Dynamic Compression in conventional loudspeakers

morricab

Well-Known Member
Apr 25, 2014
7,370
3,279
653
Switzerland
I was reading about the Thrax Lyra speaker as I find some of their designs quite interesting. I hadn't noticed before but I saw that inside this waveguide (horn) that they have for the tweeter is actually a compression driver of the plastic ring radiator design concept. I think this is somewhat of a new development as I thought the original version had like a SEAS Millenium soft dome in that waveguide (the waveguide/horn now looks different from what I remembered seeing at shows). What struck me about it while I was reading was this quote from their site:

"

Dynamic compression - is an often overlooked artefact in many speakers. Voice coil heating being the primary cause of it. Having an ultra high sensitivity horn driver (108db/watt) means no electrical power ever heats the tweeter so transients and bursts are crisp and clean, but to complement this the metal diaphragms of the bass/mid drivers aid the cooling of their voice coils adding to the effect of unrestricted dynamic contrasts."​


I agree quite strongly with Thrax that this is an overlooked artifact and that it has much more impact on reproduction realism than people realize. So, I was quite pleased to see a company taking that seriously and touting their high sensitivity tweeter, which I totally agree is barely ticking over at normal listening levels. But then I got down to the system specs, thinking these speakers would be in the mid-90s dB and an easy load and I see that they are spec'd at 90dB with a 4 ohm load... While this is somewhat above average for a speaker (although with 4 ohm load it really takes 2 watts to get that 90dB at 2.83V) it is by no means what I would consider to be a high sensitivity speaker, which IMO starts from around 95dB and up (that can be debated of course as some will say it only starts at 100dB and up).

What are your thoughts on this topic of dynamic compression and it's relative importance?
 

tima

Industry Expert
Mar 4, 2014
4,704
5,246
1,195
the Upper Midwest
I have been reading about speaker technology to understand it better. The issues of voice coil heating and heat dissapation appear as a somewhat regular topic. As the temperature of the coil rises, the resistance of the coil increases with the claim that in multi-driver speakers with different frequency bands, and different heat dissaption rates, the overall tonal balance is altered.

A by-product of this "thermal compression" is that the harder the speaker is driven, the less responsive it becomes. Increased voice coil resistance can reduce sensitivity. Different voice coil wire choices yield different results. Different driver materials are also a consideration regarding dissapation.

Also voice coil heating can lead to partial demagnetization of the driver's magnet. The claim is this effect is small and reversed when the magnet cools. But, so I theorize, a field coil driver would not be subject to demagnetization resulting from voice cool heating.

However, I have not run across discussion of voice coil heating and dynamic compression specifically - what you are talking about. That is not a denial, just that I have not read about it. But I have read that as voice coil resistance increases from heat, the driver's electrical damping decreases - which I"m guessing can have an impact of dynamics.

This is all interesting stuff. What I do not know is how does the effect of voice coil heating cash out in terms of what we here. What I gather you're suggesting is with higher sensitivity speakers we do not hear (or are less likely to hear) a fading transient response -- in effect the absense of dynamic compression tells the tale. But in less efficient speakers what do we experience as a result of voice coil heating? Or what should we listen for? Do bass drivers become less articulate the longer that Metallica plays?

Your's is an interesting post, Brad, thanks for making it.
 

adrianywu

Active Member
Nov 15, 2021
223
249
43
56
The question is, with passive crossovers, do the crossover components heat up as well ? Power dissipation is proportional to the square of the current. With low impedance bass drivers, this problem is especially important and increased resistance of the crossover components will limit current delivery to the voice coil. I wonder if this problem is more important than actual voice coil heating in most cases.
 

