ProSoundWeb Community

Sound Reinforcement - Forums for Live Sound Professionals - Your Displayed Name Must Be Your Real Full Name To Post In The Live Sound Forums => SR Forum Archives => LAB Subwoofer FUD Forum Archive => Topic started by: Grant Rider on September 11, 2006, 09:38:36 pm

Title: Cooling fan (Patent 4757547)
Post by: Grant Rider on September 11, 2006, 09:38:36 pm
I tried Tom Danley's cooling idea from the patent at http://www.freepatentsonline.com/4757547.pdf. I have some questions. I got cooling plugs and they work great. Now I want to try Danley's idea too. The cooling plug is threaded at one end so it makes it easy to add a pipe fitting for fan air.

I did a experiment and hooked up a fan to give a blast of air. I took Wayne Parham's sugestion to use the cooling plug as a fitting for the fan. The fan is in a funnel with caulk all around so it makes a blast of air out the small end. I hooked that to a tube and used a pipe fiting on my cooling plug.

The air rushing out of the magnet hole blasts with woofer movement even if the fan isn't connected. The air blasts are enough to move my pants leg! It turns the fan just from woofer air blasts!

The fan moves the woofer cone forward. How bad is that? Parham said it makes the speaker suspension stiffer in one direction and looser in the other. How much is too much? I know it makes more distortion and less xmax by shifting offset, but Danley must have figured something out. What can I do about the offset besides taking off the fan?

Does it only work with a specially designed woofer that has a cone that can move further out than in to make up the difference? Maybe a special speaker is a requirement for this to work? If there are exit holes maybe that will help. I think the problem is the restriction at the voice coil gap so the fan air will still push on the back cap a lot but not on front of the cone. That is my dilemma.

Here is what I tried. If I don't use much fan power, the woofer moving air blasts spin the fan by itself, so it isn't much use. If I use more fan power, the fan spins normally but the cone moves out further, sometimes almost as far as it can go. I can change fan speed in between, and that works sort of. The Danley patent uses power from the signal so it won't make the fan go unless power is high. I think the woofer will be pushed forward a lot by the fan when power is high though. That seems like it would still be a problem. Even if power is high, I don't think the cone should be shoved forward. Thoughts anyone?

Title: Re: Cooling fan (Patent 4757547)
Post by: Antone Atmarama Bajor on September 12, 2006, 04:46:51 pm
     I'm not familiar with the particular patent that you are talking about but the cooling fan used in the BT7 does not blow air through a drivers pole piece but through a servomotor, and is vented back out of the box which never interacts with the diaphragm of the driver nor enters the horn chamber.  It runs in parallel with the motor using a rectifier and I think a filter to give DC signal proportional to the audio input.  

Antone-
Title: Re: Cooling fan (Patent 4757547)
Post by: Grant Rider on September 12, 2006, 07:56:25 pm
I can see blowing air through round dc motors with a fan. Open the can and run air through it. It can cool the brushes and windings and doesn't blow the cone forward or backward. What I don't see how to do is use a fan on a regular speaker without pushing the cone with fan air pressure. Wayne Parham said air cooling was a good idea but the problem was how to do without making a pressure differential between the front and back of the cone.

http://www.audioroundtable.com/ProSpeakers/messages/350.html

Tom Danley must have some idea because it is shown in his patent.

http://www.freepatentsonline.com/4757547.pdf.

I want to know how this can work.
Title: Re: Cooling fan (Patent 4757547)
Post by: John Roberts {JR} on September 12, 2006, 08:24:51 pm
Grant Rider wrote on Tue, 12 September 2006 18:56



I want to know how this can work.


I suspect a great deal of benefit could be realized with only modest air flow. Just removing the hot air from inside the box at a regular interval helps so you don't need much pressure. If you know in advance that you will have a pressurized box you can provide an asymmetrical bias in the surround (easier to say than do).

Now just off the top of my pointy head how about using two fans with one fan blowing in while another fan is sucking out. In principle if their flow rates are matched there would be no net pressurization of the chamber.  Very Happy

I suspect the fans will also look like a lossy port so that would need to be factored in for box tuning.

JR


Title: Re: Cooling fan (Patent 4757547)
Post by: Grant Rider on September 12, 2006, 11:52:13 pm
Maybe so. Sounds tricky.  If the fan sucking is very strong it can balance the one that blows. I think it would be hard to balance because the one that blows has a lot of cone area to push against and the one that sucks only pulls on the small area through the voice coil gap. That's a good idea but I think it must be tricky to get right. Fan speed is probably critical too.

Maybe Tom Danley has other ideas we aren't thinking about. Otherwise I guess it needs a tricky balance. I guess I can't do it with a normal driver. A special one has to be made with extra ports or asymmetrical bias in the surround. Darn!


