Anonther way to restore balance function
The balance pot in PAS3 and SCA35 is unique. Half the track is fully conducting, the other half is the 750k resistive track.
This way dynaco managed to make a balance pot that does not reduce signal.
Replacing with a dual 750k pot ( congrats if you can find such a beast) would
reduce the signal with 50%, which in the SCA35 would be fatal.
What can be done is to replace the dynaco pot with a dual 250k, and instead of
grounding one side as per the 750k , insert a resistor in the ground connection.
This way the signal is reduced by 125/625 in the middle position, the balance
range is very reduced. The balance range might not be such a bad thing as
balance often needs only minor adjustments from center.
Dual 250k pots are available, replace everything as with the existing 750k
with two exceptions: the ground wire that grounds one end of the track
should be replaced with a 470k resistor between the pot and ground.
The balance pot in PAS3 and SCA35 is unique. Half the track is fully conducting, the other half is the 750k resistive track.
This way dynaco managed to make a balance pot that does not reduce signal.
Replacing with a dual 750k pot ( congrats if you can find such a beast) would
reduce the signal with 50%, which in the SCA35 would be fatal.
What can be done is to replace the dynaco pot with a dual 250k, and instead of
grounding one side as per the 750k , insert a resistor in the ground connection.
This way the signal is reduced by 125/625 in the middle position, the balance
range is very reduced. The balance range might not be such a bad thing as
balance often needs only minor adjustments from center.
Dual 250k pots are available, replace everything as with the existing 750k
with two exceptions: the ground wire that grounds one end of the track
should be replaced with a 470k resistor between the pot and ground.
Last edited:
You could use a 12 pos switch as they are easier to get hold of - and make the two centre pos the same - nobody will know. I don't think you need such fine steps - maybe 2dB per pos. You may want to make the max out positions just L or R for testing as PPR suggested. To centre position should be really -3dB both channels so you don't get a shift in apparent volume as you alter the position.
This may work for you using a dual linear 1Meg pot. The extra resistors will make it more log like.
Older tube amplifiers have normally more gain than that today is required. If I wanted to add a balance control (and I was seriously thinking about it when I made my Audio Note clone) then I would use Bernd Ludwig's "a better balance and volume control". Simple linear pots can be used. I've attached a PDF describing this.
Sorry I should have said anti-log for the balance rather than log which is why the resistor goes to signal rather than ground.
Adding a resistor does have two disadvantages. First the input impedance changes with position, so for the balance control it may affect the tone control near maximum attenuation.
Second it does put some signal current through the pot wiper which can make a gritty sound as the volume is adjusted. That's why a log pot is better for volume.
Your switch idea will have a similar issue. A make before break switch will sound better as the audio won't disappear between positions.
Adding a resistor does have two disadvantages. First the input impedance changes with position, so for the balance control it may affect the tone control near maximum attenuation.
Second it does put some signal current through the pot wiper which can make a gritty sound as the volume is adjusted. That's why a log pot is better for volume.
Your switch idea will have a similar issue. A make before break switch will sound better as the audio won't disappear between positions.
If there is a change in tone when you change volume then there is something seriously wrong with the preceding stage.
I have taken a great dislike to stepped attenuators: when listening at low volume one never seems to be able to set it just right.
I've gone to using TKD potentiometers as the Alps that I measured did vary too much between channels when value changed.
I have taken a great dislike to stepped attenuators: when listening at low volume one never seems to be able to set it just right.
I've gone to using TKD potentiometers as the Alps that I measured did vary too much between channels when value changed.
SCA35 has no extra gain to burn in an inefficient balance control. Added
on that is that the tonecontrol needs a constant loading to be accurate.
The 11 step dual switch (or greater than 11 steps) can give you:
No Loss at center.
No loss for Left channel for 5 more CCW steps.
No loss for Right channel for 5 more CW steps.
The same input impedance per attenuation (as the original Dyna control).
The same output impedance per attenuation (as the original Dyna control).
A few steps to balance any complete system that does not need to be otherwise fixed (signal sources, speakers, room, etc.).
And, infinite attenuation at CW (Left channel), and CCW (Right Channel).
Perfect repeatability.
Find me a control that does all that!
