Damping Factor

[rantan]

50 minutes ago, MLXXX said:

You appear to have decided to focus on a particular amplifier spec.  And you appear to have decided it was responsible for the differences in sound you heard.

I am sorry if this appears to be the reason for creating the thread. I am actually seeking information about how the damping factor may influence sonic characteristics and why it varies so much in different amplifiers

[MLXXX]

21 minutes ago, rantan said:

I am sorry if this appears to be the reason for creating the thread. I am actually seeking information about how the damping factor may influence sonic characteristics and why it varies so much in different amplifiers

Fair enough.

 

 

In your opening post you said:

22 hours ago, rantan said:

Are there particular advantages in a high factor and why do some amplifiers have 20-30 times more than another?

 

I think the 20-30 times factor only applies to extremes.

 

1000 would be exceptionally high. 20 would be unusually low. 100 would be more typical.

 

20 would suggest an old-fashioned classic valve amplifier where losses in the output transformer (winding resistance and coupling losses ) made it difficult to achieve a high damping factor.   That type of amplifier could vary its performance depending on the characteristics of the speakers connected to it.

 

100 or 200 would be fine. There is enough "stiffness" to the driving voltage  from the amp for it not to be noticeably affected by speakers with varying impedance vs frequency.

 

1000 suggests overkill, or some unusual design. However of itself that need not affect the sound quality.

 

Why is it that the rated damping factor differs as between different power amplifiers?  The damping factor depends on the combination of how low the inherent output impedance would be without any negative feedback applied for the output stage, and how much negative feedback is used to bolster the inherent performance of the output stage, lowering its effective output impedance.

Edited March 15, 2023 by MLXXX

[Addicted to music]

1 hour ago, Hydrology said:

Amplifier topology, amplifier design, component/semiconducter choices, synergy with partnering equipment etc etc 

^^^^
 

This is spot on.   
Never relate to the SQ of and amp based on figures, especial dampening factor.   The product that presents the SQ is based on device choices, circuit implementing and passive components….

The other thing that can effect amplifier performance is that something is right:  faulty etc.

[georgehifi]

2 hours ago, Keith_W said:

he makes a convincing argument that a damping factor > 200 is of marginal benefit. 

 

Nick Barofsky Riverina HiFi dealer for both Linn/Naim/ME/AMW etc  did a demo for me back in the 80's

Naim proved with the first gen 250 that a damping factor of DF=20 (.4ohm) was all that was needed for tight dynamic bass into the 8ohm Isobarics amazing bass. (It had only local feedback from memory).

But believe it or not Nick then did a demo for me with the Isobarics with an big ME150 similar to the 1400 DF=220 (.03ohm), it had less bass and slam into them, which proved to me the Isobarics were an over-damped design, with a Q of around .2 or .3 which didn't gel with amps with good damping factor. They preferred the lowish 20 DF on the early Naim 250

 

Cheers George

Edited March 15, 2023 by georgehifi

[rantan]

On 15/03/2023 at 3:21 PM, MLXXX said:

1000 suggests overkill, or some unusual design. However of itself that need not affect the sound quality.

 

Having done a little research I have found that several Class D amplifiers have damping factors of >1000, so it is not an unusual design.

I should ad that the amplifier I heard with the high DF was NOT a class D design.

[March Audio]

On 15/03/2023 at 12:36 PM, rantan said:

I am actually seeking information about how the damping factor may influence sonic characteristics and why it varies so much in different amplifiers

 

Low output impedance results in high damping factor.  A low output impedance will generally help control the movement of the driver.  However as I mentioned, beyond a certain level WRT driver control it becomes a bit meaningless.  This is due to the much higher driver impedance being the limiting factor to the current flowing in the circuit.

 

Low output impedance is a good thing however for other reasons.  It means (all things being equal) that the amps frequency response will remain flat regardless of load.  High output impedance amps, such as tubes, will have a variable frequency response (sound) depending on what load (speaker) you attach.  

