Home » Sponsored » Pi Speakers » Heat exchanger effectiveness
Heat exchanger effectiveness [message #49650] Wed, 21 June 2006 13:27 Go to next message
Wayne Parham is currently offline  Wayne Parham
Messages: 18670
Registered: January 2001
Illuminati (33rd Degree)

I have recently answered a lot of E-Mail's about the effectiveness of the heat exchanger on the 12π basshorn. Some of the questions are how much power handing is increased, is it a "fix" for small sealed chambers and does it work in applications other than the 12π. We covered this thoroughly last year when the device was being developed but I suppose it merits another examination here.

In short, speakers run at high power levels dump a lot of heat into the magnet and pole piece. Some voice coil heat is removed by air cooling through the vent but this does almost nothing to cool the motor core. Heat is generated in the motor core from magnetic losses. It is also radiated from the voice coil into the pole piece. This heat tends to buildup in the core, and within a few minutes it can become hot enough to boil water. This heat surrounds the voice coil raising its local ambient temperature. This makes it less able to sustain additional heat from signal current flowing through it. Eventually, the heat causes the voice coil adhesive to weaken and fail. The coil separates from the former. This is the most common failure mode of any speaker.

The heat exchanger is simple. It wicks the heat away from the pole piece, sinking it into a large plate which then radiates it away. The same mechanism can be used by virtually any speaker to improve thermal performance. Power handling is increased and thermal compression reduced.

For the LAB12, power handling increased over 225% over a driver in free air. If placed in a small contstrained space where the air can become superheated, the performance increase may be more. But the improvement was measured with a driver surrounded by air conditioned cool air. The problem isn't limited to systems with small sealed rear chambers; The problem is that the heat is retained in the magnet and pole piece.

If you think about it, a speaker voice coil is applied several hundred watts, so it gets hot like a large soldering iron. Even if the speaker system is very efficient, you still have hundreds of watts dissipated as heat. Take a theoretical 400 watt speaker at a very optimistic 50% efficiency level - You still have 200 watts of heat. This heat source is surrounded by steel and then covered by a large chunk of ceramic. This is a pretty good heat container, one that is almost made to hold heat. So one of the best things you can do is to get a good conductor of heat down inside the motor, in contact with the pole piece. Wick the heat out of the core and radiate it away.

Every loudspeaker can take advantage of this technology, including those with open backs.

Re: Heat exchanger effectiveness [message #49659 is a reply to message #49650] Fri, 23 June 2006 05:45 Go to previous messageGo to next message
Leland Crooks is currently offline  Leland Crooks
Messages: 212
Registered: May 2009
Master
I like mine ../emoticons/smiley.gif

I routinely run 550w into hl10's which are 300w speakers. And I run them right at that limit. They never break a sweat. When I'm running just one pair, it's 1100w from the bridged amp. I don't push it to the limit when it's just one pair, but I also don't worry about it.

I've even got them in Beta 10's. Pretty small sinks in the cab I'm using, but I'll take any protection I can get. I see them more as insurance rather than for increased power handling. If it it keeps you from losing a driver at a gig, they're worth every penny.

I've still got to get around to some more heat testing. I got an RF thermometer to put in the vc and be able to read it remotely, but it's my busy season. Probably won't happen until the fall.
Re: Heat exchanger effectiveness [message #49660 is a reply to message #49659] Fri, 23 June 2006 10:08 Go to previous messageGo to next message
Wayne Parham is currently offline  Wayne Parham
Messages: 18670
Registered: January 2001
Illuminati (33rd Degree)

They work very well, don't they?

What do you think of the size of the vent on the new HL10's? I was considering making a 10π subwoofer using them, but when Eminence decreased the size of the vent to 3/4", I was concerned about vent airspeed and pressure behind the cone. What have you found?


Re: Heat exchanger effectiveness [message #49661 is a reply to message #49660] Fri, 23 June 2006 10:17 Go to previous messageGo to next message
Leland Crooks is currently offline  Leland Crooks
Messages: 212
Registered: May 2009
Master
I have sinks in them. With adequate venting, minimizing the wall thickness and turning a venturi type opening at the cone end, the T/S parameters did not change a bit, except for the better at higher power. My total surface area of the vent holes in the tube exceeds the actual vent size of the c.

I love my C's, and as finances permit will probably relegate the a's to some other cab (Home theater) and buy another pair of c's for the PA subs.

Re: Heat exchanger effectiveness [message #49662 is a reply to message #49661] Fri, 23 June 2006 11:05 Go to previous messageGo to next message
Wayne Parham is currently offline  Wayne Parham
Messages: 18670
Registered: January 2001
Illuminati (33rd Degree)

I may have to re-examine the HL10C device then. I had been working on the 10π design but then shelved it when Eminence dropped the HL10A. I might blow the dust off those plans and make a prototype at some point later this year.

I've been working on the 9π design, which is a top/main, actually a collection of three horns. It's an excellent speaker worthy of complimenting the 12π. Much easier to carry too. ../emoticons/laughie.gif

Re: Heat exchanger effectiveness [message #49719 is a reply to message #49650] Sat, 15 July 2006 14:19 Go to previous messageGo to next message
Tako Tamas is currently offline  Tako Tamas
Messages: 4
Registered: May 2009
Esquire
Hi Wayne,

Did you directly measure the temp of the VC when testing the difference between the original LAB12 and the one with the cooling device?

I mean by measuring the DC resistance or by adding a small temp sensor on the VC?

It would be interesting to know...

Thanks,

Tamas


Re: Heat exchanger effectiveness [message #49724 is a reply to message #49719] Sat, 15 July 2006 18:02 Go to previous messageGo to next message
Wayne Parham is currently offline  Wayne Parham
Messages: 18670
Registered: January 2001
Illuminati (33rd Degree)

Yes, DC resistance was measured and the variance was much less with the heat exchanger installed. You can also see it in the response curve. When measuring response with the cooling plugs installed, the response curve looks the same even at extremely high power levels. Take the cooling plugs out and a low frequency peak appears as power goes up. You can really see the increased resistance in the response curve as the motor gets hot.


Re: Heat exchanger effectiveness [message #49727 is a reply to message #49724] Sun, 16 July 2006 15:05 Go to previous messageGo to next message
Tako Tamas is currently offline  Tako Tamas
Messages: 4
Registered: May 2009
Esquire
Hi Wayne,

Could you give me (us) some details about the resistance measurement and the results?
It could be very interesting...

Thanks,

Tamas

Re: Heat exchanger effectiveness [message #49728 is a reply to message #49727] Mon, 17 July 2006 02:27 Go to previous messageGo to next message
Wayne Parham is currently offline  Wayne Parham
Messages: 18670
Registered: January 2001
Illuminati (33rd Degree)

The links below have several links within that show various tests, setup conditions and results. I've taken measurements at several power levels and durations, and the power levels and temperatures are shown for each. You'll find temperature/power/time measurements in the links in the post called "Heat exchanger effectiveness" (from this thread), as well as a destructive test that sets a baseline for a stock LAB12 without a heat exchanger.
I'd also like to draw your attention to the response charts for the 12π taken at the Prosound Shootout, and compare that with other horns that didn't have a heat exchanger. That will show you a comparison of electro-mechanical shifts. Refer to the response curves of the 12π basshorn at various power levels. Notice that the response curve is rock-solid, changing very little at any power level. Now look at some of the other horns (that didn't have a heat exchanger) from the "Hornsub shootout results", and you'll see their response curves shift. This is particularly noticeable at low frequencies where Qes increases, making a corresponding peak in LF output, and creating a small bass shelf. This is an indication of electro-mechanical parameter shift on the unprotected drivers.

I think the most important thing is how the speaker acts in regards to response and compression, and the fact that it is able to safely handle more power for extended periods of time. Electro-mechanical shift is visible in the response curves at various power levels, seen for example in the test datasets from the Prosound Shootout. The increased DCR value is what causes the response shift and thermal compression when the driver is pushed hard at high power levels.

I'm in Austin right now, and so I'm not where I can run any additional tests or look through my notes right now. But I'd be happy to provide more information when I'm back in Tulsa if you need more data. One of the datasets I measured was DC resistance, another was power/time and another was temperature, so I would be happy to provide this information in whatever format you'd like. Leland Crooks has also done similar testing, using a cooling plug heat exchanger on an HL10 driver. His test results included temperature, DC resistance and Qes, as I recall. I think he has them in an Excel spreadsheet or something so that may be interesting for you. Hopefully he'll see this and post a link to his data here.

Re: Heat exchanger effectiveness [message #49733 is a reply to message #49728] Tue, 18 July 2006 11:26 Go to previous messageGo to previous message
Tako Tamas is currently offline  Tako Tamas
Messages: 4
Registered: May 2009
Esquire
DC resistance results both from you and from Leland would be nice...
BTW How did you measure the DC resistance? I know it is quite simple when the speaker is cold. You just need a good multimeter..
But while it is driven by an AC source it could be a bit difficult.
And when you remove the signal from the speaker, it starts to cool down quite quick....

Thanks,

Tamas


Previous Topic: 12π basshorn subwoofer - the best yet!
Next Topic: Looking for a video
Goto Forum:
  


Current Time: Thu Mar 28 14:04:12 CDT 2024

Sponsoring Organizations

DIY Audio Projects
DIY Audio Projects
OddWatt Audio
OddWatt Audio
Pi Speakers
Pi Speakers
Prosound Shootout
Prosound Shootout
Smith & Larson Audio
Smith & Larson Audio
Tubes For Amps
TubesForAmps.com

Lone Star Audiofest