Can amplifier speed and resolution be measured?

[BillH]

How did we get into the topics of LDRs???

Anyway, here are measurements of one from own member @tomchr : https://www.neurochrome.com/tortuga-audio-ldr3/

Yippee!
We have data and a very nice narrative.
Thank you Amir

 

[amirm]

There are many differences one LDR offering to the next, which just goes to show they are a device needing
much needed further exploration - and never giving up on.

Well, then send one in and I will measure it!

 

[stereo coffee]

Hi Amir
Great will do. can you send me a PM to let me know where to send please

 

[tomchr]

Showing as I see it, digital techniques of driving the anode and cathode are not kind to measured performance.

An LDR volume control is a resistive divider formed by one or two LDRs. As with any resistive divider, the distortion in the divider stems from the temperature coefficient and voltage coefficient of the resistive material used. The temperature coefficient can be dealt with easily by sizing the resistor so that its self-heating is minimized (large resistor and/or low power dissipation). The voltage coefficient depends on the resistor material, thus, can only be dealt with by selecting a resistor type with a low voltage coefficient. That's why we use metal film resistors in attenuators...

The voltage and temperature coefficients of the CdSe film used in LDRs are quite horrid. See example here: https://www.researchgate.net/figure...Se-ITO-thin-film-heating-cycle_fig6_320273333
Also note Fig. 11. You can download the full paper here: https://www.researchgate.net/profil...rochemical-Cell.pdf?origin=publication_detail

Now, obviously the paper linked to above is not a measurement of the LDRs used it Tortuga's circuit, but it is a measurement of an LDR with similar chemistry, thus, indicative of the performance that can be expected from Tortuga's LDRs. CdSe is just not a good resistor material for an attenuator. At least not if you want low distortion.

Driving the LED to illuminate the LDR is easy. I find it hard to believe that Tortuga should have screwed that up.

Tom

 

[Blumlein 88]

I messed with these a long time ago. When lesser components were available. They were rather too many issues for me to be happy with them. Ended up with something like a binary ladder of resistors to ground that could give lots of small steps. But I'd like to see this tested (as apparently will happen) to see what has been achieved.

To the guys posting about them, don't be too put off by the hard line, skeptical questions. That is sort of what this place is about. Also when it comes down to it, we'll test things and see what is what as much as is possible.

 

[amirm]

Hi Amir
Great will do. can you send me a PM to let me know where to send please

Message sent....

 

[stereo coffee]

Received thanks, and working on building one up for you to send as soon as I can

 

[solderdude]

interesting read.

From the descriptions in the DIYaudio thread it seems that poor sound quality originates from mechanical contacts (wipers and switches) and not from non-linearities, these seem to improve sound quality. You learn something new every day.

How can added distortion contribute to 'better' sound reproduction when compared to no added distortion ?
I do see that this attenuator can sound different from a 'regular' volume control.

I figure it won't measure that different from the one Tom measured as the principle (even the used LDR ?) is the same.
Difference is that it seems Chris's version also is a passive mixer, like the Tortuga LDR6, but with 3 inputs.
It probably only differs in the LED driver and used LED currents. That latter part may be interesting.

Will wait for Amir's findings.

Things that could also be interesting to measure outside of the usual suite of measurements:
1: Input resistance vs set attenuation level (expect it to vary).
2: FR at different volume settings while connected to an actual load (RCA cable of normal length + typical amplifier load) up to 50kHz.

 

[Lbstyling]

Davids Life Beyond 20khz article http://wilson-benesch.com/reviews/Life_Beyond_20kHz_Blackmer_SVC_Sep-1998.pdf
gives insight that there is with our auditory system ability to perceive - not necessarily directly hear
frequencies above 20khz, that influence the frequencies we can hear. Suggest you have a read of the article, it is quite interesting.

To offer an alternative opinion on this, he compares soft domes to hard with equivalent FR. Power response (off axis SPL at different frequencies) will be very different though, as will distortion spectra.

There is no way 2 domes should sound like one another when eq'ed flat anyway. Wilson well knows this. Read Toole.

 

[Sir Sanders Zingmore]

What is this thread about again?

 

[solderdude]

LDR volume controls and dome tweeters I suppose ?

Ohh... my mistake ... it says: "Can amplifier speed and resolution be measured ?"
(Someone must have derailed it...)

Well yes it can, just not the 'amplifier speed and resolution' as it is 'described' by (mostly non technical) individuals that use these words to describe something they don't know how to define based on what their ears have told them must be there.

 

[GGroch]

Two additional pejorative listening phrases I also see applied to (especially lightweight) amplifiers (JDS Atom & most Class D as examples) are "Weight" and "Slam".

The earlier referenced "Life Beyond 20kHz" article may allude to this. "Many audio signal sources have extremely high transient peaks, often as high as 20 dB above the level read on a volume indicator........Many of the available peak reading meters do not read true instantaneous peak levels but respond to something closer to a 300 IJ.S to 1 ms averaged peak approximation. All system Errors in reverberant components, including power amps and loudspeakers. should be designed to reproduce the original peaks accurately."

