I have a Dynaco SCA-35 that I've re-done for and with my son, using the Dave Gillespie update boards.
On the whole it has turned out great but one problem I'm left chasing is the balance control, which keeps going intermittent despite being carefully disassembled and cleaned.
Since 750K M-N taper balance pots of high quality appear to be as scarce as unicorns, I am looking at putting in an 11-step 2 pole rotary switch, with resistors.
I recognize that to work like an M-N taper, each deck should have six of the 11 positions with no resistors, in mirror image (opposite directions) to one another, with the center position having no attenuation of either channel.
I also understand that the "taper" of the remaining five positions that will have resistors should be logarithmic, but that's where I start to get lost - abstract math has never been one of my areas of skill/comfort.
Can anyone please advise how I would "taper" 750K over five positions to work like a logarithmic balance control? Or a "for dummies" way that I could calculate it?
Thanks in advance!
On the whole it has turned out great but one problem I'm left chasing is the balance control, which keeps going intermittent despite being carefully disassembled and cleaned.
Since 750K M-N taper balance pots of high quality appear to be as scarce as unicorns, I am looking at putting in an 11-step 2 pole rotary switch, with resistors.
I recognize that to work like an M-N taper, each deck should have six of the 11 positions with no resistors, in mirror image (opposite directions) to one another, with the center position having no attenuation of either channel.
I also understand that the "taper" of the remaining five positions that will have resistors should be logarithmic, but that's where I start to get lost - abstract math has never been one of my areas of skill/comfort.
Can anyone please advise how I would "taper" 750K over five positions to work like a logarithmic balance control? Or a "for dummies" way that I could calculate it?
Thanks in advance!
Do we really need balance controls?
I always hated the balance control on my Harmon Kardon A30 kit amplifier.
It had two capacitors that changed the frequency response slightly of one channel at a time, as you rotated either clockwise or counterclockwise.
Other than keeping the amp stock, I wonder why I did not pull the balance control out.
For decades after that, I have not used balance controls, except for special applications, such as to put the sound near the TV when I use the Hi Fi for better sound (the other speaker is far from the TV).
In the old days, some recordings were poorly balanced; some mono Hi Fi's were doubled up in order to make it stereo, but they had different L and R speakers; the speaker placement in the room was acoustically unbalanced; etc.
Question:
How many of you actually use a balance control now?
I always hated the balance control on my Harmon Kardon A30 kit amplifier.
It had two capacitors that changed the frequency response slightly of one channel at a time, as you rotated either clockwise or counterclockwise.
Other than keeping the amp stock, I wonder why I did not pull the balance control out.
For decades after that, I have not used balance controls, except for special applications, such as to put the sound near the TV when I use the Hi Fi for better sound (the other speaker is far from the TV).
In the old days, some recordings were poorly balanced; some mono Hi Fi's were doubled up in order to make it stereo, but they had different L and R speakers; the speaker placement in the room was acoustically unbalanced; etc.
Question:
How many of you actually use a balance control now?
It would be a real advantage to instead replace the volume control with a switched resistor control.
This would give much better and more consistent channel balance. No balance or loudness needed.
DACT Type 21 Stepped Attenuator / Volume Control (Hi Fi Grade) * D shape Shaft * | eBay
This would give much better and more consistent channel balance. No balance or loudness needed.
DACT Type 21 Stepped Attenuator / Volume Control (Hi Fi Grade) * D shape Shaft * | eBay
About the only time I use a balance control is if something is sounding odd and I want to try to isolate whether it's one channel or the other, or equally odd in both channels. So if I were designing from scratch I'd seriously consider omitting a balance control.
But, this is an existing piece of equipment, and so I'd just as soon not either leave a void on the front panel or have a knob that does nothing, and it does seem useful for what I mention above. I've thought of just a switch, switching out one channel, or the other but it doesn't seem like a lot of extra bother to have a quasi-functional stepped arrangement with an 11-step switch, which I have already bought at modest cost on eBay.
