xformer DCR questions

First, I was wondering if it's possible to ballpark transformer ratio with DCR figures the same way you would calculate it with impedance ratios. I've tried it with a couple of transformers where I know the turns ratio, & it seems to get relatively close, but I don't know if this would work with transformers in general.

Second, I know that you can often guess the impedance of a winding if you know its DCR (eg, 600 ohm windings tend to measure 30-40 ohms.) Is this a pretty consistent thing, or does it vary with other attributes of the transformer? Would a 600 ohm winding on a smallish line input xformer measure about the same as a chunky 600 ohm output transformer?

Tom

There's no hard and fast relationship between d.c.r. and turns ratio, unless of course you know the gauge of the wire for each winding. And if the gauge is the same for each winding that's the easiest of all, but even there the details of the construction will have a small effect.

The only reliable means without getting access to the innards is to look at the unloaded voltage stepup or stepdown ratio---drive a winding with a low Z generator somewhere at a midband frequency and look at the voltage on the others with no more than the voltmeter as a load. Then know that the impedances go as the square of the stepup ratio.

Certainly if you measure a d.c.r. that is getting significant compared to the considered impedance, it's a pretty good indication that the design target impedance is higher than the guess.

If you can get into the beast and introduce a turn, or a few of them, passing through the core, you can measure its inductance with a suitable meter and then deduce what the number of turns on the other windings are by measuring their inductance and knowing that the inductance goes as the square of the number of turns. This can give you an acutal number of turns per winding, more information than you usually need unless you plan on duplicating the transformer.

If a winding's DCR is more than 10% of the load, loss will be bad.

If it is less than 10% of the load, cost will be high for no good reason.

So to some degree, you can take audio impedance as about 5X to 20X the total DCR. Anything else is unlikely on cost/performance grounds.

That's for hi-fi audio. Voice-only transformers may have very low inductance (hardly enough to support the rated impedance), few turns, wire size a bit big because it reduces breakage, etc, and the DCR may be less than 10% of the nominal impedance.

Transformers with DC in them (SET outputs) generally can't fit enough turns around their poor over-soaked iron to keep losses low. DCR may be 20% of rated impedance.

Very high impedance windings are unpredictable. Many-many turns implies very small wire and high DCR. But you may not want to work with such fine wire, and use fatter wire (lower DCR) for production reasons. OTOH, hi-Z windings want to resonate, open-grid inputs have infinite resistance and significant capacitance, and it has been done to use a high-resistance metal (like iron) for a hi-Z grid winding, so the DCR damps the high frequency resonance.

This UTC A-25 I am taking apart has a dcr of 36 ohms per secondary (600 Ohm windings)

The A-10 has about 36 ohms on each of it's primaries. (600 ohm imp), if that is any help.

[quote author="PRR"]If a winding's DCR is more than 10% of the load, loss will be bad.

[/quote]

So if I'm looking at a mic input transformer with primary DCR of 80R & secondary DCR of 10K6, it's likely not to be ideal for the job? (This is from an old Bozak mixer, btw.) I'm thinking about trying NYDave's single bottle pre, & was tryinng to figure out if anything I've got on hand might be a decent input transformer.

Tom

> if I'm looking at a mic input transformer with primary DCR of 80R & secondary DCR of 10K6, it's likely not to be ideal for the job?

What is "ideal"? What is truth? beauty?

80 ohms on one side suggests maybe 1600 ohms input, not an odd value for a mike input. The 80 ohms adds directly to the ~150 ohms of the mike to generate random noise; there are better transformers, but this ain't bad.

The 10K secondary suggests loading with more than 100K-200K; you are diving an open grid, infinite resistance, fine.

That is as far as you can go with DC tests. Pad a signal generator to 150 ohms and demagnetize the cores that you soaked with the ohm meter. Bring up 1V 20Hz, then turn down to less than 10mV.

Pad to ~150 ohms and 10mV, load the secondary with an ACVM, measure the output voltage (turns ratio) and run response sweeps. This won't be exactly right since the tube grid impedance isn't quite the same as an ACVM, but will rule-out a crappy transformer.

> is from an old Bozak mixer

From a tube Bozak? Then it will be excellent for the purpose, don't even bother testing, just build and enjoy.

[quote author="PRR"]>

From a tube Bozak? Then it will be excellent for the purpose, don't even bother testing, just build and enjoy.[/quote]

It's from a solid state mixer, but it seems like a decent candidate for the project. I'm going to give it a shot.

Thanks
Tom