Transformer reversing... CJ?

[SSLtech]

So...

Let's say I wanted to reverse a transformer... let's say it looks like a DI transformer... that kinda size...

I measure the primary and secondary resistance, I put 100mV RMS at a very low source impedance into it, and I measure what comes out. I measure the inductance of the primary but the secndary reads off the scale on my tester, so I can't directly read it...

What else do we need to know?

(this is a potted, shielded transformer from an interesting piece of gear that is on its way out the door, back to the client, and has to remain essentially un-molested...)

Keith

 

[gyraf]

From the primary inductance and the turns ratio, you can easily calculate secondary inductance.

Shorting the secondary and measuring primary inductance should show the leakage inductance (I'm not sure about this, haven't tried that)

Jakob E.

 

[NewYorkDave]

I've posted this link before, but in case you didn't see it... I scanned these old Norman Crowhurst articles that include some information about transformer testing
PDF

 

[SSLtech]

Primary:... ... ... ...Secondary:
100mV -voltage ratio- 775mV
530Ω... -resistance- ...1400Ω
9.36H ...-inductance- ...????

..This is from a piece of equipment that keeps coming up, used when returning signals to an amplifier after recording... please... let's not mention the product name... search engines and all!!!

Keith

 

[Steffen]

so ratio is 1:8?

frequency response -1dB@20 Hz when driven from a 600 ohm source

What else do we need to know?

does it sound good?

steff

 

[PRR]

> What else do we need to know?

Distortion: level, frequency.

Capacitance is nice.

A "perfect" clone would require far more data than you can conveniently measure.

 

[SSLtech]

The total component count is as follows:

1x XLR/TRS(combo) input socket
1x ¼" output socket
1x SPST switch (ground lift)
1x log potentiometer (with knob)
6x resistors (1% metal film)
1x capacitor
1x transformer
1x circuit board
1x case

We can assume the target frequency range to be 20Hz-20kHz, though even if it falls slightly short at the top end, that's probably acceptable... bear in mind the use to which it's put.

Keith

 

[SonsOfThunder]

Keith,

I don't understand why you need to de-eng. the xformer, since that is stated in the aforementioned patent articles. Is there some big diff in this new xformer? If you only want to build one or two, what would be wrong with the original ckt? Have I missed some posts?

Gosh, there is so much cool info going here, I can barely keep up!! Thanks guys for sharing the knowledge with those of us who tend to lurk. I promise not to sell any of this stuff on evilbay, lest the "sismofyt police" come after me :shock: :razz:

As someone said the other day, "If only I didn't have to go to work!"

Peace!
Charlie

 

[Kev]

[quote author="SSLtech"]... This is from a piece of equipment that keeps coming up, ... [/quote]

:shock:
I'd love to re-ply with an answer BUT I've never actually used one of these before.

A friend on mine is trying to import these things and from the little he will tell me, I feel that his buy price is way too high for what the components in the product must be.
This is just begging for a Group D ... sorry a LAB version ... I think I just thought of a name for it.

 

[peter purpose]

[/quote]

This is just begging for a Group D ... sorry a LAB version ... I think I just thought of a name for it.[/quote]

The 'LAB1A'.... :green:

 

[CJ]

OK, sec. inductance should be about 600 henries.
Inductance changes as does frequency, how it does is good to know.

Also, self resonance is good to know.
You can kind of estimate capacitance from the self res curve.

Leakage like Jalob said can be measured by shorting the pri or sec and checking inductance of the opposite winding with a bridge.

Distortion should probably measured in the circuit. Use your ears if you don't have a distortion analyzer.
Most distortion will occur at the lower frequencies.

Phase shift is also good to know. I usually write down where the signal gets reversed 180 degrees in case I want to compare a DIY transformer with a real one that I am cloning.

High end drop off point is good to know.

Knowing the winding struture is a must if you want an exact duplicate.

The typwe and amount of insulation is nice to know.
The distance from layer to layer is important, and this is set up by the insulation and wire size.

Wire size is nice to know. From the dcr, you can estimate the length of each winding.

Lam dimensions: This determines the magnetic path, which can effect the overall sound.

Core structure: Gapped or ungapped, Lams grouped at all?

Core material: Usually M6, 49 or 80 Nickel. Sometimes Cobalt. Thickness of the lams is usually always around 14 mil. Grain or non grain lams-a lot of the real old transformers used non grain Nickel, which I don't believe is made anymore.

Shielding: The number of cans of mu metal can effect the frequency response, particulary at resonance. Noise level will vary also.

I am looking into a method for getting the B-H curve of the core with the scope. Don't know how yet.

There are other things I am forgetting, but the above will tell most of the story.

cj
:guinness: