Sledgehammer to crack a Lightspeed walnut

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I liked the idea of the Lightspeed volume control, but couldn't live with the channel imbalance.

I liked the idea of a stepped attenuator, but couldn't live with all those dirty contacts.

Why choose? I decided to marry the two technologies by making 4 banks of 24 presets, operated by a 24 pole / 4 way switch which could then adjust the LDR voltages precisely, at each stage, whilst monitoring the output. I managed to get perfect channel balance and a perfect log taper. I also like the idea of being able to dial in different tapers.

This is a video of it in action:

https://www.youtube.com/watch?v=0Px0BDsCEL8
 
Not the same but similar. If those contacts are dirty and the ldr control voltage is jumping around what do you think will happen to the audio output?

'Jumping around' ? I suspect the problem would be more about an increase of capacitance / resistance.

Just to recap then:

All those who think a stepped attenuator is the best solution for a volume control are wrong.

All those who think an LDR is the best solution for a volume control are wrong.

A normal potentiometer is the best solution for a volume control.

Blimey.

(Have you ever looked at how they produce a logarithmic taper in pots?)
 
A dodgy contact feeding an LED+LDR has a similar effect on the LDR resistance as a dodgy contact has directly. LEDs react very quickly to current changes; LDRs are perhaps a little slower but you will still get LF intermodulation. If you don't trust switches don't use them.

number7 said:
All those who think an LDR is the best solution for a volume control are wrong.
Definitely. Unless they prefer a little distortion.

The best solution is a good quality pot, used in the correct circuit.
 
A pot track is a reasonably good resistor, so more linear than any LDR. Provided that the end connections are well made, any non-linearity in the track itself will not matter too much as it is a potential divider. The track-slider interface is the problem, but that can be dealt with by ensuring that the slider sees a high impedance so little signal current crosses the interface.

Those who are worried about resistor linearity will prefer to avoid a pot and use high quality resistors in a switched attenuator. The downside of this is that you need to keep all DC off the switches to avoid clicks, so you need coupling caps (and possibly ground leaks) both sides. People who worry about resistor linearity are sometimes Faradaphobes too, so they have a problem.

People who don't like switches or pots can use LDRs, but then have to accept some non-linearity - much worse than any resistor or pot. However, if signal levels are kept low then there won't be too much distortion.

So you choose which set of problems to avoid, but you also unavoidably choose which set of problems you can tolerate.
 
A pot track is a reasonably good resistor, so more linear than any LDR. Provided that the end connections are well made, any non-linearity in the track itself will not matter too much as it is a potential divider. The track-slider interface is the problem, but that can be dealt with by ensuring that the slider sees a high impedance so little signal current crosses the interface.

Those who are worried about resistor linearity will prefer to avoid a pot and use high quality resistors in a switched attenuator. The downside of this is that you need to keep all DC off the switches to avoid clicks, so you need coupling caps (and possibly ground leaks) both sides. People who worry about resistor linearity are sometimes Faradaphobes too, so they have a problem.

People who don't like switches or pots can use LDRs, but then have to accept some non-linearity - much worse than any resistor or pot. However, if signal levels are kept low then there won't be too much distortion.

So you choose which set of problems to avoid, but you also unavoidably choose which set of problems you can tolerate.

Do you actually read other people's posts? The whole point of what I did was to completely get rid of any LDR non-linearity!

Like it or not, audio is subjective. A few months ago I saw Beck in concert, the sound quality was simply appalling. Overblown bass, massive mid/high distortions etc - it was almost unbearable. Now if you asked some of the crowd who were severely inebriated and jumping up and down, they thought it sounded wonderful. Perceptions.

Your hi-fi system is good? No, it's rubbish - if you put it in a tiled bathroom or an anechoic chamber. Your listening room will fall within those two extremes, but unless you have access to FFT analysis tools, your room will have a far bigger affect on the resulting sound quality than any pot you care to mention. If, like me, your large flat TV is positioned between your speakers, you will be getting all sorts of crosstalk, phase anomalies, flutter echoes etc. Try putting a bath towel over the TV and see (hear!) the difference.

In my system, in my room, with my ears, a stepped LDR attenuator is the best solution. If you prefer a pot - fill your boots:)
 
number7 said:
Do you actually read other people's posts?
Sadly, yes.

The whole point of what I did was to completely get rid of any LDR non-linearity!
You may be confusing two quite different issues:
1. linearity of volume control action
2. linearity of signal voltage

People selling LDR volume controls sometimes allow confusion of these two quite different issues to occur, so I suppose it is not surprising that some users confuse them too. You can linearise the control action by using a sufficiently sophisticated LED driver system. This makes absolutely no difference whatsoever to the LDR signal voltage nonlinearity; it is this which create signal distortion.

Audio is subjective. Hi-fi is much less so. If you like the sound of LDR then so be it. I prefer a volume control with less sound, ideally none at all.
 
An LDR is a variable resistor. The resistor value is set by light rather than a control knob. By adjusting the light very carefully, you can get the resistance change to follow any law you want so you can have linear or logarithmic etc.

Unfortunately an LDR is also a non-linear resistor. The resistance depends on signal voltage, much more than for an ordinary fixed resistor. LDR manufacturers often give figures or graphs for this. Put an audio signal through a non-linear resistance and you will get distortion. If the signal level is sufficiently low then the distortion will be low, but you can't go too low as then noise becomes a problem. Hence, like everything in electronics, an LDR volume control is a compromise. It will have much less distortion than a typical loudspeaker, but much more distortion than a typical amplifier. Some people are happy with this, which is fine provided they don't try to claim that an LDR volume control is somehow more accurate/transparent or less distorting than other volume control methods.
 
Founder of XSA-Labs
Joined 2012
Paid Member
Are we talking about non linear resistance as a function of frequency? That will cause harmonic distortion to arise out of non linear components. Non linear behavior of amplitude vs knob position just gives inaccurate volume amplitude but should not distort the signal.

Some resistors (materials) are more non linear than others. Carbon composition vs metal thin film for example are different.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.