High vs low damping factor in amps (Bass "suspension")

[terryforsythe]

Damping factor is a property of an amplifer, just an weird presentation of amplifier output impedance. You seem to use some other definition. Unless you specify what it means to you, your graphs tell nothing.

Specifying "damping factor" of an amplifier is a bit of a misnomer, probably more for marketing purposes than providing relevant data. What amplifier manufacturers should specify is minimum output impedance or, better yet, a chart of output impedance vs. frequency.

Specifying "damping factor" for an amplifier is problematic because, typically, manufacturers specify it with reference to an 8 ohm resistor. The input impedance of a real speaker is nowhere close to being purely 8 ohms resistive. Just look at the impedance plot of any typical speaker. I have seen many speakers that have impedances dropping down below 4 ohms, and some even below 3 ohms, and shooting up to over 40 ohms.

As a useful value, damping factor is determined at the terminals of the driver, comparing the impedance of the driver to the impedance measured from the driver's terminals all the way back into the output stage of the amplifier. Amlifiers are used with many different speakers, different cable resistances, etc., and it would be nearly impossible to specify the true damping factor for every possible setup.

 

[AJM1981]

You can't have a good loudness without acoustical measurements. It is properly done in AVRs. Examples are Yamaha YPAO Volume and Audyssey Dynamic EQ.
I liked Yamaha implementation and was curious, how it works in WXA50.

Looked it up. Apparently it is not the extra bold auto EQ I mentioned.

· What it is: Volume Adaptive EQ is a smart loudness management system . Traditional loudness buttons simply boost bass and treble at a fixed level. In contrast, this feature automatically adjusts the balance of bass and treble in real-time based on the volume level you set .
· Why it's used: This ensures you hear full, well-balanced sound even at low listening levels. It compensates for the human ear's natural tendency to perceive less bass and treble when music is quiet .
· How to set it up: The system is calibrated by selecting the type of speakers you have connected (e.g., bookshelf, floor-standing, in-wall) through the setup menu, which optimizes the effect for your specific setup

Have it activated for years. My standmounts are oddly between what is considered bookshelf and floorstanding. Went for the latter.

 

[mhardy6647]

Damping factor is a relic from tube era. It had some meaning, when it was a single digit value.

Emphasis added, and: Guilty as charged.



Output impedance ca. 1 ohm. I do use nominally 16 ohm loudspeakers, though, so there's that.

 

[popej]

Specifying "damping factor" of an amplifier is a bit of a misnomer, probably more for marketing purposes than providing relevant data.

Yes, but this is what OP is testing. There is no reason to look for other manings and deviate from posted problem.

As a useful value, damping factor is determined at the terminals of the driver, and comparing the impedance measured there all the way back into the output stage of the amplifier.

Can't see the use of the value. I don't want to calculate\simulate but I doubt that added inductance reduce Q factor of speaker resonance. Actually it shouldn't significantly change anything in pass band, so maybe your calculation are wrong?

 

[popej]

Have it activated for years.

It definitely can change bass. Try without.

 

[terryforsythe]

I doubt that added inductance reduce Q factor of speaker resonance.

I didn't say that added inductance reduces Q factor.

Actually it shouldn't significantly change anything in pass band, so maybe your calculation are wrong?

Watch the video I posted in post #30. Then watch the video below. The two videos explain why that is not correct (not referring to Q, but pertaining to amplifier output impedance and passive filters on a speaker - the first video discusses both, the second video discusses just the output impedance).

 

[popej]

I didn't say the added inductance reduces Q factor.

Damping implies Q factor. You didn't explain the use of your "damping factor" so I'm guessing.

Watch the video I posted in post #30. Then watch the video below.

Second video is irrelevant, since we don't talk about high impedance amplifiers. Not going to watch the other, sorry, already wasted enough time.

 

[terryforsythe]

You didn't explain the use of your "damping factor" so I'm guessing.

The charts are a computation of damping factor vs frequency, as clearly labled on the charts. Damping factor (DF) is computed as:

DF = (ZL + ZS)/ZS

where ZL is the impedance of the driver and ZS is the source impedance as seen from the driver's terminals.

For ZL, I used the magnitude and phase of the woofer's measured impedance at the specific frequencies where you see the markers on the curves. For the amplifier's output impedance I used the impedance claimed by the manufacturer of my amplifier (it may not perfect because they only provided a single resistance value, not output impedance vs frequency). I also included the resistance, inductance and capacitance of the speaker cable.

The first chart includes in the model the passive filter from my speaker, including the DCR I measured for the inductors. For the second chart I removed the passive filter from the model, e.g., as implemented in an active speaker.

 

[mhardy6647]

Your Sansui amp has 12W power output. Your Yamaha has around 80W available.

Admittedly not quite a trivial difference, but still "only" 8.2 dB.
12 watts continuous is about 6 dB more than my aforementioned (aforepictured? ) SE 2A3 amp.
Horses for courses.

 

[antcollinet]

Admittedly not quite a trivial difference, but still "only" 8.2 dB.
12 watts continuous is about 6 dB more than my aforementioned (aforepictured? ) SE 2A3 amp.
Horses for courses.

Enough for almost double percieved volume change. And plenty to make a huge audible difference if the lower powered amp clips the peaks.


