Can amplifier speed and resolution be measured?

[tomchr]

Probing for resistance with signal voltage present can create distortion in itself, as the meter despite its high impedance is a parallel in the case of the series pairs input to output. A basic set up would assess the attachment of measuring equipment by having music playing to see said equipment and its many settings was not the actual distortion contributor, caused by being a parallel.

I think we may be talking slightly past each other here.

Parameters such as resistance, temperature coefficient, and voltage coefficient are all measured on the resistor in isolation. To get the resistance and voltage coefficient, you measure the I/V curve of the resistor. To get the temperature coefficient, you measure the I/V curve at various temperatures. You can see an example of this in the paper I linked to in Post #64. You can then develop a SPICE model that matches the measurements and use it in simulation to determine the performance of any circuit that includes that resistor. A circuit as simple as an LDR-based volume control (two resistors) would be trivial to simulate and as long as the resistor model is reasonably accurate, it's reasonable to expect the simulation results to match reality.

You can measure the resistance with the music playing by measuring the voltage across the resistor, the current through it, and calculating the derivative. That will give you the same result as just measuring the I/V curve, but is much more complicated to set up and replicate, which is why we use source meters and such rather than music players for resistor characterization. Similarly for the AC parameters of the resistor. Swept sine wave vs music will ultimately yield the same parasitics (which can also be included in the simulation model).

Tom

 

[dragonspit4]

Yeah, in perceptible terms resolution as used in photography or video doesn't fit much with hearing.

We all know and can see what happens if you in step wise fashion reduced resolution of an image. What is the equivalent of step wise resolution reduction of a sound file? Would it be reducing frequency response gradually? Would it be reducing dynamic range gradually or similarly letting noise levels grow till it obscures sound? Would it be some combination of reducing dynamic range and frequency response?

Of course if any of that is true the real issue is we have all the dynamic range and frequency response already that any human can hear. So called hires might be higher resolution, but only in a measured sense. Not at all in a perceptible sense. Just like at some viewing distance photo resolution exceeds the eye's angular resolution, and increasing it more does nothing to the seen image.

I thought lower resolution is caused by distortion.(I could be wrong)
When I compare my tube amp to my solid state amp, they definitely sound different.
Not all amp sound the same.
As soon as u have different sound, u can measure it for sure.
If all amp sound the same, tube amp wouldn't exist.

 

[amirm]

Yes friendly is important. I hope this applies to everyone.

No. Not everyone is situated the same. I explained that. Please focus your posts on new information than protests.

 

[solderdude]

The LDR pre will most likely measure similar to the one tested by Tom. The reason for this is that the used LDR's are the same and the principle is the same, namely 2 LDR's in the signal path and no other components as shown below on the right.

The only real difference is wire routing and the way the LED's are driven.
When this is done with less noise one gets less noise in the audio as well as the LED intensity modulates the signal (music or test).

What might be different with other LDR circuits is that the LED currents used are either higher (becomes sort of an equivalent of a lower resistance potmeter) or lower currents are used (becomes the equivalent of a higher resistance potmeter). This could result in different noise levels and FR behaviour.

A difference with LDR versus a normal potmeter is that the input resistance changes with the volume setting. This can be much more substantial than using a normal pot. The reason is that both LDR's (that are effectively in series when it comes to the source) vary in resistance.
In a normal pot the resistance only changes slightly and depends on the ratio between the pots own resistance and that of the load at the wiper.
In a well designed circuit the input resistance of a pot varies just a few %. In case of the LDR this can be many factors !
The input R of the circuit could potentially vary between a few hundred Ohms and Mega Ohms !
This can have an influence on the source's output behavior (R out can be anything between between 1 and a few hundred Ohm) and the load circuit.
Combined with the capacitive load of the amp + RCA the FR might change with settings.

That's why I mentioned it should be tested with a 'normal' RCA cable connected (loaded with realistic capacitance) and with say a 10k resistive load.
Possibly also with a 100k load and suggested to measure the variance of the input resistance with a few potmeter settings.

Perhaps also measure a 'normal' potmeter (like the Schiit Sys for instance) as a comparison... More workload for Amir

The whole discussion about it sounding 'better' can be ended using a null tester and real music and real loads.
When the null produces differences the sound is altered. Period.
And it will be altered as there will be little to no difference to the one Tom measured simply because the only 2 active components in the audio path are the same... the LDR's. There is no magic involved.

Edit: Sometimes I see attenuators being called 'passive pre-amplifier' but from a technical standpoint these devices only attenuate so 'attenuator' is a better description. Passive pre-amplifiers do exist but use (auto)transformers to boost the voltage at the expense of other properties.

 

[rebbiputzmaker]

The LDR pre will most likely measure similar to the one tested by Tom. The reason for this is that the used LDR's are the same and the principle is the same, namely 2 LDR's in the signal path and no other components as shown below on the right.

The only real difference is wire routing and the way the LED's are driven.
When this is done with less noise one gets less noise in the audio as well as the LED intensity modulates the signal (music or test).

What might be different with other LDR circuits is that the LED currents used are either higher (becomes sort of an equivalent of a lower resistance potmeter) or lower currents are used (becomes the equivalent of a higher resistance potmeter). This could result in different noise levels and FR behaviour.

A difference with LDR versus a normal potmeter is that the input resistance changes with the volume setting. This can be much more substantial than using a normal pot. The reason is that both LDR's (that are effectively in series when it comes to the source) vary in resistance.
In a normal pot the resistance only changes slightly and depends on the ratio between the pots own resistance and that of the load at the wiper.
In a well designed circuit the input resistance of a pot varies just a few %. In case of the LDR this can be many factors !
This can have an influence on the source's output and the load circuit.
Combined with the capacitive load of the amp + RCA the FR might change with settings.

That's why I mentioned it should be tested with a 'normal' RCA cable connected (loaded with realistic capacitance) and with say a 10k resistive load.
Possibly also with a 100k load and suggested to measure the variance of the input resistance with a few potmeter settings.

Perhaps also measure a 'normal' potmeter (like the Schiit Sys for instance) as a comparison... More workload for Amir

The whole discussion about it sounding 'better' can be ended using a null tester and real music and real loads.
When the null produces differences the sound is altered. Period.
And it will be altered as there will be little to no difference to the one Tom measured simply because the only 2 active components in the audio path are the same... the LDR's. There is no magic involved.

Load matching is important. That might be why people often hear differences with different preamps placed between dacs, power amps etc. Do you have any opinions on the use of transformers or autoformers for this application? Since we will be measuring some pre-chosen input / output impeadence values, can the results be considered universal?

 

[solderdude]

Transformers and autotransformers all limit FR and add distortion (hysteresis of the metal core).
To some this may sound 'pleasant' but for technical folks it is clear that it changes the input signal in many ways and not just amplitude which is the goal of a perfect attenuator.
Good line level transformers can have very low distortion numbers and a wide enough FR range for audio applications.
Remember, most audiophiles are older folks that rarely hear above 14kHz and don't play loud. Most listen to speakers which rarely produce 'flat' down to 20Hz (which by itself is easier to feel than hear)

The same is true for the LDR. It changes a lot more aspects than just the amplitude and are even 'worse' than good transformers.
It adds signals that aren't in the, to be attenuated, signal.
To some this may sound 'more pleasant'. Just like vinyl-, tape- and tube-added distortions can sound more pleasant to some.
This does not mean it is 'better' in fidelity, it just means it is altered in a 'pleasant' way to some listeners.

When one measures with 10k and 100k loads the extremes of most amps will be covered as that is the range most amps fall into. Tube amps sometimes may be higher up to 1M but won't make any meaningful differences with 100k which already isn't much of a load anyway.

Source resistances could be into many kOhm for some tube output sources. These would definitely be attenuated differently. Most sources will be below 100 Ohm. Remember the LDR volume control will have a very wide range of load resistance to a source (between a few hundred Ohms and MegaOhms) which isn't a good property for most loads as output stages of sources will have different output currents with the same output voltage.
The output stage of the AP will most likely not be affected BUT may not be so for many other sources.
The LDR volume control is a very poor (as it is changing per volume setting) load to any source.
The output R of the attenuator will change less as one resistor goes up and the other goes down in resistance and these are in parallel from the load's point of view.
The output Resistance of a normal pot will have about the same 'resistance variance' as that of the LDR (not exactly the same).

When a source for instance has an output capacitor of say 2.2uF the low load resistance of the LDR could really change the FR substantially which would not be shown by the AP (as it doesn't have one but certainly tube audio gear could).
When the input R for instance drops to 1k Ohm the -3dB point of such a source would drop to an audible -3dB @ 72Hz. In most cases this means folks could find bass subjectively 'tighter'.

 

[March Audio]

I have only skimmed in on this so probably missed pertinent points, but am I alone in being completely bemused by this whole LDR thing? Why would you go there? A complex problematic solution that just doesn't work as well as the most simple solution.

 

[SIY]

Why would you go there? A complex problematic solution that just doesn't work as well as the most simple solution.

Your latter answers your former. This is the world of audiophilia, where the more complex and exotic, the better, irrespective of mundane things like performance. The story is more important than the technology.

 

[March Audio]

You are of course correct, I was foolishly thinking rationally. Big mistake

 

[Blumlein 88]

Your latter answers your former. This is the world of audiophilia, where the more complex and exotic, the better, irrespective of mundane things like performance. The story is more important than the technology.

At one time the high end fashion was simpler is better. Remove tone control, remove balance, use simpler circuity. Etc.

With digital it seems to have turned around. The more complex the better. The more imagined ills fixed the better. Certainly there is no end to the market for imagined problems. Just look at trying to use a USB cable.