Here it is:
L . . . . . R
0 dB ... Ground (infinite attenuation)
0 dB ... -4 dB
0 dB ... -3 dB
0 dB ... -2 dB
0 dB ... -1 dB
0 dB ... 0 dB
-1 dB ... 0 dB
-2 dB ... 0 dB
-3 dB ... 0 dB
-4 dB ... 0 dB
Ground ... 0 dB
(infinite
Atten.)
And . . .
The required resistance at each tap:
Attenuation Resistance at switch output
0 dB Switch contacts the top 750k total . . . 750k
-1 dB Switch contacts 89% of 750k . . . 667.5k
-2 dB Switch contacts 79% of 750k . . . 592.5k
-3 dB Switch contacts 70.7% of 750k . . . 530.25k
-4 dB Switch contacts 63% of 750k . . . 472.5k
A little subtraction math gives the resistor values required between the taps.
'Silver Platter'
Done
I used to own another amplifier that had the same special log balance pot, with the conductive trace on one channel for ~ 1/2 CCW rotation, and the other channel with the conductive trace on the other channel for ~ 1/2 CW rotation.
I never needed more than 4 dB to balance, and the complete attenuation of the channels at the CW and CCW ends 'respectively' was handy too.
No Loss at center.
No loss for Left channel for 5 more CCW steps.
No loss for Right channel for 5 more CW steps.
The same input impedance per attenuation (as the original Dyna control).
The same output impedance per attenuation (as the original Dyna control).
A few steps to balance any complete system that does not need to be otherwise fixed (signal sources, speakers, room, etc.).
And, infinite attenuation at CW (Left channel), and CCW (Right Channel).
Perfect repeatability.
Find me a control that does all that!
Here it is:
L . . . . . R
0 dB ... Ground (infinite attenuation)
0 dB ... -4 dB
0 dB ... -3 dB
0 dB ... -2 dB
0 dB ... -1 dB
0 dB ... 0 dB
-1 dB ... 0 dB
-2 dB ... 0 dB
-3 dB ... 0 dB
-4 dB ... 0 dB
Ground ... 0 dB
(infinite
Atten.)
And . . .
The required resistance at each tap:
Attenuation Resistance at switch output
0 dB Switch contacts the top 750k total . . . 750k
-1 dB Switch contacts 89% of 750k . . . 667.5k
-2 dB Switch contacts 79% of 750k . . . 592.5k
-3 dB Switch contacts 70.7% of 750k . . . 530.25k
-4 dB Switch contacts 63% of 750k . . . 472.5k
A little subtraction math gives the resistor values required between the taps.
'Silver Platter'
Done
I used to own another amplifier that had the same special log balance pot, with the conductive trace on one channel for ~ 1/2 CCW rotation, and the other channel with the conductive trace on the other channel for ~ 1/2 CW rotation.
I never needed more than 4 dB to balance, and the complete attenuation of the channels at the CW and CCW ends 'respectively' was handy too.
Last edited:
I believe the original poster had an 11 position switch. Post # 1
And another poster said a 12 step is available. Post # 22
I have only seen 'mono' 12 steps, not dual.
And another poster said a 12 step is available. Post # 22
I have only seen 'mono' 12 steps, not dual.
Last edited:
I think if your stuck a dual gang linear 1M pot will work with 390k from each wiper to each hot end. It may affect the tone stack slightly but only on the channel that is wound down. At centre position the impedance will be about 750k.
Questions for Petertub on this SCA-35 balance control idea. Is the 250k pot a balance pot? Or a volume pot, and if it's a volume pot, linear or log taper? Also, do the grounds of the other controls also pass through the added 470k resistor?
I have tried this with a 250k balance pot passing all the grounds through the respective resistors and while the balance seems to work as described, the volume control then becomes ineffective -- very little change if any in level regardless of position (even fully counter-clockwise). Sending the treble, bass, and volume control grounds directly to ground while sending the balance control through the 470k resistors yields the same problem with the volume control while the bass and treble controls seem to have little effect.
Of course, I could have wired these things entirely wrongly.
Thanks for any advice.
The balance pot should be of LIN type.
All other pots is left as in the manual.
The area is cramped and one must arrange proper ground for pin 6 of the PAC, volume pot filter and loudness circuit.
All other pots is left as in the manual.
The area is cramped and one must arrange proper ground for pin 6 of the PAC, volume pot filter and loudness circuit.
Last edited: