Can output impedance measurement/plot be added to the amplifier test repertoire?

[Mnyb]

It could be mentioned if it's abnormal , almost all decent and competently designed amps amps have good enough damping factor it sort of goes along for the ride with designing a good amp , you sort of get low output impedance if everything else is proper ?

And Amir actually mentions many abnormal cases already for example Class-D amps without post filter feedback , they have unusually high impedance in the treble and can modify the response of many speakers ! And Amir carefully states this in those reviews .

 

[Sokel]

I don't know if it's mentioned but I see AP measuring it and it looks automated:

 

[BDWoody]

I added this reason to the OP:

Ok, you have 26 edits to your original post.

That can make it much more difficult to follow the thread, since it makes some answers or comments seem incomplete or out of context.

When you want to make meaningful clarifications or changes, it would be great if you could do it in-line with the rest of the conversation.

 

[Jukka]

It could be mentioned if it's abnormal , almost all decent and competently designed amps amps have good enough damping factor it sort of goes along for the ride with designing a good amp , you sort of get low output impedance if everything else is proper ?

And Amir actually mentions many abnormal cases already for example Class-D amps without post filter feedback , they have unusually high impedance in the treble and can modify the response of many speakers ! And Amir carefully states this in those reviews .

I think that is enough data. For example Topping PA5:

It's not flat and varies by load. This suggests non-flat output impedance. The effect is -0.23 dB for selected loads at 20 kHz, which is all I want to know about it with this amplifier.

If you want more, go ahead and ask Amir

 

[mike7877]

Ok, you have 26 edits to your original post.

That can make it much more difficult to follow the thread, since it makes some answers or comments seem incomplete or out of context.

When you want to make meaningful clarifications or changes, it would be great if you could do it in-line with the rest of the conversation.

The entire OP is unedited. All edits are labelled as such. Most edits are minor grammar corrections.
As someone who's been here the whole time, I can say for sure the flow of discussion hasn't been affected (you can verify this too by reading my edits and looking at contributors' posts - compare topics)
I will try to make all changes together and as quickly as possible in the future though. I'm on a phone, too

Sorry for the edit here, it's underlined.

 

[Deleted member 48726]

I think that is enough data. For example Topping PA5:

It's not flat and varies by load. This suggests non-flat output impedance. The effect is -0.23 dB for selected loads at 20 kHz, which is all I want to know about it with this amplifier.

If you want more, go ahead and ask Amir

Its behaviour with a real load is quite different though.

 

[mike7877]

Also, the edits make it easier for newcomers to join the discussion without reading everything... I'm trying to make things conducive to discussion. And (as you can see), sometimes things come to me after!

 

[mike7877]

Ok, you have 26 edits to your original post.

That can make it much more difficult to follow the thread, since it makes some answers or comments seem incomplete or out of context.

When you want to make meaningful clarifications or changes, it would be great if you could do it in-line with the rest of the conversation.

And what do you mean "in-line"?
Other than edit3, which I prefaced with as being my own addition that I think is important to discussion (and if you read it you will too), what have I added that hasn't been from others' relevant points?

 

[Jukka]

Its behaviour with a real load is quite different though.

And that's exactly why DF is mostly useless information.

 

[mike7877]

Its behaviour with a real load is quite different though.

We've been over this already: [amplifier] frequency response is not only affected by output impedance

[Added for clarity]

 

[notsodeadlizard]

Probably true, but they don't believe anything here, anyways.

Let's just say, it's up to each individual.
But.
From a purely engineering point of view, the task is solvable, but it generates a huge amount of data with which it is not entirely clear what to do.
But what is obviously clear from this is also quite obvious: in these nuances of big data, there is a rationale for blind tests, for example.
Moreover, there are many potential real world effects that justify to some extent and in some cases even such exotic devices as active cable impedance cancellers.
For example, the simplest AWG cable is not pure resistance, it has capacity about 100 picofarads per meter by datasheet:

https://us.rs-online.com/m/d/7fb47615971ffce6fbeb905473d98f88.pdf

And, for example, this cable is directly connected to the feedback circuit of the amplifier (that's how it connects). And the amplifier, for example, with a high-impedance input (for example, there are FET transistors). Can an extra 500 picofarads from 5 meters of cable affect the feedback? Why not? How will it influence? Who knows... There are too many options, too many schematics, too many feedback types etc etc etc.
Perhaps that is why no one has yet refused to listen to sets of equipment.
There is nothing subjective in this.

 

[mike7877]

And that's exactly why DF is mostly useless information.

Read edit3 at the top of OP.
Output impedance significantly affects passive crossover networks.
Edit: if not a plot, an approximation at 1kHz minimally.
But it's been stated a plot would be just as easy an add.
Why are you so against output impedance?