 

This is one major reason why testing with resistive dummy loads is unhelpful in judging how an amp might sound.  You need reactive, variable impedance loads that simulate real speakers.

Edited March 16, 2023 by March Audio

[davewantsmoore]

On 14/03/2023 at 6:29 PM, Keith_W said:

A high damping factor means that the amplifier has more control over the speaker cones

No.

https://forums.melaudia.net/attachment.php?aid=32566

 

 

Anybody who says this.....  either actually doesn't understand "damping factor," or (like with some manufacturers of amplifiers and speaker, I suspect) doesn't care that their customers don't.

[georgehifi]

Here's something to ponder over

 

DF to a speakers "minimum impedance" ( lets say 2ohm), could very well be like impedance matching of your electronics outpts to inputs all the way down the line to the poweramp

Output impedance of a source to input impedance of the pre, output impedance of the pre to the input impedance of the amp. 

 

When I manufactured my Lightspeed Attenuator (LDR) it was 2.5kohm max output impedance and I found if the power amp was >33kohm input, and there was no change in sound if the amp was 100khm. So 33kohm was the min for it to work optimally into. That's a ratio of 33 divided by 2.5.  1 to I3.2 impedance ratio, I worked on 1 to 10

I think it would be a safe bet if the minimum of a speaker hits 2ohm then the output impedance of the amp should be 10 x lower .2ohm or lower to give a DF of at least 10! (but then the Q of the speaker also comes into play, being .7 they say is optium, but as I found the Linn Isobarics were a Q .4 which is over damped and needed an amp with high output impedance, too low and they had no bass with high feedback amps.

Now that into your hypothetical speaker 8ohms,  is or case .2ohm into 8ohms = a DF of 40! Any DF over this, even 1000 may not be detectable and may not sound any better.

 

Cheers George       

[March Audio]

1 hour ago, davewantsmoore said:

No.

https://forums.melaudia.net/attachment.php?aid=32566

 

 

Anybody who says this.....  either actually doesn't understand "damping factor," or (like with some manufacturers of amplifiers and speaker, I suspect) doesn't care that their customers don't.

 

I would respectfully have to disagree with you there. Your link actually confirms the driver control mechanism.

 

Before this turns into an argument , yes there is a critical point to be optimally damped, but low output impedance is also desirable for other reasons.

 

I think you might want to treat that paper with a bit of caution.  It's not really applicable to typical passive speaker voltage source amps.

 

Have you ever seen a speaker response vary by 14dB with different amps?

 

 

 

If you look at the examples later in the paper it shows some extremely high levels of resistance.  These would never be encountered with real world amplifiers.

 

I know where this is going, current drive, but whatever its advantages might be, it requires active dsp designs specifically tuned for the driver and box.  Valid course maybe, but not what 99.9% of audiophiles are contending with typical passive speakers and amps.

 

 

Edited March 16, 2023 by March Audio

[Addicted to music]

34 minutes ago, March Audio said:

 

 

 

Have you ever seen a speaker response vary by 14dB with different amps?

 

 

The paper was reprinted in 1954-55.  It talks about $2 speakers with 6 ounce and 5 pound magnets......so maybe they so by 14db 

[davewantsmoore]

35 minutes ago, March Audio said:

Have you ever seen a speaker response vary by 14dB with different amps?

Sure, when the Zout goes from ~"0" to ~"infinite"

 

35 minutes ago, March Audio said:

If you look at the examples later in the paper it shows some extremely high levels of resistance.  These would never be encountered with real world amplifiers.

If by "real world" you mean ones with an Zout of like >> 1ohm... sure.

 

But, I don't see the point, re: the understanding of damping factor, in the context of "larger damping factors provive more 'control' to the louderspeaker".

 

Quote

I would respectfully have to disagree with you there and your link confirms it.

I'm not quite certain what you're disagreeing with.

Is it that:  Large(r) damping factors don't prove "more control" to the loudspeaker.

