Hi Folks
Has anyone ever modded one of these amps? They are actually quite good for what they are. This amp is the stereo version of the UA-1 or UA-2 mono amps while it also has a preamp and phono section. The output iron is ultralinear and the circuit of course uses the tri/pent 6AN8 which I like.
It's no Sherwood S-5000 but I see some potential in this little gem. I also have the AA-151 Heath Integrated which is very similar as well.
I did change the input filter cap and added two more sections underneath.
I didn't change the coupling caps yet but what I am going to so a partial film cap supply and put in a modern selector switch. There are two separate 6.3v taps and it does use a hum pot for each channel but I was going to do DC fil for just the phono and preamp tubes.
Has anyone modded one of these with a separate choke along with doing the tweaked RCA phono stage? Maybe Eli can jump in.
Has anyone ever modded one of these amps? They are actually quite good for what they are. This amp is the stereo version of the UA-1 or UA-2 mono amps while it also has a preamp and phono section. The output iron is ultralinear and the circuit of course uses the tri/pent 6AN8 which I like.
It's no Sherwood S-5000 but I see some potential in this little gem. I also have the AA-151 Heath Integrated which is very similar as well.
I did change the input filter cap and added two more sections underneath.
I didn't change the coupling caps yet but what I am going to so a partial film cap supply and put in a modern selector switch. There are two separate 6.3v taps and it does use a hum pot for each channel but I was going to do DC fil for just the phono and preamp tubes.
Has anyone modded one of these with a separate choke along with doing the tweaked RCA phono stage? Maybe Eli can jump in.
Attachments
Last edited:
Mike,
You've got "iron", sockets, and sheet metal. 😀 There's something to work with.
I'm providing some additional photos, which I acquired off the WWW some time ago. A "shot" of the back panel would be nice.
A proper, DC, heater supply for the phono section requires switching to SS rectifying the B+ and voltage multiplying the 5 VAC winding. Mount a B+ PSU filter choke on the outside of the chassis' side wall near the power trafo.
You've got "iron", sockets, and sheet metal. 😀 There's something to work with.
I'm providing some additional photos, which I acquired off the WWW some time ago. A "shot" of the back panel would be nice.
A proper, DC, heater supply for the phono section requires switching to SS rectifying the B+ and voltage multiplying the 5 VAC winding. Mount a B+ PSU filter choke on the outside of the chassis' side wall near the power trafo.
Attachments
It appears that you have only one 100 Ohm self bias resistor, and only one bypass cap for all 4 output tubes. This is a case where sharing is a bad thing.
What is the cost of a Matched Quad of output tubes?
Versus . . .
What is the cost of four 400 Ohm individual self bias resistors, and four individual bypass caps? (Todays modern electrolytic caps are much smaller for the same C x V rating).
Fit them in there.
That is much less expensive than a very well Matched Quad of output tubes.
The output transformers will love the matched currents they get even if the tubes are not closely matched. Your Bass distortion will go way down.
What is the cost of a Matched Quad of output tubes?
Versus . . .
What is the cost of four 400 Ohm individual self bias resistors, and four individual bypass caps? (Todays modern electrolytic caps are much smaller for the same C x V rating).
Fit them in there.
That is much less expensive than a very well Matched Quad of output tubes.
The output transformers will love the matched currents they get even if the tubes are not closely matched. Your Bass distortion will go way down.
I agree
That's all going to be changed. You could use a 400ohm cathode resistor on each tube but I would love to go with combination bias.
That's all going to be changed. You could use a 400ohm cathode resistor on each tube but I would love to go with combination bias.
Thanks Eli
I was going to split off each 6.3v filament supply and just rectify it for the phono tubes. I was going to leave the rest ACv.
I was going to split off each 6.3v filament supply and just rectify it for the phono tubes. I was going to leave the rest ACv.
Cathode or Fixed Bias?
First of all, the bias voltage for EL84 / 6BQ5 is only a few volts.
It is not a large Beam Power tube, or a large Pentode, they require 30, 40, or 50V of bias.
Those large bias voltages do rob lots of B+ voltage when you use self bias, and that robs output power.
If you need 17.22 Watts at 20kHz going to your Tweeter, then you need to use fixed bias for that pair of EL84 /6BQ5. 1dB less power of 13.69 Watts is so much quieter.
1 dB difference of continuous power will not make any difference, both will burn out 95% of all tweeters.
In regards to power at 20Hz, nothing will ruin the distortion faster than unbalanced DC in the push and pull windings.
If you have any doubts about the extra distortion caused by un-balanced DC currents, look up the very last 2 pages of the Heathkit W5 amplifier manual.
It shows the increased distortion at 20Hz and even at 100 Hz, when there is just a small un-balancing of the push and pull quiescent DC currents.
And, that graph was taken With the global negative feedback.
Global negative feedback can not fix some of the errors in the design and implementation of an amplifier, and this is one of them it can not fix.
Combination bias gives both some auto adjustment range, and some (adjustable fixed) manual bias adjustment range of the current.
But the amount of auto adjustment due to the self bias resistor is proportional to the ratio of the resistance of the self bias resistor, versus the impedance of the cathode.
That means the (fixed adjustable) manual bias robs from the range of the self bias circuit (less auto adjustment effect).
A self bias resistor that gives all of the bias has a good self adjustment range (no fixed bias circuit).
A self bias resistor that gives only 1/2 of the bias voltage, only has 1/2 of the self bias auto range (1/2 of the bias in the fixed bias circuit).
Just like so many things, it is a tradeoff.
Self bias for a pair of push pull tubes:
2 resistors, 2 bypass caps.
Multiply that by 2 for Stereo.
Fixed adjustable bias for a pair of push pull tubes:
Bias voltage supply, 2 potentiometers.
Multiply the potentiometers by 2 for Stereo.
Combination bias:
2 resistors, 2 bypass caps.
Bias voltage supply, 2 potentiometers.
Now, multiply that by 2 for Stereo (well, only one bias supply).
First of all, the bias voltage for EL84 / 6BQ5 is only a few volts.
It is not a large Beam Power tube, or a large Pentode, they require 30, 40, or 50V of bias.
Those large bias voltages do rob lots of B+ voltage when you use self bias, and that robs output power.
If you need 17.22 Watts at 20kHz going to your Tweeter, then you need to use fixed bias for that pair of EL84 /6BQ5. 1dB less power of 13.69 Watts is so much quieter.
1 dB difference of continuous power will not make any difference, both will burn out 95% of all tweeters.
In regards to power at 20Hz, nothing will ruin the distortion faster than unbalanced DC in the push and pull windings.
If you have any doubts about the extra distortion caused by un-balanced DC currents, look up the very last 2 pages of the Heathkit W5 amplifier manual.
It shows the increased distortion at 20Hz and even at 100 Hz, when there is just a small un-balancing of the push and pull quiescent DC currents.
And, that graph was taken With the global negative feedback.
Global negative feedback can not fix some of the errors in the design and implementation of an amplifier, and this is one of them it can not fix.
Combination bias gives both some auto adjustment range, and some (adjustable fixed) manual bias adjustment range of the current.
But the amount of auto adjustment due to the self bias resistor is proportional to the ratio of the resistance of the self bias resistor, versus the impedance of the cathode.
That means the (fixed adjustable) manual bias robs from the range of the self bias circuit (less auto adjustment effect).
A self bias resistor that gives all of the bias has a good self adjustment range (no fixed bias circuit).
A self bias resistor that gives only 1/2 of the bias voltage, only has 1/2 of the self bias auto range (1/2 of the bias in the fixed bias circuit).
Just like so many things, it is a tradeoff.
Self bias for a pair of push pull tubes:
2 resistors, 2 bypass caps.
Multiply that by 2 for Stereo.
Fixed adjustable bias for a pair of push pull tubes:
Bias voltage supply, 2 potentiometers.
Multiply the potentiometers by 2 for Stereo.
Combination bias:
2 resistors, 2 bypass caps.
Bias voltage supply, 2 potentiometers.
Now, multiply that by 2 for Stereo (well, only one bias supply).
Last edited:
Take a pair of O/P tubes that are very well matched in gm (always mandatory) and somewhat well matched in cathode current and "stand" the pair on a shared 100 Ω/470 μF. bias network, which also provides a convenient "idle" current test point. Now, only 1 bias trim pot./channel is needed and minor cathode current imbalances auto-correct.
Sourcing suitably matched pairs for the technique described above is easier and less costly than the very well matched (both gm and cathode current) quartet that Heath used in the SA-2 and Dyna used in the ST-35.
Sourcing suitably matched pairs for the technique described above is easier and less costly than the very well matched (both gm and cathode current) quartet that Heath used in the SA-2 and Dyna used in the ST-35.
I "harvested" the iron from an AA-151 and used it with a Tubelab.com SPP board. I also built one with new Hammond iron. Yes, I know, the Hammond isn't the greatest but it's OK. Anyway, the SPP with the Hammond iron had noticeably clearer mids and highs. The SPP with Heath iron was OK, but clearly out-classed by the Hammond-equipped amp.
S.
S.