Title: Re: Cooling fan (Patent 4757547)
Post by: Grant Rider on September 13, 2006, 09:41:34 pm
Something I thought about all day keeps bothering me:

John Roberts  {JR} wrote on Wed, 13 September 2006 01:24

I suspect a great deal of benefit could be realized with only modest air flow.


The air rushing out of the magnet hole blasts with woofer movement already. The air blasts are enough to move my pants leg! I think it must do a lot by itself. How can the fan do more with "only modest air flow"?
Title: Re: Cooling fan (Patent 4757547)
Post by: John Roberts {JR} on September 13, 2006, 11:04:54 pm
Grant Rider wrote on Wed, 13 September 2006 20:41

Something I thought about all day keeps bothering me:

=John Roberts  {JR} wrote on Wed, 13 September 2006 01:24 I suspect a great deal of benefit could be realized with only modest air flow.

The air rushing out of the magnet hole blasts with woofer movement already. The air blasts are enough to move my pants leg! I think it must do a lot by itself. How can the fan do more with "only modest air flow"?


At issue is whether it's the same hot air moving back and forth vigorously or new cool air replacing the old hot air.

JR
Title: Re: Cooling fan (Patent 4757547)
Post by: Grant Rider on September 14, 2006, 12:18:28 am
John Roberts  {JR} wrote on Thu, 14 September 2006 04:04

At issue is whether it's the same hot air moving back and forth vigorously or new cool air replacing the old hot air.


I think you are right there. The air must be cool.

The fan isn't needed because the air blasts from the cone are strong. Using the cone as the pump makes the fan not needed. There is no pressure differential offset problem that way either.
Title: Re: Cooling fan (Patent 4757547)
Post by: Al Limberg on September 14, 2006, 12:29:11 am
You might contact the good folks at Community and see if they can provide you with any of their papers concerning their 'Air Force' concert system which used forced air cooling on traditional voice coil drivers.

Al
Title: Re: Cooling fan (Patent 6837333)
Post by: Grant Rider on September 14, 2006, 01:32:11 am
Thanks for the tip!  I found the patent:

http://www.freepatentsonline.com/6837333.pdf

This answers my questions. The Community Light and Sound patent references Danley's patent but says his invention resulted in too much noise and distortion as a result of the air pressure through the gap. That is what I thought was happening too.

I have disconected the fan. Danley's idea was worth a try though!

Do you know if Community still sells Air Force speakers?  I don't see them on their website. Have they been discontinued?
Title: Re: Cooling fan (Patent 4757547)
Post by: John Roberts {JR} on September 14, 2006, 01:34:12 am
Grant Rider wrote on Wed, 13 September 2006 23:18



I think you are right there. The air must be cool.

The fan isn't needed because the air blasts from the cone are strong. Using the cone as the pump makes the fan not needed. There is no pressure differential offset problem that way either.



If you go back to the patent search website and look in the same category or patent class as the one referenced above you will probably find a long list of other related patents. This is not a new problem and smart folks have been working on it for years.

JR
Title: Re: Cooling fan (Patent 6837333)
Post by: Al Limberg on September 14, 2006, 09:27:15 am
From my understanding (and I am NOT the resident expert!) only two complete Air Force Systems were sold.  The cabinets were quite large and very heavy and VERY loud and VERY expensive.  As I recall Community also went thru two different ways of powering the cooling fans, initially running them from the actual power amp signal and then running them from a seperate power supply running thru pins 7 & 8 of an NL8 connector.

Al
Title: Re: Cooling fan (Patent 4757547)
Post by: Tom Danley on September 15, 2006, 12:15:36 pm
Hi

The power cooling trick in the patent you mention was used on all Servodrive BT-7’s and most SDL-4 and SDL-5’s and proved to be very effective.
It can also be applied to a normal loudspeaker if configured as shown in the patent.
Intersonics inc, the company I worked for at the time licensed the patent to Bond acoustics in Canada who was headed by Steve Hutt, now director of R&D at Harmon Motive.
While I was prohibited from using “conventional drivers” most of the time, Steve applied it to a range of products and got more than a factor of two in increased RMS power capacity.
As the power to the fan tracks the input power, it is not difficult to make the fans operation inaudible.

It is important to do the venting as show or risk cone offset proportional to air pressure.
On the other hand, if the air is withdrawn from the VC pocket, then the cone is not affected.
Also, it doesn’t take much flow past the coil to cause a large increase in PHC.
Keep in mind that the “un-glued” temp in a modern VC is around 300 – 350 deg C. that is darn hot and any net flow across that coil causes a nice temperature drop.

In the patent drawing, is shown the power cooling applied to a conventional VC driver, note where the air is drawn out.  
Attached is a photo from the “Olden days” back at Intersonics (around the time I applied for the patent).  
I was demonstrating a switching amplifier and control circuitry on a subwoofer at the time.  
In the foreground was the Eminance 12 inch driver I had modified to add power cooling.
This driver (like most) did not have access to the VC pocket, I drilled a pair of holes and pressed in he copper tubes sticking out the back.
I am on the far left side and John Halliburton (a lab poster) is on the far right.
Hope that helps

Tom Danley
index.php/fa/5880/0/
Title: Re: Cooling fan (Patent 4757547)
Post by: Grant Rider on September 15, 2006, 03:55:03 pm
Thank you for responding.