morricab

Well-Known Member
Apr 25, 2014
7,370
3,279
653
Switzerland
The question is, with passive crossovers, do the crossover components heat up as well ? Power dissipation is proportional to the square of the current. With low impedance bass drivers, this problem is especially important and increased resistance of the crossover components will limit current delivery to the voice coil. I wonder if this problem is more important than actual voice coil heating in most cases.
Heating of resistors would be the most likely I think. This could change their resistance (but by how much I am not sure...would have to research it) and a change of resistance could impact the amount of attenuation on the driver and maybe crossover points. Pro speakers are specifically designed to minimize the impact but home speaker drivers tend to have smaller less well ventilated voice coils and are likely much more susceptible as they were designed with other goals.
 

morricab

Well-Known Member
Apr 25, 2014
7,370
3,279
653
Switzerland
I have been reading about speaker technology to understand it better. The issues of voice coil heating and heat dissapation appear as a somewhat regular topic. As the temperature of the coil rises, the resistance of the coil increases with the claim that in multi-driver speakers with different frequency bands, and different heat dissaption rates, the overall tonal balance is altered.

A by-product of this "thermal compression" is that the harder the speaker is driven, the less responsive it becomes. Increased voice coil resistance can reduce sensitivity. Different voice coil wire choices yield different results. Different driver materials are also a consideration regarding dissapation.

Also voice coil heating can lead to partial demagnetization of the driver's magnet. The claim is this effect is small and reversed when the magnet cools. But, so I theorize, a field coil driver would not be subject to demagnetization resulting from voice cool heating.

However, I have not run across discussion of voice coil heating and dynamic compression specifically - what you are talking about. That is not a denial, just that I have not read about it. But I have read that as voice coil resistance increases from heat, the driver's electrical damping decreases - which I"m guessing can have an impact of dynamics.

This is all interesting stuff. What I do not know is how does the effect of voice coil heating cash out in terms of what we here. What I gather you're suggesting is with higher sensitivity speakers we do not hear (or are less likely to hear) a fading transient response -- in effect the absense of dynamic compression tells the tale. But in less efficient speakers what do we experience as a result of voice coil heating? Or what should we listen for? Do bass drivers become less articulate the longer that Metallica plays?

Your's is an interesting post, Brad, thanks for making it.
I found some papers (need to look again) that showed up to a 3db compression on peaks, which would definitely affect dynamic perception (feeling unrestrained and expansive) and impact realism. I think you also have a situation where, as the driver's physical parameters are changing so are the crossover points then changing as they interact with the passive filter. You will of course have different compression rates between drivers in a multiway speaker and this will result in tonal shifts with level as well. It is not that easy to track all the effects because of the dynamic nature of the phenomenon.

You can imagine though a speaker where you keep putting in power but eventually it refuses to get any louder at all.

I used to have a pair of Dynaudio Contour 1.8MKII speakers, which were very neutral, smooth and had good bass (with a big SS amp) and clean highs (Dynaudio always made good soft dome tweeters). When I listened to typical pop/rock music they were about all the speaker one would need; however, as I gravitated more towards well recorded Jazz and Classical music I found that they tended to sound boring and dynamically flat. Adding more power didn't help the situation. If I listened to Peter Gabriel or even Pearl Jam they were great but listening to Oscar Peterson, Jan Garbarek or a Beethoven Symphony, not so great. I tried tube amps on them but they sounded anemic with poor bass. In the end I sold them and got big electrostats, which although on paper were just as insensitive in reality were much more alive and all that more dynamically recorded music came alive.

it was only later that I realized that the Dynaudio speakers were really only comfortable in playing a limited volume range from mid 70s dB to around 90dB, which suited compressed commercial pop/rock fine for the most part. I could set the volume level at around 80dB and only get a few dB swing in the music and the speaker was quite happy and did well what it was supposed to do. But when the swings were 30dB or more then it didn't do so well and lost the drama such big swings are supposed to create.

Contrast that with the electrostatic speaker, which doesn't heat up and therefore doesn't suffer this type of compression. It's limitations are more mechanical (panel travel limits excursion) and so they do the micro scale beautifully, all the low level info on ambient and decay is nicely captured despite low electrical sensitivity) but are capped at the macro unless the panel is truly huge...then they work on the macro too.

Big ribbon panels also don't have much of a heating problem because the cooling is so good for the exposed voicecoil traces. Same mechanical limitations for the big swings but large panels like Apogee Divas could do a nice job there too. Still, the micro for older panels was not as good as electrostats probably due to heavier panels materials (metal traces add a lot of weight) and needed a bit of juice to wake up...also low impedance on some models...

The accurate tracking of the recording dynamics then is, IMO, a serious issue for the realism of reproduction as it is the one thing most speakers do poorly.

The drivers in horn systems will also suffer dynamic compression but the question is at what SPL does it become a serious issue for them. This is where sensitivity and voice coil construction etc. matter to push that point of serious onset (one could argue that it is always there the moment any current flows through the coil) to a high enough SPL not to impact realism at least in a home setting where your max SPL will be much lower than a live concert PA system.

If compression starts to really kick in at much less than 100 watts then you can see that the crest of even 15dB peaks is going to be rounded over ( say average level of 80dB with peaks to 95dB) but with the 100dB driver there will be well under 1 watt going through the driver and compression will be negligible. The 85dB driver will have about 10 watts going through, a significant amount of current into an 8 ohm driver.

The crossover will also contribute and modify the sound as this fluctuations are occurring as well, and thus a passive multi-way speaker will have both drivers and filters impacting the dynamics. This is one area where an active speaker system has a big advantage, IMO.
 

DasguteOhr

Well-Known Member
Sep 26, 2013
1,278
1,246
415
Germany
With horn drivers, the rear chamber is what creates distortion if it is not built stable enough. it provides dynamic compression and a wavy frequency response when you overdo it with volume. two possible solutions either you open them or dampen them from the inside. if you add a damper you lose some efficiency but you get a smoother frequency response. opening brings a bigger sound stage for horns interesting, you lost maximum sound pessure level some db it definitely sounds better. you have to die one death real life
 

tima

Industry Expert
Mar 4, 2014
4,704
5,246
1,195
the Upper Midwest

I don't understand all the details but still an interesting read.

This fellow writes: "More realistically, on the wider-dynamic-range material I more usually listen to, and at my habitual replay levels, the rise in voice-coil temperature and the concomitant thermal compression will be lower still. So I strongly suspect that, for most hi-fi users—those who don't habitually wind the volume control to its highest position and indulge in PA listening levels—thermal compression is a paper tiger."
 

morricab

Well-Known Member
Apr 25, 2014
7,370
3,279
653
Switzerland

I don't understand all the details but still an interesting read.

This fellow writes: "More realistically, on the wider-dynamic-range material I more usually listen to, and at my habitual replay levels, the rise in voice-coil temperature and the concomitant thermal compression will be lower still. So I strongly suspect that, for most hi-fi users—those who don't habitually wind the volume control to its highest position and indulge in PA listening levels—thermal compression is a paper tiger."
I think he is talking about continuous temperatures rather than short term instantaneous rises from dynamic peaks. It turns out that the coil heats up VERY fast (think light bulb filaments) and cools down relatively slowly due to the restricted airflow in tight air gaps. So, while the average temps is one thing it is the instantaneous heating that I am concerned with.
 

morricab

Well-Known Member
Apr 25, 2014
7,370
3,279
653
Switzerland
I think he is talking about continuous temperatures rather than short term instantaneous rises from dynamic peaks. It turns out that the coil heats up VERY fast (think light bulb filaments) and cools down relatively slowly due to the restricted airflow in tight air gaps. So, while the average temps is one thing it is the instantaneous heating that I am concerned with.
"
To be absolutely certain that it doesn't interfere in any way with subtle dynamic contrasts, it would be necessary to use the data from resistance vs time measurements to construct a thermal model of each drive-unit, then write software to apply the appropriate thermal compression to a source file (or, perhaps better, undo it) for the purposes of comparison.