?

... and I don't see how "the link confirms it".

 

Varying the "damping factor" (As I said before) is just going to (if anything) change the frequecny response and the settling time of the cone..... the right amount is the right amount.    It is not a case of "more is (simply) better".

[March Audio]

1 minute ago, Addicted to music said:

 

The paper was reprinted in 1954-55.  It talks about $2 speakers with 6 ounce and 5 pound magnets......so maybe they so by 14db 

 

Whilst it is fundamentally correct in what its saying, the lack of current drive speakers out there would suggest its benefits are not overwhelming.  

 

Previously difficult to implement, with modern dsp it becomes a much easier challenge, yet it's not gained any traction.

[davewantsmoore]

5 minutes ago, March Audio said:

Whilst it is fundamentally correct in what its saying, the lack of current drive speakers out there would suggest its benefits are not overwhelming.  

Previously difficult to implement, with modern dsp it becomes a much easier challenge, yet it's not gained any traction.

We seem to be missing the point.

 

It isn't about low dam,ping factors being superior.

 

It is about high damping factors not offering any inherent benefit.    An audiophiles inherent understaning of damping factor boils down "it should be high (or higher) to provide control for the cone".   It's as much nonsense today (on todays speakers) as it was 70 years ago.

Edited March 16, 2023 by davewantsmoore

[Addicted to music]

1 minute ago, davewantsmoore said:

We seem to be missing the point.

 

It isn't about low dam,ping factors being superior.

 

It is about high damping factors not offering any inherent benefit.    An audiophiles inherent understaning of damping factor boils down "it should be high (or higher) to provide control for the cone".   It's as much nonsense today (on todays speakers) as it was 70 years ago.

 

here's a Benchmark view on the topic,  look at the excel spreed sheet,  check the bottom with the DF of 1000 and look at the error deviation in db.

 

https://benchmarkmedia.com/blogs/application_notes/audio-myth-damping-factor-isnt-much-of-a-factor

[March Audio]

2 hours ago, davewantsmoore said:

I'm not quite certain what you're disagreeing with.

Is it that:  Large(r) damping factors don't prove "more control" to the loudspeaker

 

You objected to the terms used implying that damping doesn't control driver motion.  It does, and your paper supports that.

 

2 hours ago, davewantsmoore said:

Varying the "damping factor" (As I said before) is just going to (if anything) change the frequecny response and the settling time of the cone..... the right amount is the right amount.    It is not a case of "more is (simply) better".

 

Actually, in the case of traditional amps and speakers a low output impedance will provide a flatter frequency response.

 

Just see what high output impedance tube amps do to frequency response. (Black trace)

And you are worried about sub optimal damping factor?

 

NOTE: To whome it may concern.  I am not bashing tube amps. They can sound great.  It's just a technical fact that they suffer from load dependant FR variations.

 

 

Your arguments are valid, but only in the context of active (dsp) speakers with current drive amps, which by definition have to be an integrated design.

 

With the passive speakers most people have, a low output impedance will offer lower frequency response deviations.

 

Damping factor (output impedance) can't be optimised, so all things considered low output impedance is the best objective.

Edited March 16, 2023 by March Audio

[March Audio]

27 minutes ago, davewantsmoore said:

We seem to be missing the point.

 

It isn't about low dam,ping factors being superior.

 

It is about high damping factors not offering any inherent benefit.    An audiophiles inherent understaning of damping factor boils down "it should be high (or higher) to provide control for the cone".   It's as much nonsense today (on todays speakers) as it was 70 years ago.

But they do (within reason).  Low output z provides load independence.

Edited March 16, 2023 by March Audio

[March Audio]

4 minutes ago, Addicted to music said:

 

here's a Benchmark view on the topic,  look at the excel spreed sheet,  check the bottom with the DF of 1000 and look at the error deviation in db.

