I recently constructed an LDR attenuator and volume control from Tortuga Audio. I was rather surprised and very impressed once I connected this up in my music system, replacing mono series resistor attenuators in each channel. The series resistor attenuator is a very purist connection between player output stage and amp input stage, a varied series of resistors and a switch contact. The music was good and I thought it would be hard to improve on.
I constructed the LDR unit mostly out of curiosity. I have read about them for years, but nevertheless a lot of things get talked up. But some times curiosity gets the best of everything.
Much to my surprise, I found that all the comments on LDR attenuators were not really exaggerating anything. The channel balance became more precise in that the music images became tighter and more focused from what already were tight and focused music images. Bass response was more pronounced and deeper. Dynamics were much more, as in the softs were softer and the louds were louder than before. The dynamics were the first thing I picked up on and then I began listening for other things.
Well that's all good and all, and then I immediately posted about the build here, trying to share my profound new enlightenment, as one is likely to do in such instances.
Now, I occassionally browse the threads about LDR attenuators and volume controls looking at what others have done, and reading all the arguments by those who think LDR's are all a bunch of rubbish and unfit for such an application and implementation such as this.
I really like the latter postings. It is clear from reading that the posters have no experience with LDR attenuators and volume controls. But the posters do seem to have profound knowledge about LDR's and why they are no good for this application.
But all the designers know the foibles of LDR's and their shortcomings. It's just that they know how to use them in an application for a specific purpose.
Theory be damned, it's the implementation that matters. And in the case of LDR's the implementations are really good.
For the curious or anyone wanting a simple project, constructing one of these is a real good use of time and money with benefits unforeseen.
While I went with Tortuga, there are other less expensive options. They all have their followers here.
I constructed the LDR unit mostly out of curiosity. I have read about them for years, but nevertheless a lot of things get talked up. But some times curiosity gets the best of everything.
Much to my surprise, I found that all the comments on LDR attenuators were not really exaggerating anything. The channel balance became more precise in that the music images became tighter and more focused from what already were tight and focused music images. Bass response was more pronounced and deeper. Dynamics were much more, as in the softs were softer and the louds were louder than before. The dynamics were the first thing I picked up on and then I began listening for other things.
Well that's all good and all, and then I immediately posted about the build here, trying to share my profound new enlightenment, as one is likely to do in such instances.
Now, I occassionally browse the threads about LDR attenuators and volume controls looking at what others have done, and reading all the arguments by those who think LDR's are all a bunch of rubbish and unfit for such an application and implementation such as this.
I really like the latter postings. It is clear from reading that the posters have no experience with LDR attenuators and volume controls. But the posters do seem to have profound knowledge about LDR's and why they are no good for this application.
But all the designers know the foibles of LDR's and their shortcomings. It's just that they know how to use them in an application for a specific purpose.
Theory be damned, it's the implementation that matters. And in the case of LDR's the implementations are really good.
For the curious or anyone wanting a simple project, constructing one of these is a real good use of time and money with benefits unforeseen.
While I went with Tortuga, there are other less expensive options. They all have their followers here.
Last edited:
I'm not sure what is meant by a "series resistor attenuator". Taken at face value it appears to mean simply inserting a resistor in series with the signal. If so, it would be hard to imagine a worse way to control volume (apart from a purely shunt resistor attenuator!).
Two good ways to control volume:
1. variable feedback - but may have problems with stability if not thought through carefully
2. potential divider made from linear resistance (i.e. pot or switched resistors)
An attenuator which acts as a tone control is poorly designed. An attenuator which affects signal dynamics is even worse. Appropriate value resistors are unlikely to do this, but an LDR might as LDRs are known to be less linear than a resistor.stvnharr said:
Two good ways to control volume:
1. variable feedback - but may have problems with stability if not thought through carefully
2. potential divider made from linear resistance (i.e. pot or switched resistors)
DF,
I assumed that the 3 types of resistor based stepped attenuators, series, shunt, and ladder were well known. I was using a series attenuator prior to installing the LDR unit. Perhaps my wording was not clear.
The sound of the music was improved in the ways I described, with better bass, extended dynamics and improved channel balance making for more precise music images.