That, maybe, is a project for the future. Right now, the prospect of thermal compression in my listening-room loudspeakers causes me no lost sleep whatsoever."


It is clear his measurements are not fast enough or sensitive enough to capture all that needs to be captured and he kind of realizes it with this statement.

Still, an interesting attempt.
 

Robh3606

Well-Known Member
Aug 25, 2010
1,295
291
505
Destiny
It's an interesting topic and it can be measured. These are 3 examples from Erin's Corner. As you can see it runs the gamut. The M2 woofer as an example uses wire with a TCR that helps limit the rise in resistance as the voice coils heat up. On the compression driver it's simply not needed.

Rob :)
 

Attachments

  • JBL M2_Compression.png
    JBL M2_Compression.png
    77.7 KB · Views: 8
  • KLH Model 5 (Mid Setting, Grille On)_Compression.png
    KLH Model 5 (Mid Setting, Grille On)_Compression.png
    90.1 KB · Views: 8
  • Endow Audio Bravura 7_Compression.png
    Endow Audio Bravura 7_Compression.png
    108.1 KB · Views: 8

rob

Well-Known Member
Jul 12, 2012
49
54
323
It's an interesting topic and it can be measured. These are 3 examples from Erin's Corner. As you can see it runs the gamut. The M2 woofer as an example uses wire with a TCR that helps limit the rise in resistance as the voice coils heat up. On the compression driver it's simply not needed.

Rob :)
Wow, interesting.
Thanks for posting.
 

microstrip

VIP/Donor
May 30, 2010
19,757
3,792
910
Portugal

The third link does not work - it points to your own computer. :)

Anyway, such study using transients should be very simple to carry - if people do not do it is probably because they already know the results at common audiophile sound levels. This effect is however relevant at professional high sound levels in large spaces.
 

morricab

Well-Known Member
Apr 25, 2014
7,370
3,279
653
Switzerland
It's an interesting topic and it can be measured. These are 3 examples from Erin's Corner. As you can see it runs the gamut. The M2 woofer as an example uses wire with a TCR that helps limit the rise in resistance as the voice coils heat up. On the compression driver it's simply not needed.

Rob :)
Yes, I would expect the M2 to be quite good this way and it’s active xover right? Then no effects of the passive filter. The other two are significant and profound differences with SPL, respectively.
 

morricab

Well-Known Member
Apr 25, 2014
7,370
3,279
653
Switzerland
The third link does not work - it points to your own computer. :)

Anyway, such study using transients should be very simple to carry - if people do not do it is probably because they already know the results at common audiophile sound levels. This effect is however relevant at professional high sound levels in large spaces.
You are perhaps underestimating the effect on short peaks. With dynamic recordings it can be easily 20-30dB above average. I think there is a subtle more insidious effect that we are very sensitive to that ultimately robs realism. Kind of like distortion and jitter, how low of an effect is inaudible??
 

DasguteOhr

Well-Known Member
Sep 26, 2013
1,278
1,246
415
Germany

I don't understand all the details but still an interesting read.

This fellow writes: "More realistically, on the wider-dynamic-range material I more usually listen to, and at my habitual replay levels, the rise in voice-coil temperature and the concomitant thermal compression will be lower still. So I strongly suspect that, for most hi-fi users—those who don't habitually wind the volume control to its highest position and indulge in PA listening levels—thermal compression is a paper tiger."
I rather think that the crossover components in conventional loudspeakers ensure that the loudspeaker is limited. For example, a bass coil saturates at higher levels and thus causes distortion and ripple, or a cheap resistor that is operated at the limit ensures that the sound becomes harder and more aggressive, for example. the ear thinks it's dynamic compression. My opinon the red pencil is often applied to companies where you can save money and maximize profits. the xover are calculated and measured, they work well, but you could use better components that improve the result.
 

