A long time ago there was an article in The Audio Amateur, or its posterior siblings, on a potentiometer using I think 3 pots that allowed very fine level setting.
If I am not wrong one pot set the coarse level, and the other two the fine levels, working both as a balance and fine step arrangement.
As we can now find good quality 21-step pots on eBay for a cheap price, I was thinking of building one.
Am I right on this recall or my memory is failing?
Is there a similar project in the web somewhere, maybe here on DIYAudio?
Carlos
If I am not wrong one pot set the coarse level, and the other two the fine levels, working both as a balance and fine step arrangement.
As we can now find good quality 21-step pots on eBay for a cheap price, I was thinking of building one.
Am I right on this recall or my memory is failing?
Is there a similar project in the web somewhere, maybe here on DIYAudio?
Carlos
I think this is the one Bill was talking about - one coarse control on both channels, and a fine control for each channel:
NEW! The New-Old TCJ Stepped Attenuator
NEW! The New-Old TCJ Stepped Attenuator
Yes. I think the original Old Colony project was something like that.
Unfortunately it demands one of the pots to be shunt type and the other series type, and these eBay pots are series type only.
The other thing is that you need four poles to make a shunt pot, two for each channel. These eBay pots are only two pole.
Unfortunately it demands one of the pots to be shunt type and the other series type, and these eBay pots are series type only.
The other thing is that you need four poles to make a shunt pot, two for each channel. These eBay pots are only two pole.
Last edited:
Yes, that is a ladder type. With it you choose between different resistance dividers at each position with only 2 resistors in the circuit at any position.
A shunt is strictly a variable resistance. You can use a series attenuator by connecting the bottom to ground, and the wiper to the output, leaving the top unconnected or also connected to the output. It then acts as a variable resistance instead of a voltage divider.
You can use 2 attenuators wired as variable resistors in a series-parallel configuration to do exactly what you are looking for. That lets you adjust both halves of the overall voltage divider independently. The biggest drawback is both the input and output impedance changes with the position, but that can be dealt with if you choose the values wisely.
Best, Bill
A shunt is strictly a variable resistance. You can use a series attenuator by connecting the bottom to ground, and the wiper to the output, leaving the top unconnected or also connected to the output. It then acts as a variable resistance instead of a voltage divider.
You can use 2 attenuators wired as variable resistors in a series-parallel configuration to do exactly what you are looking for. That lets you adjust both halves of the overall voltage divider independently. The biggest drawback is both the input and output impedance changes with the position, but that can be dealt with if you choose the values wisely.
Best, Bill
Standard attenuators are usually a resisting film (oxide) where a wiper slides on, so it's really a resistor divider. At the wiper end there's always two resistors: one series/one parallel. That combination sets the attenuation.
There are several problems in attenuators that have to be considered, the first one being the quality of film track and wiper contact. That can be better or great in expensive types.
Film tracks wears off and may compromise audio quality. And it's where step attenuators are better.
Series attenuators problems is that they put all the resistors and solder joints in the signal path, and it's also something to consider.
All pots, by definition and according to how you connect them, will present a variable impedance to the source output or the (amp/pre) input. That's why you need a buffer that will not be loaded by it in a serious way.
If you don't use a buffer or intermediate stage you may have only one or two pot positions where that is not a problem.
I have been using passive interfacing for a long time, and it's more transparent than an active one in many ways, but not the best solution perhaps. I'm willing to listen to alternatives, simple ones. Particularly simple buffers, interfaced with quality affordable pots.
Carlos
There are several problems in attenuators that have to be considered, the first one being the quality of film track and wiper contact. That can be better or great in expensive types.
Film tracks wears off and may compromise audio quality. And it's where step attenuators are better.
Series attenuators problems is that they put all the resistors and solder joints in the signal path, and it's also something to consider.
All pots, by definition and according to how you connect them, will present a variable impedance to the source output or the (amp/pre) input. That's why you need a buffer that will not be loaded by it in a serious way.
If you don't use a buffer or intermediate stage you may have only one or two pot positions where that is not a problem.
I have been using passive interfacing for a long time, and it's more transparent than an active one in many ways, but not the best solution perhaps. I'm willing to listen to alternatives, simple ones. Particularly simple buffers, interfaced with quality affordable pots.
Carlos
With a standard resistive pot used as an attenuator, the POT output resistance/impedance is usually chosen to be at least 5times smaller than the receiver's input impedance. <20times is even better.
Using a receiver with Rin=50k, then a pot output impedance of ~5k would achieve a 10:1 ratio.
A 20k pot will give that maximum output impedance when Rs is kept low.
Now let's consider what effect varying the attenuation has on the resistance/impedance seen by the Source.
At zero volume (= infinite attenuation), the source sees the 20k pot in parallel with the first cable capacitance, i.e. 20k//100pF
At maximum volume (= 0dB attenuation), the source sees the 20k pot // first cable capacitance // Rin // second cable capacitance. If we assume the high output impedance of the pot requires a short (low capacitance) cable to be chosen for the connection between pot and receiver, then let's allow 50pF for this second cable.
The source sees:
20k//50k //100p//50pF,
the equivalent full volume loading is 14k3//150pF
If a source can drive a 20k pot properly then there must be something seriously wrong with the source if the sound quality changes audibly by reducing the loading to 14k3.
Personally, I'd choose a 10k pot for Rin=50k. The range of loading seen by the source is 8k3 to 10k.
The 8k3 loading requires an extra 210uA of peak source current (@~2Vac maximum output) compared to driving a 20k load. I don't see that 60uApk (20k to 14k3) causing any problem of "a variable impedance to the source output"
Last edited:
Also, getting back to the attenuators themselves, you can also wire them to put only one resistor in the circuit instead of a string by soldering the proper attenuation values to the switch terminals then wiring the free ends together. Now you merely switch between resistors instead of using a string of them. It's very simple. For the coarse series attenuator section you could probably use a switch with only 4-6 positions.
Best, Bill
Best, Bill
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.