Analog was assumed good so the less done the better. Digital is assumed a problem in need of fixing. So the more fixed the better.

 

[stereo coffee]

I have only skimmed in on this so probably missed pertinent points, but am I alone in being completely bemused by this whole LDR thing? Why would you go there? A complex problematic solution that just doesn't work as well as the most simple solution.

Until you actually try it.

 

[SIY]

Until you actually try it.

And then you find out that it's a less good solution than conventional methods which are audibly transparent, and much cheaper.

 

[stereo coffee]

And then you find out that it's a less good solution than conventional methods which are audibly transparent, and much cheaper.

..but if you have no experience, or much worse no wish of trying each method... suggest politely you gain that experience

 

[SIY]

Major assumption. And incorrect, of course.

 

[Frank Dernie]

..but if you have no experience, or much worse no wish of trying each method... suggest politely you gain that experience

I am interested in the measurements but as an engineer do believe that a knowledgeable engineer can rule out certain designs as not worth trying from knowledge of the components.
It has saved me lots of time and money, personally.
The salient point (IMO) in any case is that the conventional designs properly implemented are already transparent so why do something different - unless it will be much easier to manufacture and/or cheaper?
Reading between the lines it looks like these devices are prone to non-linearity which may make them add quite a lot of distortion if not very carefully implemented.
I am afraid it looks like an eccentric solution to a non-problem so far.
I await the measurements patiently...

 

[rebbiputzmaker]

At one time the high end fashion was simpler is better. Remove tone control, remove balance, use simpler circuity. Etc.

With digital it seems to have turned around. The more complex the better. The more imagined ills fixed the better. Certainly there is no end to the market for imagined problems. Just look at trying to use a USB cable.

Analog was assumed good so the less done the better. Digital is assumed a problem in need of fixing. So the more fixed the better.

Well not everyone thinks that way. I still think a simple chain is better. I do not see tone controls on good gear etc. I use dsp for home theater but not for two channel system.

 

[tomchr]

..but if you have no experience, or much worse no wish of trying each method... suggest politely you gain that experience

Rather than putting the onus of demonstrating the wonders of LDR volume controls on the listener, how about presenting evidence for its greatness? "You gotta hear to believe" is not a convincing argument, especially not in the face of the scientific evidence against the LDR-based volume controls.

If you truly believe that the only way to show the superiority of your circuit is by a listening trial, I suggest setting up such a trial. A scientifically valid one that is. Double-blind, significant sample size, controlled environment, statistical controls, etc. See Olive & Toole for examples. If a statistically significant proportion of your sample population prefers the LDR volume control over a regular volume pot, you have actual data. Until then, all you have is opinion.

Perception is the cognitive experience of a stimulus. It is perfectly possible for two completely identical stimuli to generate two different cognitive experiences. One such way is to tell the listener that, "this box is special". Another way is to tell the listener, "this box is really expensive". It gets even more distorted once someone has forked over significant cash for a product ($50k speaker cables anyone?). That's well-documented in psychology.

Extraordinary claims should be backed by extraordinary evidence.

Tom

 

[March Audio]

Until you actually try it.

I don't need to try it. You can examine ideas and assess their efficacy. it's obviously an inferior design choice for multiple reasons already outlined by Frank above.

 

[stereo coffee]

I am interested in the measurements but as an engineer do believe that a knowledgeable engineer can rule out certain designs as not worth trying from knowledge of the components.
It has saved me lots of time and money, personally.
The salient point (IMO) in any case is that the conventional designs properly implemented are already transparent so why do something different - unless it will be much easier to manufacture and/or cheaper?
Reading between the lines it looks like these devices are prone to non-linearity which may make them add quite a lot of distortion if not very carefully implemented.
I am afraid it looks like an eccentric solution to a non-problem so far.
I await the measurements patiently...

So can you list the components, you have concern about ?
I can assure its much harder to make - totally and absurdly contrary to normally seen practices.

 

[solderdude]

The components 'we' are concerned about are the LDR's as the LDR itself is also voltage dependent and thus adds distortion.
Not the LED's inside nor the way they are driven.
Added distortion may sound great to some, but is a loss in fidelity for 'us'.

Let the measurements tell the story.
The 'volume controlled signal reaching your amplifier is pristine because it has not been altered' may not be entirely true.
My guess.... its measurements may be very much alike those measurements Tom made.
That possibly could mean one of Amir's famed 'not recommended' labels may be dangling of one of the pink panther figurines atop.


Fortunately this might be good for your business as a 'not recommended' from Amir means it is generally loved to bits by the subjective crowd which after all is your audience. The ones believing in this kind of thing don't visit ASR anyway or do but frown on most that is said here.

Why haven't you grabbed a PC with an ADC/DAC card and run basic measurements yourself b.t.w. ?