will be the value of the shunt inside the meter, which is usually pretty low, somewhere along the lines of 0.01 to 0.1 ohms, so that if we have 10 volts at 1 ma, we have 0.001 A x .1 ohms = 0.0001 volts across the ammeter, so our sense resistor is 1000 times better than if we used the volt meter,

we need to find out what flux level to test at,

to know that, we need the number of turns and on what size and material of core,

since you probably have not hacked into the turns count, we can guess if you tell us what application this xfmr is intended for, such as mic in, line in, or tube output,

then we can calculate how many volts to apply and take it from there.


meanwhile, it turns out they do measure core loss at high frequencies, here is a graph of 50/50 Ni  from mag Met,

notice at 100 k that there is a lot of core loss, like 1000 watts per pound, so for a 1/4 pounder with cheese, we have 250 watts core loss,
looking at my dB chart, we see that this means that any mic or line signal will have no chance of exciting the core at 100 k hz,

at 60 hz, loss is 100 milli-watts per pound, so for a double cheezeburger with cheese, we will have about 25 milli-vanelli watts core  loss for 50/50 at 10 k-gauss,  this seems more reasonable, we could get it way down by dropping the flux, this would lower insertion loss for a moving coil cart, we could get it down to 1 milliwatt/lb. at 0.8 kg, = 0.25 milli-watts for a whopper with cheese,

see that the y axis is plotted on log paper,

and the x axis too,

 

[emrr]

Observation related to testing level.

I looked at a pair of UTC LS-34 and an LS-666 yesterday.  This is no load testing at 120Hz, handheld L meter.  Both transformers are rated for high power applications, well beyond what a handheld meter can possibly deliver.  Note the low L readings that result. 

LS-34
500->50 : 500->50 : 30->1.2, 10Hz-40kHzm, 40W max rating
3H9-4H3 range seen across (4) 500 ohm windings on two transformers

LS-666
500:8, 7Hz-50kHz, 50W max rating
500 ohm 5H1

 

[CJ]

some of those UTC LS series have super hi-perm L lams,

Tamura also uses an alloy in some of their stuff that gives higher Hernies than most other transformers, these hi-perm lams look like they are chrome plated,

you can put a signal generator across the 600 ohm wind and measure current from maybe 15 Hz to 100 Hz to get  a nice idea of what max inductance can be,

as frequency goes down, flux goes up so you get higher perm values as you make the B-H curve larger, as it gets larger, it changes shape, usually showing less core loss, which is the area inside the loop,

 

[bockaudio]

every meter i have used has given different readings for inductance,

but if you use the same meter, you get a idea of how different iron compares to each other,

the gen rad bridge allows for a signal generator input, so you can measure inductance at any frequency, and also adjust the signal level,

most hand held bridges measure at 1 k hz, so the bottom end inductance is not well represented,

my B&K LCR does 100Hz, thankfully.

 

[CJ]

what is the model # on that BK LCR, thanks!

 

[bockaudio]

for example, the LA2a reissue used a different winding geometry for the ha 100 x input copy,
instead of 2 pies per coil, they wound one pie.
much easier to wind this way, UTC had the dual winding thing down,
i do not know of anybody doing the side by side sections anymore,
so the old UTC HA100x has less leakage, which gives it a different curve,

Which is why my experienced clients had me install old xfmrs in their new units.
There was also something funky in the psu IIR.

 

[usekgb]

Is this the right place to add this?

I just found 3 UTC A-21 Transformers.  The seller only wanted $5 a piece for them, so I naturally bought them.  After searching through here, I couldn't find detailed specs for the A-21, so I took some basic measurements.  I measured all three transformers, and took the average to get the numbers I am posting here:

UTC A-21:

Stated on Transformer:
Primary:          Secondary:
1-2:  200R      4-5:  200R
1-3:  500R      4-6:  500R
2-3:  50R        5-6:  50R


Measured DC Resistance:
Primary:          Secondary:
1-2:  19.0R      4-5:  27.3R
1-3:  30.4R      4-6:  43.3R
2-3:  12.0R      5-6:  16.6R


Measured Inductance:
Primary:          Secondary:
1-2:  5.62H      4-5:  5.64H
1-3:  14.20H    4-6:  14.21H
2-3:  1.52H      5-6:  1.52H


I really don't have a good way to measure impedance curves, and they vary with load anyway, so I didn't do them.  I hope this helps.

Now, what to do with these buggers.  I would guess that they could be used as an output transformer for something or other, but I don't have anything solid in mind.  Anyone have some good suggestions?  I'm always up for adding more things to my to-do list. ;D

Cheers,
Zach

 

[ricardo]

I couldn't find detailed specs for the A-21, so I took some basic measurements.

How did you measure the Inductance?  Instrument?  Method?

 

[usekgb]

With my multi-meter.  I forget the brand.  It's usually pretty close.
The transformers didn't have anything connected to them.  I just connected the leads to my multimeter.

 

[ricardo]

With my multi-meter.  I forget the brand.  It's usually pretty close.