morricab

Well-Known Member
Apr 25, 2014
7,370
3,279
653
Switzerland
I rather think that the crossover components in conventional loudspeakers ensure that the loudspeaker is limited. For example, a bass coil saturates at higher levels and thus causes distortion and ripple, or a cheap resistor that is operated at the limit ensures that the sound becomes harder and more aggressive, for example. the ear thinks it's dynamic compression. My opinon the red pencil is often applied to companies where you can save money and maximize profits. the xover are calculated and measured, they work well, but you could use better components that improve the result.
One way to tell the effect is to take a speaker and test it with it's passive crossover and then take that same speaker and bypass the crossover and use an active xover (probably DSP based to give a similar frequency response) and then look at the dynamic behavior.
 

Robh3606

Well-Known Member
Aug 25, 2010
1,295
291
505
Destiny
Yes, I would expect the M2 to be quite good this way and it’s active xover right? Then no effects of the passive filter. The other two are significant and profound differences with SPL, respectively.

Same components active vs passive you can see the drivers are the main factor. The M2 may be active but it uses correction filters so you can see the effects.

Rob :)
 

Attachments

  • JBL 4367_Compression (1).png
    JBL 4367_Compression (1).png
    75.8 KB · Views: 4
  • JBL M2_Compression.png
    JBL M2_Compression.png
    77.7 KB · Views: 4

audiobomber

Active Member
Oct 13, 2020
104
58
35
Sudbury ON, Canada
I was reading about the Thrax Lyra speaker as I find some of their designs quite interesting. I hadn't noticed before but I saw that inside this waveguide (horn) that they have for the tweeter is actually a compression driver of the plastic ring radiator design concept. I think this is somewhat of a new development as I thought the original version had like a SEAS Millenium soft dome in that waveguide (the waveguide/horn now looks different from what I remembered seeing at shows). What struck me about it while I was reading was this quote from their site:

"

Dynamic compression - is an often overlooked artefact in many speakers. Voice coil heating being the primary cause of it. Having an ultra high sensitivity horn driver (108db/watt) means no electrical power ever heats the tweeter so transients and bursts are crisp and clean, but to complement this the metal diaphragms of the bass/mid drivers aid the cooling of their voice coils adding to the effect of unrestricted dynamic contrasts."​


I agree quite strongly with Thrax that this is an overlooked artifact and that it has much more impact on reproduction realism than people realize. So, I was quite pleased to see a company taking that seriously and touting their high sensitivity tweeter, which I totally agree is barely ticking over at normal listening levels. But then I got down to the system specs, thinking these speakers would be in the mid-90s dB and an easy load and I see that they are spec'd at 90dB with a 4 ohm load... While this is somewhat above average for a speaker (although with 4 ohm load it really takes 2 watts to get that 90dB at 2.83V) it is by no means what I would consider to be a high sensitivity speaker, which IMO starts from around 95dB and up (that can be debated of course as some will say it only starts at 100dB and up).

What are your thoughts on this topic of dynamic compression and it's relative importance?
My speakers are 4 ohms, Bamberg S5-MTM, with SEAS Excel Millennium silk tweeter, and dual SEAS Excel 18cm mid-bass drivers. Sensitivity is 91dB (2.83V@1-Meter). From the specifications page:

Maximum linear output:
105 dB 64W HPF2 @80Hz, all frequencies, no compression
102 dB 32W no HPF, all frequencies, no compression


 

About us

  • What’s Best Forum is THE forum for high end audio, product reviews, advice and sharing experiences on the best of everything else. This is THE place where audiophiles and audio companies discuss vintage, contemporary and new audio products, music servers, music streamers, computer audio, digital-to-analog converters, turntables, phono stages, cartridges, reel-to-reel tape machines, speakers, headphones and tube and solid-state amplification. Founded in 2010 What’s Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals, we enjoy learning about new things and meeting new people, and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing