Check out these other great forumsSoundSolutionsAudio.comAudioAddicts.netCarAudio-Forum.comCarAudioResources.comHomeAudioForum.net

Jump to content

Modified Show, Tune In!

nuff said.



Photo
* * * * * 1 votes

Understanding Power Compression


  • Please log in to reply
13 replies to this topic

#1 Neil

Neil

    Now let's re-define what it is you know

  • SSA Tech Team
  • 2,384 posts
  • Gender:Male
  • Location:Regina, Sask, Canada

Posted 23 February 2007 - 02:04 PM

Power compression is a topic that is rarely discussed, but always important. When you hear someone call various theile/small parameters a “small signal parameter”, the implication is that with larger signals (more power), there will be a shift or change in these theile/small parameters. This is what we call power compression, and occurs in 3 potential ways.

The first thing to consider is how theile/small parameters are derived. There are 5 basic parameters from which all other data is extracted. These parameters are Re, BL, Mms, Sd, and Cms. So looking at this from a mathematical standpoint, a change in any of these parameters will cause a change in all other parameters as well.

Of those 5 parameters, there are three that can change once power (rather, more power) is applied; they are: Re, BL, and Cms. Now we will review the cause and effects for each one.

Re is the measured DC resistance of the voice coil. When power is applied to the voice coil, the metal becomes hotter, and resistance increases. Heating copper by 200 degrees Celsius will double the resistance, which doubles the measured Re; this results in a 3dB loss in acoustical output. Being that this is also one of the 5 basic parameters, you will see a change in all other parameters as well. For example, a change in Re will cause a large change in your enclosure alignment because doubling Re will also double Qes, as demonstrated below:
Qes = Re*Sqrt(Mms)/(BL^2*Sqrt(Cms))
This is power compression at it’s finest, and the most commonly mentioned one, known as Thermal Power Compression.

The second type of power compression refers to motor force. As power is applied to the voice coil, causing it to progress out of the gap, the motor force (or BL) begins to decrease. This decrease in motor force causes a drop in output and, again, a shift in all other parameters. When the motor force drops by 30%, you've lost 3 dB in output. This is, in part, one of the reasons why more recent definitions (from DLC and Dumax as well as Klippel) of Xmax have defined it as 70% of rest BL. This type of compression is referred to as BL Power Compression.

The third type is related to the compliance of the suspension. As power is applied, the voice coil begins to move out of the gap, causing the cone, spider, and surround to move as well. This movement changes the compliance of the suspension, and is known Cms Power Compression.

Of the three, power compression is a more gradual process. A quick burst of power will heat the voice coil quickly but it will take a while to cool back down and remain fairly constant. Over a 30 second period, the coil will have cooled down and been reheated several times, so that while you thermal power compression occurs, it's not it does not occur nearly as fast as BL and Cms. In other words, thermal power compression is a much less dynamic issue.

BL and Cms power compression are very frequent and occur on every stroke, as a stroke is simply the voice coil and attached assembly moving. For example, playing a 60hz sine wave will cause 60 full strokes in a one second period of time, and since BL and Cms compression occurs at both the positive and negative ends of the stroke, you would actually encounter BL and Cms power compression 120 times a second! Of course, as power is applied, excursion increases, and BL and Cms compression increases as well. This is why it’s so important to find a driver that has very flat BL and Cms curves (measured on Dumax or Klippel) because these curves represent the behaviour of BL and Cms over the stroke; if the curve is relatively flat and extends out fairly far on either side, BL and Cms compression will be of little effect until you reach the extremes.

For most installs, BL and Cms compression will be of the greatest concern. For those who compete or have fairly high power stereos, thermal compression will become a growing concern as more and more power is applied.

To summarize, Thermal, BL, and Cms Power Compression are all issues to consider, but it is important to understand how to work around these issues while knowing which drivers suffer more than others. A driver with adequate cooling and flat BL and Cms curves out to 20+ mm will likely experience very little parameter shift and, as such, is likely a better candidate for your high end stereo.
  • 0

#2 Neil

Neil

    Now let's re-define what it is you know

  • SSA Tech Team
  • 2,384 posts
  • Gender:Male
  • Location:Regina, Sask, Canada

Posted 23 February 2007 - 02:13 PM

To very briefly expand on a thought:

BL Compression is the least resolved issue. Cms compression has been something understood very well for quite some time; things like wide roll surrounds and various spider designs have made Cms compression a relatively solvable issue. BL compression is still a big problem. There are treatments out there that intend to resolve this issue (split coil, XBL^2, and LMT), but each has their own inherent issues. One thing often forget by those who are big fans of certain designs is that there is no free lunch in engineering. To name a couple, XBL^2 sacrifices BL and efficiency, split coil suffers from significant flux modulation and inductance issues, and LMT suffers from high Mms, high inductance, flux modulation, and generally poor transient response.

Speaker engineering is often about making the most appropriate compromise, rather than trying to avoid compromise at all; you'll find the second route very challenging. ;)

Edited by DevilDriver, 23 February 2007 - 02:15 PM.

  • 0

#3 Chris

Chris

    Regular Member

  • Members
  • 699 posts

Posted 25 February 2007 - 10:06 PM

When power is applied to the voice coil, the metal becomes hotter, and resistance increases.

