Bypass a tube to reduce power?

[petertub]

I am only referimg to the measurments i have done .

There is no crossover in any power on a sane st70. Hum would be a more urgent issue, but as the poster does not mention hum i guess that the only thing that bothers him is plain level. Adding a resistor on the volume pot input as suggested seems an

elegant method. If the pot is 100k, then a resistor on between source and pot of 200k would make life better ( is should be mounted between source and outer connection on pot NOT the viper)

 

[RJM1]

You could also reduce the value of the 1k feedback resistor to something like 330 ohms, this would
also lower distortion and increase frequency response, The feedback capacitor would not have to be changed as the zero db gain / frequency intercept point would not move up.

 

[rayma]

I would like to have a wider range of volume control. I have Klipsch La Scalas which have a sensitivity of 105db.

Then you need a preamp with much less gain (0 to 6dB) in the line stage, or a passive line stage with just a volume control. Don't change or attempt to redesign the amplifier at all.

Any theoretical or empirical evidence that crossover is not detectable with a ST70 amp and Klipsch Las Scala speakers? I would be interested to know if I heard something else and not crossover distortion.

The output stage operates more and more completely in class A push-pull as the power level is reduced. The ST70 was a favorite of many with Klipsch speakers for many years.

 

[petertub]

You could also reduce the value of the 1k feedback resistor to something like 330 ohms, this would
also lower distortion and increase frequency response, The feedback capacitor would not have to be changed as the zero db gain / frequency intercept point would not move up.

This i would advice against. It might cause instability. Feedback in st70
is quite optimal.

 

[RJM1]

It's not like an opamp, it's unity gain stable.

 

[hpeter]

Put a resistor in series with the pot, that would shift the usable range downward.

Jan

i would put 5-10k between pot wiper and ground.

 

[JMFahey]

* Silly way: attenuate after amplifier or reduce power output by mutilating it.

It will be as uncontrollable as before and distort a ton, in a very ugly way.

* Intelligent way: put a resistor in series with volume pot.

Will NOT distort , be controllable and have huge headroom.


Pick one.

 

[johnhogie]

Pragmatic solution for now.

I am interested in experimenting with this amp but need to be realistic about my current knowledge and skills. (Pun not intended.) I do not want to get rid of the speakers so I am going to start looking for a low power (someone suggested an EL84 based SE) amp. Thanks again for everyone's help.

 

[6A3sUMMER]

RJM1,

Get a Dyna Stereo 70, and change the 1k feedback resistor to 330 Ohms. Do not change the capacitor that was across the 1k, and do not change the capacitor that was from the UL tap to earlier circuitry, and do not change the series RC dominant pole from the pentode plate to ground.
We are still 10dB or more away from unity gain, even though we increased the negative feedback by 9.6 dB.

But, just try different loudspeakers until you find one that causes the amp to go un-stable.

Who has ever tried it?

 

[pblix]

I see two good options:
1) Keep the amp, sell the speakers. Buy something less sensitive, say under 90dB/W/m. Additional benefit is the WAF will increase
2) Keep the speakers, sell the amp. Same in reverse. Get a flea-power amp.

YMMV

 

[jan.didden]

I am interested in experimenting with this amp but need to be realistic about my current knowledge and skills. (Pun not intended.) I do not want to get rid of the speakers so I am going to start looking for a low power (someone suggested an EL84 based SE) amp. Thanks again for everyone's help.

This seems a very drastic and non-optimal plan to solve something that can be fixed by a couple of $ 0.10 resistors ...

And, lets call it what it is: you are NOT looking for a lower power amp. If the new amp has the same gain as the old one, it will sound exactly as loud as the old one!!
A 50W amp with 20x gain will sound just as loud on your speakers as a 20W amp with gain 20x. Get that. You want lower gain!!

Jan

 

[petertub]

This seems a very drastic and non-optimal plan to solve something that can be fixed by a couple of $ 0.10 resistors ...

And, lets call it what it is: you are NOT looking for a lower power amp. If the new amp has the same gain as the old one, it will sound exactly as loud as the old one!!
A 50W amp with 20x gain will sound just as loud on your speakers as a 20W amp with gain 20x. Get that. You want lower gain!!

Jan

A very good summary.
I would like to add that keeping an amp that works fine, that has a history
you know, and that works, is much better then selling it ( for less then your
affection-value), and buying an unknown amp with unknown faults is a
potential looser.
What you need is to reduce the gain, and the easiest way is to add
two resistors to the existing volume pot. Or, if that feels better, build
a 2 resistor voltage divider for each channel. Either way you will reduce
the total gain.

 

[huggygood]

A very good summary.
I would like to add that keeping an amp that works fine, that has a history
you know, and that works, is much better then selling it ( for less then your
affection-value), and buying an unknown amp with unknown faults is a
potential looser.
What you need is to reduce the gain, and the easiest way is to add
two resistors to the existing volume pot. Or, if that feels better, build
a 2 resistor voltage divider for each channel. Either way you will reduce
the total gain.

+4859625458963215478

 

[Wavebourn]

The best way to reduce power increasing sound quality is to increase primary impedance of the output transformer using more turns of thinner wire, and increasing bias current of the output tubes up to the maximum that tubes can dissipate.

 

[kevinkr]

Look into triode connection in the output stage, this will reduce output power by roughly half.

Not sure what your level of technical skill is - can you solder, follow basic schematic/written instructions?

Triode strapping EL34 is pretty easy, remove and insulate with heat shrink tubing the wires coming from output transformers connected to pin 4 on each of the 4 output tube sockets. Solder a 100 ohm 1W 5% MO resistor between pin 3 and 4 of each

Check the operating bias after making the modification and adjust to the correct value. (Lower than stock is OK, I typically used 1.2 - 1.3V)

 

[jan.didden]

Kevin he doesn't need less power, he needs less gain!

Jan

 

[kevinkr]

Given the sensitivity of the speaker system I would recommend both. Having reduced the open loop gain by going to triode connection he can experiment with either feedback adjustments (unlikely since I doubt he has a scope and the experience to do so, but a reduction of 3dB or so should be safe) or he can add an attenuator in front of the amplifier or both. (Note also that just switching to triode output connection will probably reduce the closed loop gain by roughly a couple of dB under load)

Truthfully he should get rid of the ST-70 and purchase or build an appropriate SE amplifier. One or the other of the Tubelab SE amplifiers would seem to be a better match.

No one has asked what he is driving the amplifier with, it could be simply that they are too hot. The ST-70 does not have particularly unreasonable CLG.

 

[Wavebourn]

Look into triode connection in the output stage, this will reduce output power by roughly half.

One more option, more realistic. It would decrease OL gain, so twice lower values of feedback resistors would be safer.

 

[johnhogie]

Front End

Given the sensitivity of the speaker system I would recommend both. Having reduced the open loop gain by going to triode connection he can experiment with either feedback adjustments (unlikely since I doubt he has a scope and the experience to do so, but a reduction of 3dB or so should be safe) or he can add an attenuator in front of the amplifier or both. (Note also that just switching to triode output connection will probably reduce the closed loop gain by roughly a couple of dB under load)

Truthfully he should get rid of the ST-70 and purchase or build an appropriate SE amplifier. One or the other of the Tubelab SE amplifiers would seem to be a better match.

No one has asked what he is driving the amplifier with, it could be simply that they are too hot. The ST-70 does not have particularly unreasonable CLG.

That was asked in a PM. I was using an Adcom 555 preamp. Today I was able to test a First Watt B1 and will be replacing the Adcom. Ahead of the preamp is a Lounge Audio LCR phono preamp which is connected to a Bang and Olufsen turntable.

 

[jan.didden]

Given the sensitivity of the speaker system I would recommend both. Having reduced the open loop gain by going to triode connection he can experiment with either feedback adjustments (unlikely since I doubt he has a scope and the experience to do so, but a reduction of 3dB or so should be safe) or he can add an attenuator in front of the amplifier or both. (Note also that just switching to triode output connection will probably reduce the closed loop gain by roughly a couple of dB under load)

Truthfully he should get rid of the ST-70 and purchase or build an appropriate SE amplifier. One or the other of the Tubelab SE amplifiers would seem to be a better match.

No one has asked what he is driving the amplifier with, it could be simply that they are too hot. The ST-70 does not have particularly unreasonable CLG.

So now we are moving from a simple 2-resistor mod to replacing the entire equipment chain? Well that's diy for you I guess ;-)

Jan