Last edited:
Rayma,
I like that idea. Accurate matched stepped volume control.
In fact I have a remote controlled precision accuracy stepped volume control in one of my systems.
Even if the gains of the tubes are not perfectly matched, the precision tolerance of the negative feedback resistors in the amp matches the gain of the channels.
I like that idea. Accurate matched stepped volume control.
In fact I have a remote controlled precision accuracy stepped volume control in one of my systems.
Even if the gains of the tubes are not perfectly matched, the precision tolerance of the negative feedback resistors in the amp matches the gain of the channels.
I've looked at the real-make DACT controls and they seem like good quality but the prices are outrageous. And I don't trust a pseudo-DACT "type" bargain priced control from eBay out of the far east.
I already have a real, not fake, ALPS stepped attenuator for the main volume control.
What I am intending to do are switched resistors on a nice ceramic double-deck 11 position switch that I already have, as a balance control of sorts. I just need to get to the right value of resistors.
Last edited:
Respectfully, I am not asking what someone else wants or suggests doing with some other part and different method on their own other piece of equipment.
I'm interested in some help in getting to the right values for a "log-equivalent" of five stepped resistors
Rather than a pseudo-log taper, decide how much imbalance you want in dB for each position,
and then calculate the resistors accordingly. For example, for each side the first position could
be -1dB, the next -2dB, then -3dB, -6dB, and off. You really only want to correct for
a small imbalance, since anything more indicates a problem somewhere.
and then calculate the resistors accordingly. For example, for each side the first position could
be -1dB, the next -2dB, then -3dB, -6dB, and off. You really only want to correct for
a small imbalance, since anything more indicates a problem somewhere.
On this circuit, the balance pot hangs off the tone pots and PECs, and is then loaded by a 4.7M resistor.
With this passive tone circuit that balances the tone impedances, you'd have to estimate this to calculate
the actual attenuation. It's still a 750k balance pot, like Dyna's other tube preamps.
The easiest way would be to add a resistor between top of the Dyna balance pot and its previous connection,
and adjust the value of that resistor to get -1dB, or -2dB, or -3dB , etc. at the line stage output. Those values
will give you enough info to calculate the needed resistor values.
Last edited:
Thanks - one of the more legible sets of schematics is here, in two parts:
http://www.quadesl.com/schematics/sca35a.jpg
http://www.quadesl.com/schematics/sca35b.jpg
Thanks!
Last edited:
Unless your balance errors are exceptionally well disciplined, there is no strong reason to make them exactly equal-in-dB.
And good reason to avoid brain-strain for theoretical purity.
We know it will be 750k total. And 1dB is 10%. So start with a 75k resistor. Get lazy, throw in some more. In fact 3 steps comes out 525/750 or -3.1dB which is "3" for any practical purpose. (Again: no reason to suspect balance errors will trend to whole-dB intervals.) So throw in a fourth 75k and finish off with 450k to make the 750k total.
If your rig is more than 4dB off center maybe you should fix it. However sometimes we got what we got. Try 2dB, 20% near-enough. Throwing 150k at the poles comes out to 750k but the steps grow to 6dB. If you have to fix a large unbalance, that rising-step sequence may be useful.
So split the difference. Use 100k. This grows from 1.2dB to 2.2dB.
If you need to be fussier than that you probably want many more steps.
Attachments
OK.
You have to preserve the total resistance of 750k to make the rest of the circuitry work as per original (original max gain, and original frequency response).
Attenuation Resistance at switch output
0 dB Switch contacts the top 750k total.
-1 dB Switch contacts 89% of 750k
-2 dB Switch contacts 79% of 750k
-3 dB Switch contacts 70.7% of 750k
-4 dB Switch contacts 63% of 750k
-5 dB Switch contacts 56% of 750k 375k in series with a 45k resistor = 420k to ground)
-6 dB Switch contacts 50% of 750k (375k resistor to ground)
I think you can calculate the additional resistors for each step in the series string to get 750k to ground at the 0 dB setting.