How is this measured? I know it happens. I am just curious. It seems to be logical to assume this increase in resistance would be linear to some degree. The measurement of 200deg C will double the resistance and result in a 3db loss in acoustical output is an great example. This also makes me wonder about different coil compositions and what implications said variable would have on not only to the rise itself but the equation used to measure and/or estimate the rate and amount of change.


Edit: I just went through the Pinned explaination of how to measure impedance rise. I posted too soon. :)

Edited by Chris, 25 February 2007 - 10:14 PM.

  • 0

#4 Chris

Chris

    Regular Member

  • Members
  • 699 posts

Posted 25 February 2007 - 10:07 PM

Great thread btw. I love learning new things about my hobbies! ;)
  • 0

#5 Chris

Chris

    Regular Member

  • Members
  • 699 posts

Posted 25 February 2007 - 10:10 PM

One more quick question. What is "split coil" and what is its function?

Great read!
  • 0

#6 Neil

Neil

    Now let's re-define what it is you know

  • SSA Tech Team
  • 2,384 posts
  • Gender:Male
  • Location:Regina, Sask, Canada

Posted 25 February 2007 - 10:27 PM

One more quick question. What is "split coil" and what is its function?

Great read!


I am working on a detailed explanation of the different BL technologies and their respective advantages/disadvantages. I will try to get that up by next weekend.
  • 0

#7 Chris

Chris

    Regular Member

  • Members
  • 699 posts

Posted 25 February 2007 - 10:34 PM

One more quick question. What is "split coil" and what is its function?

Great read!


I am working on a detailed explanation of the different BL technologies and their respective advantages/disadvantages. I will try to get that up by next weekend.



Awesome. I look forward to it. Thank you for taking your time to share this information with everyone.
  • 0

#8 Neil

Neil

    Now let's re-define what it is you know

  • SSA Tech Team
  • 2,384 posts
  • Gender:Male
  • Location:Regina, Sask, Canada

Posted 25 February 2007 - 10:40 PM

One more quick question. What is "split coil" and what is its function?

Great read!


I am working on a detailed explanation of the different BL technologies and their respective advantages/disadvantages. I will try to get that up by next weekend.



Awesome. I look forward to it. Thank you for taking your time to share this information with everyone.

No problem. Big thanks to you and those who reply. It makes things more gratifying and is also conducive to expanding on each topic.

Greatly appreciated guys!
  • 0

#9 mrray13

mrray13

    5-27-20 to 7-20-08

  • Super Moderators
  • 16,020 posts
  • Location:southern Illinois

Posted 26 February 2007 - 11:20 AM

very nice read bro!


like chris mentioned, it's awesome to learn things in a hobby one loves, and in my case, provides my income as well.


also, how does one build a box around these things? i've heard of building a box twords a specific imedence rise, or one that controls it, so that you stay at or around the impedence your amplifiers crave. this is for competition purposes of course. i mean, is it just trial and error in playing with box volume? or can a decent program, like bb6p, actually forecast this with reliable accuracy?



wheeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee :slayer:
  • 0

I'm not a big fan of ignorant, blanket statements that run rampant in DIY audio. That and my mommy didn't love me enough.


#10 Tommythecat

Tommythecat

    Newbie

  • Members
  • 91 posts

Posted 27 February 2007 - 04:01 PM

To me the concerns over BL and CMS power compression are limited to high excursion (20mm+) woofers. Even then, to achieve such extreme throw most designs require a large box or lots of power - most of the time well over 500W - to approach Xmax. A typical 4" coil (pro audio) can see over 3db of power compression at less than 500W (pink noise - low crest factor) without exceeding 1/2 Xmax. "Car subs" which are long-throw dont use 4" coils, but do have much more magnet area and most likely exhibit the same levels of power compression. Now, when we look at midbass, midrange, and tweeters we are also much more concerned over thermal power compression as well as they usually dont reach Xmax very often.

BL and CMS linearity are much more important in limiting distortion. As was stated above, at any frequency, the coil is shuffling through the gap many, many times a second and non-linearities in BL and CMS show up as distortion products.
  • 0
Ask me why ThomasW is a moron.

#11 bdawson72

bdawson72

    Advanced Member

  • SSA Regular
  • 1,052 posts
  • Location:Southern Virginia

Posted 05 March 2007 - 01:45 AM

Good read
  • 0
Posted Image

#12 soydeedo

soydeedo

    Newbie

  • New Members
  • 23 posts

Posted 02 March 2008 - 03:42 PM

I had never really heard of this topic until today via one of audioholic's posts. I did a google search and this article came up, but on s10forum.com. I thought to myself, wow, that's a pretty in depth topic for an s10 forum, but then I saw the author and did a quick search here. Awesome writeup as always, DD. =)

#13 ford302redneck

ford302redneck

    SSA Regular

  • SSA Regular
  • 2,752 posts
  • Location:Darlington, SC

Posted 07 March 2008 - 05:27 PM

What would the car audio community be without Neil.
  • 0
Posted Image

#14 mpimm

mpimm

    Member

  • Members
  • 159 posts
  • Location:Edmonton, Alberta, Canada

Posted 05 April 2008 - 02:00 PM

Thanks Neil. Always find your write-ups enlightening.
  • 0




1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users

#1 Car Audio Forum Since 2001!