Instantaneous Peak or Dynamic power....headroom, used to be included in measurement of most amplifiers. I am not seeing it mentioned in current amp specifications or measurements. Is this no longer a consideration? How do we know my 9.7 Oz Atom has enough of it?

 

[andreasmaaan]

Instantaneous Peak or Dynamic power....headroom, used to be included in measurement of most amplifiers. I am not seeing it mentioned in current amp specifications or measurements. Is this no longer a consideration? How do we know my 9.7 Oz Atom has enough of it?

I agree that peak power would be good to know, but even if you assume that the peak output is no higher than the steady-state output, the Atom almost certainly has enough of it. Let's say you have a pair of headphones with an input impedance of 300 Ohm that requires 0.2 V to reach 90dB RMS (this is approximately what's needed by something like the HD650, which is considered difficult to drive by some). With the Atom's 8+ V of output into 300 Ohm, you can expect to reach over 120dB RMS without clipping.

Music with 20dB of dynamic range is considered very dynamic. If your RMS listening level were 100dB, you'd be easily able to hit the 120dB peaks with the Atom. 100dB RMS is extremely loud (dangerously so, even). So you could be satisfied that with this combination, you'd have plenty of power, even with the most dynamic of recordings.

 

[maxxevv]

Not forgetting that the impedance values of headphones such as the HD650 and HD800 varies with frequency too.
And as far as I recall, the HD650 hits a peak somewhere north of 500 Ohms, while the HD800 in the 650-700 Ohm range.

So dynamic range aside, the power for an amp probably has to cater for these transient peaks in voltage too?

 

[andreasmaaan]

Not forgetting that the impedance values of headphones such as the HD650 and HD800 varies with frequency too.
And as far as I recall, the HD650 hits a peak somewhere north of 500 Ohms, while the HD800 in the 650-700 Ohm range.

So dynamic range aside, the power for an amp probably has to cater for these transient peaks in voltage too?

That's a fair point. I don't think it would be significant in this case since we're already talking about levels much higher than any sensible person would listen at, plus there isn't a huge difference between 300 and 500 Ohm, but in general it's something to take into consideration.

 

[SplitTime]

interesting read.
Will wait for Amir's findings.

Things that could also be interesting to measure outside of the usual suite of measurements:
1: Input resistance vs set attenuation level (expect it to vary).
2: FR at different volume settings while connected to an actual load (RCA cable of normal length + typical amplifier load) up to 50kHz.

I’d add to the other interesting things to potentially measure (if possible)
3: temperature sensitivity

Many of the devices used to make these end up having strong temperature sensitivies (carrier_generation(T), resistance(T), and noise(T), which can also change V and other sensitivities). Maybe 20C - 40C difference (stacked unit in a garage with poor ventilation, compared with inside desktop in AC)... Maybe only do this if the initial measurements look reasonable at 25C? Just an idea... ;-)

 

[rebbiputzmaker]

I messed with these a long time ago. When lesser components were available. They were rather too many issues for me to be happy with them. Ended up with something like a binary ladder of resistors to ground that could give lots of small steps. But I'd like to see this tested (as apparently will happen) to see what has been achieved.

To the guys posting about them, don't be too put off by the hard line, skeptical questions. That is sort of what this place is about. Also when it comes down to it, we'll test things and see what is what as much as is possible.

What did you try, been using ldr lightspeed for many years and have been happy with it.

 

[Blumlein 88]

What did you try, been using ldr lightspeed for many years and have been happy with it.

These were DIY efforts with light sensitive resistors. I've not used any manufactured products featuring this. So I probably designed a lousy way to do it. Based on that experience I didn't intend to imply it should be ignored. Only that making such a thing work well isn't trivial or easy. It takes a little more than that.

I think testing some is a splendid idea.

My initial attraction was to make a nice precisely controlled volume unit that I could make remotely controlled with ease.

 

[rebbiputzmaker]

These were DIY efforts with light sensitive resistors. I've not used any manufactured products featuring this. So I probably designed a lousy way to do it. Based on that experience I didn't intend to imply it should be ignored. Only that making such a thing work well isn't trivial or easy. It takes a little more than that.

I think testing some is a splendid idea.

My initial attraction was to make a nice precisely controlled volume unit that I could make remotely controlled with ease.

There is a lot of good diy info on building ldr volume controls. Matching can take some effort though.

http://diyaudioprojects.com/Solid/DIY-Lightspeed-Passive-Attenuator/

https://www.stereo.net.au/forums/to...-lightspeed-diy-ldr-pre-vs-stereo-coffee-diy/

 

[SIY]

And when you're all done, at best you have something which performs significantly less well than a conventional potentiometer. And is more expensive and fiddly.

It's the perfect audiophile solution to a non-existent problem.