Measured Inductance of Iron Cored chokes & transformers depend on

• frequency
• level

So EXACT method is important.  There's loadsa evidence of this in this and other threads.

 

[joaquins]

Do you know the frequency the multimeter use? really important, change a lot, we usually interested in the inductance at the low end, where it affects most, which may be more than twice than the inductance at 1kHz. You don't need anything fancy to measure at different freq, just a signal generator, which may be your computer, a driver so the output is not overload which may be a headphone amp, or the headphone output of your sound card or mixer. Then one resistor which value you know, even better a potentiometer. You play the 20Hz freq, measure the voltage across the winding and the resistor, and then some math... XL=I/V=wL being the voltage (V) the voltage measured across the winding, the current (I) the voltage measured across the resistor divided by the resistance of this resistor, and then, w 2*pi*f so 125.6638... for 20Hz let's say 126. So, L=Vres/(Rres*126*Vwind). I think I should make a plugin to do this with the sound card easily... for L and C, never thought about that, but may be useful for many people here... I will when I get some time. Even complex impedance could be done that way, because we usually don't know the actual inductance and core/wire loss at different frequencies. Because with the method I'm describing you don't know that and may be way off, usually not so much for low freq, but a lot at high freq. maybe some around, I've never one.

JS

 

[usekgb]

Ok.  I'll see if I can throw a test rig together tomorrow morning.

 

[lassoharp]

After searching through here, I couldn't find detailed specs for the A-21, so I took some basic measurements. 

They are here and in most of the UTC catalogs out there

http://groupdiy.com/index.php?topic=44876.0

I would guess that they could be used as an output transformer for something or other

Or an input . . .. . cheaper than an O-12   

 

[usekgb]

After searching through here, I couldn't find detailed specs for the A-21, so I took some basic measurements. 

They are here and in most of the UTC catalogs out there

http://groupdiy.com/index.php?topic=44876.0

I searched and searched, but did not find this page.  Thanks!

I would guess that they could be used as an output transformer for something or other

Or an input . . .. . cheaper than an O-12   
[/quote]

I was considering the same thing after looking a little harder.  I've never used UTC's before, but I couldn't turn them down for $5 each!

 

[joaquins]

Is it battery-powered?

Yes, nothing fancy but it works http://gsmserver.com/shop/equipment/measuring_equipment/multimeter_mastech/digital_multimeter_mastech_my64.php

The limited frequency range may be a problem though. You may not properly see the resonance peak, which may be a problem (or not).

In the "almost perfect" region, the impedance Z is equal to sqrt[Rdc²+(2pi.F.L)]. At LF, the DCR is dominant, but as frequency increases, the impedance gets much higher than DCR, then you can use L=Z/2pi.F as a very good approximation.
You should draw a graph of Z vs. frequency, with a log scales.
You can see clearly the DCR-controlled part (parallel to the X-axis), the L.omega part (6dB/octave) and the resonance peak.

This part I still can't figure out. maybe I have a problem to visualize the connections and therms. I still don't understand what is Z in this situation :/ any drawing please ?

Pls see attached.
You have to experiment with the value of R. You must make sure that at any frequency you wish to measure the inductor, R is at least 10x larger than the impedance.
Starting with a rough estimate, let's say 1mH, at 20kHz Z is 125 ohms, so you can use 2k, which will give you a measured voltage of .6V at 20kHz, about 30mV at 1kHz, and probably a few millivolts at 20Hz, depending on the DCR of the inductor.

just in case... Other interesting thing to do is run a transfer curve with the same setup, one input the resistor, other the winding, smart for example may be useful, better with a linear amplitude trace I guess but I don't know if you can do this with smart... Look at 20Hz as your number but you can upload the pic here so we all know, then you'll have mag and phase between current and amplitude, and the relation between them, which together with the value of the resistor you have the wanted impedance as a function of frequency, and you can get the inductance at any freq you are intrested.

JS

 

[ruffrecords]

I have just updated the transformer inductance document. I corrected a few typos and added some further tests of my own on the OEP Z3003E that I completed a couple of days ago. For the price (£13 at Canfield) this is an excellent 10K:10K bridging transformer that I would recommend for the budget conscious DIYer. I also added the test I did on the Edcor XSM 2.4K/600 which I now recommend as a lower cost alternative to the already low cost Carnhill VTB2291.

You will find the old and new version of the document in the iron folder in the DIY tab of my web site:

http://www.customtubeconsoles.com/diy

Cheers

Ian

 

[trobbins]

There was some discussion in this thread about how winding shunt capacitance would also be measured. 

I recently noted Pat Turner's comments that deducing shunt capacitance from resonance frequency and leakage inductance may have significant error.  Pat proposed that lumped shunt capacitance is best measured using an RC LPF -3dB frequency technique, where R is ~5x nominal winding impedance and is inserted in series with the signal generator driving the winding, with other windings open and connected to ground to suit the normal operating condition.

Subtract measurement voltmeter or oscilloscope probe capacitance off the measured capacitance.