Help determining output impedance.

[DLNB]

I'm trying to determine the output impedance of my Lifco Cobra Mark II amp head. I'm using this guide Output Transformers and Impedance and a few others and since I'm very new to this I just want someone to help me confirm that I'm doing it correctly and not missing something.

I'm connecting a 12v AC wall wart to the tube side of the output transformer (the smaller one near the output tubes). I've removed the tubes (not sure if I have to) and connected the 12 volts with aligator clips to the pins of the tube that lead to the transformer. My volt meter gives me a measurement of 0.2 volts at the output jack. Which leads directly from the transformer.

12 divided by .2 gives me 60. so the turn ratio is 60:1.

60x60=3600

3600 x 2 ohm load would give me 7200 which is probably correct for this amp with 2 6V6 power tubes.

So does that seem right? Am I missing something? One thing I am wondering is should I have disconnected the leads of the transformer from the circuit before I do this?

I have been running this amp with a 16 ohm cabinet. :O

Thanks very much for any input.

[DLNB]

I tried this again using a slightly larger ac power supply. This one puts out 14.7 volts. I get a return of .3 volts. My last reading was probably .29 or something. My multi tester only measure to one decimal.

This gives me a turn ratio of 49:1
49 x 49 = 2401

4 ohm load = 9.6K
2 ohm load = 4.8k

- Two 6v6's in the amp so I want around 8k?

So a 4 ohm cab would be my best bet? Just looking for confirmation that I'm not making a newbie mistake here. I really like the amp and I want to use the best matching cab possible.

[MWJB]

Your math is sound, either will work fine depending on plate voltage...old cathode biased 6V6 amps may have had primary Z in the region of 4.8K, a higher plate voltage might work better with a higher primary Z, so try 2ohm, 4ohms & maybe even 2.67 if you have 3x8ohm speakers. Whichever sounds best IS best.

[DLNB]

Cool, thanks for the reply.

[Chuck H]

One thing that escapes most is that you need to measure the input voltage with the unit under test. The OT usually loads down the input voltage at least a little and sometimes a lot. Also, I usually go the other way around by putting my test voltage into the secondary and measuring at the primary. Just for kicks do it both ways and see if your results change. And measure all input voltages with the test under way for an accurate base measure.

[Wakculloch]

Are the numbers used for speaker load not a root of the impedance.

[DLNB]

weird

So I tried doing it the other way as Chuck suggested and I get 14.7 volts in to 410 volts out. that gives me a turn ratio of 27.89:1.

Is it possible that my cheapo multi tester is too inaccurate to measure properly down to .2 and .3 volts ac?

If so then this is a better way to do it and probably the correct turn ratio?

[Chuck H]

So I tried doing it the other way as Chuck suggested and I get 14.7 volts in to 410 volts out. that gives me a turn ratio of 27.89:1.

Is it possible that my cheapo multi tester is too inaccurate to measure properly down to .2 and .3 volts ac?

Yes. But most are. My old B&K that I've been using for twenty years doesn't have a zero feature. This allows you to touch the probes together and tell the meter 'this is the starting point to measure from'. So if your meter always reads .2V with the probes touched together this feature allows to calibrate that to zero. Most good meters have this feature. I don't trust my old meter to accurately read low voltage.

If so then this is a better way to do it and probably the correct turn ratio?

Yepper. Bingo.

BTW, you mentioned earlier that you had a 14.7V source available. And now you mention that you put 14.7V in for the test... Did you test the voltage going in with the unit under test??? You really must do this. That 14.7V may have loaded down quite a bit and not knowing the actual input voltage makes the test result moot.

[DLNB]

BTW, you mentioned earlier that you had a 14.7V source available. And now you mention that you put 14.7V in for the test... Did you test the voltage going in with the unit under test??? You really must do this. That 14.7V may have loaded down quite a bit and not knowing the actual input voltage makes the test result moot.

No I did not. I will try that now. Thanks for the info.

[Chuck H]

How incredibly forthright and honest

[DLNB]

Okay so measuring the test voltage of 14.7 while it is connected to the OT I get 13.1 volts. That gives me a turn ratio of 31.3:1. If I do the same test with my 12V power supply it drops down to 10.3V under load. The return voltage is 324v so that gives me a turn ratio of 31.46.
I'm not sure why the values are slightly different with the two different voltages but it's close.
If I'm doing this correctly now it looks like this Lifco head wants an 8 ohm speaker load. Which is about what I would have expected for a head like this.

[J M Fahey]

Youy are doing it very well.
Your math is perfect, no quibble about that.
Real life says that we must work within the limitations of our measuring instruments.
Many forget that.
In this case, no matter how accurate the meter is in theory, the "last digit" is the practical limiting factor: it can only display a certain reading or that one "+1" or "-1".
Applying 12V to the primary, is the safest way to go, but gives you a quite innacurate secondary reading.
Doing it the other way, it can display numbers with much higher resolution, but obviously is more dangerous.
Meaningful voltages are those measured under load, of course.
31.3:1 and 31.46:1 *is* the same value.
After all, since your resolution is .1V , those 13.1V it displayed with the first transformer may in fact be anywhere between 13.05 and 13.15 V ; the 10.3 with the second one, anywhere from 10.25 and 10.35.
That's the error "window" the instrument has.
And as I said before, your Math is OK.

[Chuck H]

FWIW I kinda knew we would end up at an 8 ohm load. I wasn't going to say it outright because the OP didn't ask for that. He asked for the correct way to determine it

[DLNB]

Thats great. Thanks a lot for the help guys.