impedance matching

[fragman56]

Just a quick and simple question, please.

How do you go about selecting an output transformer for a tube amp?

Output impedance for tube amp (say SE) with OPT is given by:

Rout=Rl*rp/Rl+rp right?

where: Rl=tranny primary impedance (assuming a secondary loading from speakers)
rp=tube internal resistance

For best power and results, best to match amp output impedance to prim tranny impedance, right? So, can Rout ever equal Rl?

What am I missing here?

Thanks,
Rick

[fragman56]

I must be that dense...where's Joel? Frank? John?

Anybody there?

[dhaen]

Rick,

I was waiting for someone to explain it more eloquently than I can.
The way I see it, you are right about maximum power transfer, but that does not correspond with "best" distortion.
I believe the distortion goes down and down as the load impedance is raised. At the same time, the power also goes down.
The "rule of thumb" of Rl being 2 to 3 times Ra is just a compromise of power vs distortion.

Cheers,

[fragman56]

John,

I appreciate your response...this stuff is difficult to comprehend without a background in electronics, I think.

The relationship you described power vs distortion and impedance seems to bear out since I use a Hammond 125ese which has different connections.

Your explanations are perfect for me!

Rick

[Gabevee]

OK, this is one where I only know things like ratios and the like. But... I thought distortion goes down as current goes up? Wouldn't a lower primary give less distortion due to higher current for signal?

I generally find this true for small signal design. But perhaps not for power output.

Higher impedance gives narrower bandwidth, no?


Curious,
Gabe

[dhaen]

Gabe,

I think you are confusing the distortion due to the valve being loaded, and the distortion inherent in the transformer.

BTW I didn't mention current

Cheers,

[fdegrove]

Hi,

Choosing optimum load for output tube /SE amp/:

For triodes choose - RL = (2-6) RP /optimum is around 4 Rp/

Higher RL for triodes means lower distortion and lower output power.

The output power is about 1/4 of PP - plate dissipation. Pout= 0.25 PP

For pentodes or beam power tubes - RL = (0.1) RP

For pentodes, tetrodes or beam power tubes the impedance is critical for best performance - higher power and lower distortion.

The output power is about 1/2 of PP - plate dissipation. Pout= 0.5 PP

WHERE: RL - load resistance /impedance of transformer primary/; RP - plate resistance of chosen valve.


Also determine loudspeaker impedance Z to match with the transformer - that is usually 4, 6, 8 or 16 Ohm. That will be reflected on the primary /transformed/ with the transformer.

Hope this helps.

Gabe,

The current flow through the outputstage may well affect distortion figures, it won't have a direct relationship to the OPT however.

Cheers,

[Gabevee]

dhaen,

I know you didn't mention current... but it does come into play doesn't it?

fdegrove,

The current flow through the outputstage may well affect distortion figures, it won't have a direct relationship to the OPT however.

Why?

As an interesting side note, I measured the voltage across the primary of an SE amplifier with a signal at full output and it measured something like 30-40 volts higher than B+. Interesting, no?

Gabe

[fragman56]

Thank you John, Gabe and Frank,

Very helpful!

Appreciate the info,

While I've got you, another simple question, please: How does the impedance from a vol pot affect the hum at the speaker in a two stage RC coupled, se triode?

That is, as the impedance of the vol pot goes up, the hum at the speaker increases. Using a 10k vol pot produces no hum but changing to a 100k or 250k ohm vol pot increases hum proportionately.

Source is CDP.

Any assistance is greatly appreciated,

Rick

[fdegrove]

Hi,

The current flow through the outputstage may well affect distortion figures, it won't have a direct relationship to the OPT however.

Gabe,

The only reason I see for this becoming a problem is with badly specified OPTs, i.e. an SE OPT with a gap for say 100 mA and 120 mA run through it.

If an OPT is overloaded it will distort but that isn't actually related to the standing current of the outputstage.

Rick,

Using a 10k vol pot produces no hum but changing to a 100k or 250k ohm vol pot increases hum proportionately.

The 10K pot shunts alot more to ground than the 100 or 250K pot.
If this is an actual problem you have than there definetely is a hum problem with that circuit.

Cheers,