 

https://benchmarkmedia.com/blogs/application_notes/audio-myth-damping-factor-isnt-much-of-a-factor

 

Yep, as I already mentioned, beyond certain level it becomes irrelevant.

 

Damping factor is frankly a bit of useless metric.  Just look at the output impedance and so long as it's below a reasonable level it will be fine.

[Addicted to music]

1 hour ago, March Audio said:

 

 

 

I know where this is going, current drive, but whatever its advantages might be, it requires active dsp designs specifically tuned for the driver and box.  Valid course maybe, but not what 99.9% of audiophiles are contending with typical passive speakers and amps.

 

 

Spot on, There was a thread here many moons ago in reference to DF,  that discounted  typical class A and B amps are Voltage Source.  The way the thread went was a designed that's turned to that electronics and speaker,  a change in either makes it void.   

[March Audio]

1 hour ago, Addicted to music said:

 

Spot on, There was a thread here many moons ago in reference to DF,  that discounted  typical class A and B amps are Voltage Source.  The way the thread went was a designed that's turned to that electronics and speaker,  a change in either makes it void.   

 

If you design for optimal damping factor you will require an amplifier with a specific output impedance.  This will inevitably be high enough to create frequency response aberrations for the speaker box combination, just like tube amps suffer.  To properly correct for these aberrations you will need dsp.

 

As I said, this may well be a valid path to take, but it's a specific product for a certain segment of the market.  By definition, expensive. 

 

So for most of us who are using passive speakers, a low output impedance is a good thing.

Edited March 16, 2023 by March Audio

[March Audio]

1 hour ago, Addicted to music said:

 

here's a Benchmark view on the topic,  look at the excel spreed sheet,  check the bottom with the DF of 1000 and look at the error deviation in db.

 

https://benchmarkmedia.com/blogs/application_notes/audio-myth-damping-factor-isnt-much-of-a-factor

 

The Benchmark article is great and gets the salient points across very well.

 

I think we have taken a jump with the conversation which would be difficult to follow.

Edited March 16, 2023 by March Audio

[Addicted to music]

19 minutes ago, March Audio said:

 

If you design for optimal damping factor you will require an amplifier with a specific output impedance.  This will inevitably be high enough to create frequency response aberrations for the speaker box combination, just like tube amps suffer.  To properly correct for these aberrations you will need dsp.

 been awhile since that thread,  lots of eyes on it because the designer taught at Sydney University,  however I’m sure that DSP wasn’t used,  probably defeat his main intent.  What I do remember was a “transconductance amplifer”  current drive!   And a resistor value calculated to flattened  the the peak freq.     

[Addicted to music]

3 hours ago, rantan said:

 

Having done a little research I have found that several Class D amplifiers have damping factors of >1000, so it is not an unusual design.

I should ad that the amplifier I heard with the high DF was NOT a class D design.

I’ve probably missed something,  if the amplifier sounded that different why are you still blaming it on DF figures?   

[March Audio]

Just now, Addicted to music said:

 been awhile since that thread,  lots of eyes on it because the designer taught at Sydney University,  however I’m sure that DSP wasn’t used,  probably defeat his main intent.  What I do remember was a “transconductance amplifer”  current drive!   And a resistor value calculated to flattened  the the peak freq.     

 

Yes that would work. It certainly can be done "active analogue", but when you dial in crossover filtering, baffle step, other corrections as well, these days dsp is the only sensible path.

[Addicted to music]

11 minutes ago, March Audio said:

 

Yes that would work. It certainly can be done "active analogue", but when you dial in crossover filtering, baffle step, other corrections as well, these days dsp is the only sensible path.

absolutely, issue resolved, and you can use any combination as desired so it’s flexible.  

[The Mad Scientist]

56 minutes ago, March Audio said:

 

The Benchmark article is great and gets the salient points across very well.

 

I think we have taken a jump with the conversation which would be difficult to follow.

There are quite a few good articles on Benchmark's website. Mostly no-nonsense and based on solid science and engineering.