Much of my music listening is to the big classical orchestral pieces, many of which feature wide dynamic range. In my initial comparison of the LDR unit v. the stepped attenuator unit, I noted that the dynamic range with the LDR unit was greater that with the stepped attenuator. While this difference was not much, it was noted and easily heard.
In listening to other musics, some which have significant single bass instrument parts, I noticed that the bass seemed a little louder than before and more precise. Again the difference wasn't much, but it was noted.
As for the channel balance, this is a little harder to describe, but the music images just seemed a little more real than before. Again, the difference wasn't much, but it was noted.
I was surprised at all of the above. I didn't expect to hear any differences. I was sceptical of all the comments I had read. I don't care about the wherefore and the why, I only care about what I heard. And I heard as described above.
Your comments on "tone control" are a bit baffling to me. I suppose if the assumption is that a resistor is the "neutral" and the standard reference then of course anything that differs from that standard could be considered a "tone control".
Don't change the subject of the thread with your comments on what makes a good way to control volume. Sure there are various ways to control volume, but I like the LDR way and that is why I started this thread. Perhaps you should listen to one some day, just for fun if nothing else.
OK.
If I replaced a linear attenuator with something even slightly non-linear then I would attribute any change in sound to the non-linearity. Apart, that is, from shifts on tonal balance which I would assume are due to different stray capacitances. Improvements in stereo image may be due to changes in channel crosstalk, cause either by stray capacitance or poor grounding. There is an engineering explanation for every genuine change in a signal; it is unlikely that a slightly non-linear component would be 'better' than a linear component in preserving fidelity.
If I replaced a linear attenuator with something even slightly non-linear then I would attribute any change in sound to the non-linearity. Apart, that is, from shifts on tonal balance which I would assume are due to different stray capacitances. Improvements in stereo image may be due to changes in channel crosstalk, cause either by stray capacitance or poor grounding. There is an engineering explanation for every genuine change in a signal; it is unlikely that a slightly non-linear component would be 'better' than a linear component in preserving fidelity.
Hi,
Well, it seems to come down to this. You read my words and make an educated guess about why I hear what I say I hear. This is based on the assumption that my resistor based series attenuator is more linear than my LDR attenuator. This is something that you cannot possibly know.
Based on what I hear, I'd say that the exact opposite is more likely to be true.
Here is my experience, I used a LDR in a couple of preamps, the sound was clean and clear. Out of curiosity I tried a ladder type attenuator. Also very clean and clear, but it had better base, dynamics, and depth, overall I liked the ladder better. The LDR and the ladder were both 50k. It might be a system synergy thing, but this is my experience.
PJN
PJN
Well system synergy is everything in audio.
Before the LDR, I had a high quality preamp that I had built. I used the series attenuators in that. Earlier this year I built a new, and slightly better amplifier. Out of curiosity I changed the preamp to be passive and just wired the attenuators direct. I thought that way was a wee bit clearer in the sound. That's when I started getting the idea of trying the LDR as I sold the preamp, minus the attenuators, so I had some available funding for the project. As stated above, I think the LDR unit to be a wee bit better that the resistor based series attenuator.
In reading the Lightspeed thread it seems that the LDR units, at least the Lightspeed, can be dependent on amp input impedance and player/dac output impedances.
Nonsense. But it doesn't matter in this case.
Your conclusion is based on subjective evaluation and, by definition, it's incontrovertible. There's no need for you to defend it against any type of objective measurement regime, or theory, or any other alternate explanation.
Dave.
I built a Tortuga based unit, not a mod.
Yes, I'm quite pleased with it.
I didn't make anything worse.
Yes, I enjoy!
Changing your attenuation method is a mod in my my book. Facts is facts, which is that a pot or switched resistors are more linear than LDRs. Unless there was something badly wrong with your old solution the LDRs cannot have increased dynamic range.
BUT: You are happy and enjoying music, which is what really matters and is good!
BUT: You are happy and enjoying music, which is what really matters and is good!