 

[egellings]

I think that with modern day S.S. amplifiers that use loop feedback in the design, low output impedance pretty much comes with for free. Once a D.F. gets higher than about 20 or so, there's no real improvement in S.Q. by going higher in D.F. So unless there is an amplifier that is deliberately designed with a low D.F. in mind, I see no need in reporting it, which can be in the hundreds for S.S. amps. Tube amps, having an inherently low D.F, would be best served by having the D.F. measured & reported.

 

[mike7877]

It could be mentioned if it's abnormal , almost all decent and competently designed amps amps have good enough damping factor it sort of goes along for the ride with designing a good amp , you sort of get low output impedance if everything else is proper ?

And Amir actually mentions many abnormal cases already for example Class-D amps without post filter feedback , they have unusually high impedance in the treble and can modify the response of many speakers ! And Amir carefully states this in those reviews .

In some cases a damping factor of 20 is enough, in others it might be 100. It's not a difficult add either (as has been stated).
It's just like all other pieces of information.
Why doesn't Amir use an ADC with THD+n of 100? That's more than enough for most listening...

Because we want an accurate and complete picture

 

[mike7877]

I think that with modern day S.S. amplifiers that use loop feedback in the design, low output impedance pretty much comes with for free. Once a D.F. gets higher than about 20 or so, there's no real improvement in S.Q. by going higher in D.F. So unless there is an amplifier that is deliberately designed with a low D.F. in mind, I see no need in reporting it, which can be in the hundreds for S.S. amps. Tube amps, having an inherently low D.F, would be best served by having the D.F. measured & reported.

Read edit3 in the OP about how output impedance affects passive crossover networks (very delicate tuning, which when thrown off messes up phase which then affects in-room frequency response)

DF 20 will cause this

Why are a vocal minority of people so against a piece of information they don't know enough to know the importance of?

Audio SCIENCE Review

 

[Deleted member 48726]

Read edit3 at the top of OP.
Output impedance significantly affects passive crossover networks.
Edit: if not a plot, an approximation at 1kHz minimally.
But it's been stated a plot would be just as easy an add.
Why are you so against output impedance?

I think Jukka consider it solved in the way that crossover networks will be the ultimate decider anyway regardless of the DF at the output terminals it will be diminished by the crossover network.

 

[fpitas]

Damping factor can get a little complicated, like Benchmark notes here:

Audio Myth - "Damping Factor Isn't Much of a Factor"

Myth - "Damping Factor Isn't Much of a Factor" Myth - "A Damping Factor of 10 is High Enough" Myth - "All Amplifiers Have a High-Enough Damping Factor" Where did these Myths Originate? These myths seem to trace back to a well-known paper (PDF) written by Dick Pierce. His analysis shows that a...

benchmarkmedia.com

 

[mike7877]

I think Jukka consider it solved in the way that crossover networks will be the ultimate decider anyway regardless of the DF at the output terminals it will be diminished by the crossover network.

So is THD+n over 100 in the same way [being solved, not how it affects crossovers, obviously]
Let's petition Amir to downgrade his ADC (Sarcasm)

 

[mike7877]

It could be mentioned if it's abnormal , almost all decent and competently designed amps amps have good enough damping factor it sort of goes along for the ride with designing a good amp , you sort of get low output impedance if everything else is proper ?

And Amir actually mentions many abnormal cases already for example Class-D amps without post filter feedback , they have unusually high impedance in the treble and can modify the response of many speakers ! And Amir carefully states this in those reviews .

Your first sentence basically says the same thing as mine in the OP (now after a preface, but still clearly marked).

A "well designed" amp (mostly good specs) can have a DF of 30, or 300. Some speakers react badly to lower damping factors. Maybe yours don't (mine don't), but some do. So it's good to know. It also provides insight - consistency and good numbers from brands can instill confidence (when combined with other metrics also good, and consistent)

And damping factor is an often published specification. We should be checking on those claims too, if nothing else - correct?
Part of the aim of making and publishing accurate specifications is to make accountable big brands which have gotten away with way too much for way too long

 

[egellings]

Read edit3 in the OP about how output impedance affects passive crossover networks (very delicate tuning, which when thrown off messes up phase which then affects in-room frequency response)

DF 20 will cause this

Why are a vocal minority of people so against a piece of information they don't know enough to know the importance of?

Audio SCIENCE Review

If the load impedance of a speaker/crossover is low and complex, I can see how a high D.F. (low source impedance) amplifier would be needed to drive something like that, and a D.F. of 20 would be sorely inadequate. Speakers I own tend to be pretty easy to drive, and my home brewed tube amps, with a D.F. of about 25, (high for a tube amp using an OPT) handle them just as well as my Bryston 4Bst bipolar transistor amp does if I don't ask the tube amps to play at powers only the Bryston can deliver. Bass seems similarly taut & bottomless with both amps on those (home-brewed) speakers. So, yes, difficult to drive speakers will need way more than a D.F. of 20 to do their best, no question there. There is no science denial here.