And vastly more significant than a 40 to 50 DF change.

 

[mhardy6647]

Enough for almost double percieved volume change. And plenty to make a huge audible difference if the lower powered amp clips the peaks.

yup. Non-trivial... not immense, though.

 

[Chrispy]

Didn't see this thread mentioned https://www.audiosciencereview.com/...-impedance-damping-factor-and-speakers.23968/

 

[popej]

DF = (ZL + ZS)/ZS

You have already said that in earlier posts. I just don't see any use of this definition. Not for physical damping, not for drawing loudspeaker characteristic. I don't see, what you are trying to prove.
I have used simplified model of speaker to illustrate, that there is no significant differnce between damping factor 25 and 60. Damping factor as commonly used in amp specification and actual damping of speaker resonance.
I know what speaker impedance characteristic looks like. With low DF there will be problems, but not with amps used by OP. Why push this off-topic?

 

[terryforsythe]

Second video is irrelevant, since we don't talk about high impedance amplifiers. Not going to watch the other, sorry, already wasted enough time.

I just don't see any use of this definition. Not for physical damping, not for drawing loudspeaker characteristic.

You don't have time to watch the videos, but you still have time to debate. If you watch the videos, then you will understand. But, as of now, to me it seems that you don't.

The first video shows how damping factor affects the time domain response of a woofer. The second video explains why output impedance is the more appropriate value to use for an amplifier in lieu of damping factor, and shows how output impedance affects a speaker's frequency response. But, unless you understand those things, we are just talking in circles. More talk is not going to change the science.

 

[AJM1981]

It definitely can change bass. Try without.

The change is minor. What would it be.. a percent, maybe 3% hairsplitting on the scale of human perception. Definitely not double blinded. It does not change the bass to anywhere near the Sansui signature when it is a subtle gradual loudness contour. But again, the au505 is not in the loudness contour area with its default sound.

I am not dissatisfied with the Yamaha amp by the way. Reviews are also good and some underline it has really good grip on bass and probably would be the pick for most listeners having a more tight /modern detailed perceived sound instead of warmer / bloomy.

The Yamaha has the enhancer toggle as a touch of the aural exciter effect to add harmonics. Which affects the bass a little , but more in a way that it adds a bit of groove instead of bloom, and it kind of opens the mids. A bit of a digital tube emulation which in my opinion is a nice class d extra. But totally not comparible.

 

[KSTR]

Damping factor (DF) is computed as:

DF = (ZL + ZS)/ZS

That's what I'd call the marketing damping factor because values explode in an 1/x fashion for small (but irrelevant) Zs.

The true engineering damping factor would be Zl/(Zl+Zs), telling us how much of damping is lost vs true zero output impedance. This definition makes clear that once Zs is less than 1/10th of Zl the improvement from going lower is marginal. It also deals with negative output impedance properly (resulting in a damping factor > 1).

 

[solderdude]

Those output caps are in the AC feedback path though which might cause phase issues in the low frequencies.
Open loop gain is probably not very high either.
Hope these caps are not dried out.

 

[popej]

If you watch the videos, then you will understand.

I hate to watch 20 minutes of video to get information I could read in 2 minutes. But I did.
To summarize:
We both use the same model of speaker to analyze.
You simulated DF of 6.5 to show problems. I have used real values, to illustrate, that for modern amps problem is insignificant.
You have discovered, that 3-rd order Butterworth introduces ringing, but you put this problem in the same category as speaker damping. It is not, going to active speaker with the same Batterworth before amp will give ringing too.

 

[terryforsythe]

I hate to watch 20 minutes of video to get information I could read in 2 minutes. But I did.
To summarize:
We both use the same model of speaker to analyze.
You simulated DF of 6.5 to show problems. I have used real values, to illustrate, that for modern amps problem is insignificant.
You have discovered, that 3-rd order Butterworth introduces ringing, but you put this problem in the same category as speaker damping. It is not, going to active speaker with the same Batterworth before amp will give ringing too.

I simulated the actual DF of my actual woofer both with its passive filter in the circuit and without. What I discovered is that the DCR of the inductors in series with a woofer are a significant factor in the DF computation, oftentimes a far bigger issue than the DCR of the speaker cable and sometimes a bigger issue than the output impedance of the amplifier, and many people overlook that issue.

Any filter is going to create issues with a square wave. That is to be expected, being clear by taking the Fourier series of the square wave, attenuating the harmonic components as a filter does, then converting it back to the time domain. But, that is outside the scope of the damping factor discussion.

 

[terryforsythe]

That's what I'd call the marketing damping factor because values explode in an 1/x fashion for small (but irrelevant) Zs.

The true engineering damping factor would be Zl/(Zl+Zs), telling us how much of damping is lost vs true zero output impedance. This definition makes clear that once Zs is less than 1/10th of Zl the improvement from going lower is marginal. It also deals with negative output impedance properly (resulting in a damping factor > 1).

Zl/(Zl+Zs) is called the "damping ratio", and has been proposed as a new definition to be used in lieu of damping factor. I agree that it makes more sense. Nonetheless, the former definition still is used, and much more commonly shortened to Zl/Zs. Later today I will post updated charts for my speakers using the damping ratio.