True, I cannot know that for certain. However it is overwhelmingly likely to be true. Unless you used astonishingly poor resistors in your series attenuator (such as old carbon comp which had drifted and gone noisy) then I can be certain that they are more linear than an LDR. I don't know why you persist in denying facts. You prefer the mild distortion of an LDR volume control; nothing wrong with that, but please don't pretend that a nonlinear resistance provides better signal fidelity than a linear resistance.stvnharr said:
I tried the LDR's and found there to be more problems than they solve. Specifically matching pairs as they vary wildly from sample to sample, getting them balanced when in the circuit, and huge drifts with the slightest temperature change rendered them worthless for a tube amp IMO.
A series string attenuator is also a pretty poor design as you have almost all the resistors in the signal path at low volumes. I.e. with a 24 step attenuator you would have 20 odd resistors in a string.
A ladder style attenuator is a far better design than either of those options as you only have 2 resistors in the signal path forming a resistor divider, then and each step of the attenuator switches to another set of resistors. I would take two quality metal film resistors over a string or LDR any day.
Also keep in mind the overall value of the attenuator could have a drastic effect on frequency response. I learnt this the hard way comparing a 100k and 50k attenuator in a tube amp, always compare like for like.
A series string attenuator is also a pretty poor design as you have almost all the resistors in the signal path at low volumes. I.e. with a 24 step attenuator you would have 20 odd resistors in a string.
A ladder style attenuator is a far better design than either of those options as you only have 2 resistors in the signal path forming a resistor divider, then and each step of the attenuator switches to another set of resistors. I would take two quality metal film resistors over a string or LDR any day.
Also keep in mind the overall value of the attenuator could have a drastic effect on frequency response. I learnt this the hard way comparing a 100k and 50k attenuator in a tube amp, always compare like for like.
Dave,
We are going at this from different perspectives. I've read enough of your posts to know where you are coming from. May I ask one thing? And that is, are you familiar with the Tortuga controller in any way? It is significantly different from the Lightspeed, of which I know you are familiar. I tend to think that your perspective is that all LDR's are defective no matter how utilized and the application is irrelevant. FWIW, I know nothing of the Lightspeed other than reading the thread here on diyaudio.
I am coming at this completely differently. I'm coming at this from a listener's perspective. I find that the bass response with the Tortuga is better than with the series attenuator. I find the increase in dynamic range with the Tortuga to be a significant improvement in much of the music I listen to most of the time. I also find the cleaner, clearer music images with the Tortuga to be more like real acoustic instruments playing music. I go to concerts regularly and have always been impressed with how clean and clear the music sounds, especially in comparison to electronic playback.
Now what I have described may be meaningless to you.
Likewise it is simply irrelevant to me the why's of why the music now sounds more like real instruments playing music. I'm in it solely for the enjoyment.
Thanks for replying here. I always knew that there just had to be some folks out there who didn't have the near uniformly positive reception that seems to be quite common with the Lightspeed and others.
FWIW, the unit I use is not based on matched pairs of LDR's. I am quite familiar with ladder attenuators, though I have never used one. I do have a 20k shunt attenuator however. I may eventually do an A/B with this when I get the time.
Last edited:
HI Andrew,
To set the record straight. I did not modify anything.
I exchanged a passive unit with a pair of mono series attenuators for the Tortuga unit that I wired up from Tortuga parts as per a thread I posted in this forum at the end of October.
I have not used an active preamplifier in several months.
BTW, when did you sour on the Lightspeed? You used to be a big supporter. I'm going thru that long thread completely and 5-6 years ago you were helping people in their diy efforts. I've read the first 300 pages, so perhaps I will come to what I am asking you.
Oh dear ... you ...
Try to understand, why different, how are resistaces you put into the signal path tied to bass response etc. Ohterwise your thoughts carry very little value.
I did quick-and-dirty LDR: fixed value series around 10 kohm and changing shunt. Working between 300 ohm output and 100 kohm input. No capacitors in signal path. No changes in bass response. Sound is cleaner ... and I consider it to be more correct because of this, though I don't think about it in absolute terms. Just like it more, period.
Try to understand, why different, how are resistaces you put into the signal path tied to bass response etc. Ohterwise your thoughts carry very little value.
I did quick-and-dirty LDR: fixed value series around 10 kohm and changing shunt. Working between 300 ohm output and 100 kohm input. No capacitors in signal path. No changes in bass response. Sound is cleaner ... and I consider it to be more correct because of this, though I don't think about it in absolute terms. Just like it more, period.