You have to preserve the total resistance of 750k to make the rest of the circuitry work as per original (original max gain, and original frequency response).
Attenuation Resistance at switch output
0 dB Switch contacts the top 750k total.
-1 dB Switch contacts 89% of 750k
-2 dB Switch contacts 79% of 750k
-3 dB Switch contacts 70.7% of 750k
-4 dB Switch contacts 63% of 750k
-5 dB Switch contacts 56% of 750k 375k in series with a 45k resistor = 420k to ground)
-6 dB Switch contacts 50% of 750k (375k resistor to ground)
I think you can calculate the additional resistors for each step in the series string to get 750k to ground at the 0 dB setting.
Thank you all VERY much for this input and guidance- now I'll give some thought as to how "fast" a taper I want to aim for- but the info that you have all provided gets me unstuck.
I do recognize that there's not going to be a lot of subtlety with only five working steps between no attenuation and 100% attenuation. This is just an attempt to do something that leaves some degree of functionality that's better than nothing and more trouble free than a potentiometer.
I really appreciate the input.
I do recognize that there's not going to be a lot of subtlety with only five working steps between no attenuation and 100% attenuation. This is just an attempt to do something that leaves some degree of functionality that's better than nothing and more trouble free than a potentiometer.
I really appreciate the input.
You dont need 100% attenuation. 10dB of imbalance/adjustment will give the impression of the program material being completely sent to one side. Remember a balance control is for correcting minor imbalances and no recording or setup (unless you have loudspeakers in different rooms of course), will ever need more than about 6dB so rather go for resolution and a useful control.
The more I think about it, the more I like the idea of using an 11 position switch.
Use either 1 dB steps, 0 to -4 dB, and infinite attenuation.
Or 0.75 dB steps 0 to - 3dB, and infinite attenuation.
Now, you can do reasonable balancing (more than 3 dB or 4 dB . . . you have something wrong with your system, recording, or the room).
And you can do easy testing too.
Example, use a hard wire 1 to 2 adapter, and put Left channel from a CD player into both the Left and Right amp inputs.
A. You can easily test the sound of one amp channel and speaker at a time.
B. You can check the balance of your 'identical' speakers and 'identical amp channels.
C. You can compare the relative efficiency of 2 different model loudspeakers at a time.
D. And, you can do as many creative easy tests as you can think of.
Another advantage: Repeatability of settings.
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
Use either 1 dB steps, 0 to -4 dB, and infinite attenuation.
Or 0.75 dB steps 0 to - 3dB, and infinite attenuation.
Now, you can do reasonable balancing (more than 3 dB or 4 dB . . . you have something wrong with your system, recording, or the room).
And you can do easy testing too.
Example, use a hard wire 1 to 2 adapter, and put Left channel from a CD player into both the Left and Right amp inputs.
A. You can easily test the sound of one amp channel and speaker at a time.
B. You can check the balance of your 'identical' speakers and 'identical amp channels.
C. You can compare the relative efficiency of 2 different model loudspeakers at a time.
D. And, you can do as many creative easy tests as you can think of.
Another advantage: Repeatability of settings.
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
Last edited:
I'm debating between the "PRR option 1" of 1 db steps and the "PRR option 3" of mildly accelerating increments (leaning towards "3" but don't need to decide right away). Found a neat website that lets you actually try different relative db levels of test tones:
Blind Testing a 3 dB Level Difference
Blind Testing a 3 dB Level Difference
_I_ would include "dead" just for diagnosing odd problems or bad old recordings. Personal bad taste. If that function were included otherwise, I'd continue the series to the last tap, figure how short we are from 750k, and run that much to ground.
However: the whole rig in metal-film is what, $10? And there's just 5 connection to the amp. If seriously in doubt buy three or more switches, fit each one different, and try them all in the amp.
However: the whole rig in metal-film is what, $10? And there's just 5 connection to the amp. If seriously in doubt buy three or more switches, fit each one different, and try them all in the amp.