I hooked up a fan and had problems just blowing air so I took it off. Maybe I will try this again. Maybe if I drill holes like you did and use different size holes or two different fans or something I can make it work. I will try with a cheap woofer first.

The Community patent talks about the problems from pressure difference in your patent. Your patent shows a pressure side and a return side but it does not talk about matching force using different pressures through different size holes. The places air goes through are different so the drag force will be different.

What can be done to balance pressure? The area in front is different than the area in back. It has to be adjusted somehow so the force from the vacuum pulling equals the force from the pressure pushing. I think the vacuum against the gap must be much stronger than the positive pressure against the inside of the cone. The vacuum air pulls through the voice coil gap which drops pressure and reduces force. This makes them unequal. Maybe blow through a tiny hole same area as the gap to match pressure? That way the back vacuum sucking through the gap will be the same as the front pressure blowing through this tiny hole?
Title: Re: Cooling fan (Patent 4757547)
Post by: David Trotter on September 15, 2006, 04:39:27 pm


Seeing as the cone movement is sucking and blowing air in and out of the vent with every push pull, would you not benefit from using a fan blowing across the rear magnet to ensure that the hot air comming out was blown away and replaced with freash air?

has anyone tried cooling with water? what about valves which makes sure the air flow is unidirectional?

I do believe the benefits stated above. Just loading the driver backwards in the BPH horn i designed has meant we can put double the power in and the magnet only gets luke warm.

i for one look forward to day when you can use high-temp superconductors in your voice coil. Very Happy

-dave

Title: Re: Cooling fan (Patent 4757547)
Post by: Tom Danley on September 16, 2006, 09:06:36 pm
Hi Grant, all

The issue of cone offset is a hydraulic one which can be solved as this approach was used on conventional drivers under license.

The cone off set due to pressure is proportional to the pressure per area times the area the pressure is acting on.  If one simply tries to blow or suck air from the center pole vent, then one has some pressure acting on a piston the diameter of the VC.  
By using an air rectifier to “self pump” a net flow,  this tends to be highly asymmetric and “not good” so far as distortion or Qm. One can pump air both ways and that helps.

By drawing air from the rear cavity as I had done in that old old photo, one leaves the vent open and air flows to the VC gap thru the spider and through the center vent.  
Here, the total piston area  is only the VC od area minus the VC ID area (small).
I would suggest you put the fan inside the speaker rear volume, this way you will have no differential pressures on the cone etc.

Other thoughts.
Air is a decent heat insulator and is used in homes (trapped in fiberglass and in foam) but at the same time, once heated, has decent heat energy per volume.
The voice coil, is something like the filament in a light bulb, it is depending on radiant heat which only becomes significant at high temperatures and some conduction through the air gap.
While the magnet can get “hot”, keep in mind the part that matters, can get to 500 –  600 degrees at failure.
Cooling the magnet is somewhat like trying to increase the power dissipation in the light bulb filament by immersing the glass in water.  Yes you can keep it cool and increase the power some, but you are not cooling the part that maters.
Moving air past the “hot part” is very effective and routinely used on many gasoline engines now and in the past.   High performance DC servomotors have been air cooled for the same reason, they have a wire armature suspended in air, within a magnetic field.
In that application, where the armature is limited to 150 deg C, one can nearly quadruple the continuous power dissipation with air cooling.   This, done by only passing air over one side of the armature, not both sides (inside and outside) as in the VC cooling patent drawing.

Try the air path as shown in the drawing, set up a water tube manometer so you can monitor pressure at the VC rear cavity.
Use a shop vac (normally has a universal AC/DC motor) with a Variac to adjust he speed /pressure.
Once you have a few inches of water pressure (the net flow that induces) on the rear pocket, you should have increased the heat capacity significantly.
After that, it is a matter of getting the heat outside of the box in a way that doesn’t interfere acoustically.
Hope that helps,

Tom Danley

Title: Re: Cooling fan (Patent 4757547)
Post by: Wayne Parham on September 17, 2006, 01:04:47 pm

There is no doubt that forced air cooling will do some good.  There is also no doubt that other methods of cooling will do some good too.  Heat can be transferred by any of the three methods, conduction, convection and radiation.

Forced air is employed in the cooling vent, using the speaker motor and cone as the pump.  A fan can be employed, although it is difficult to do it without offset.  Whether positive or negative pressure is used, an offset is possible (probable).  One has to find a mechanism to ensure the forces are equal while at the same time introducing unidirectional flow.  It is possible, but it isn't automatic.  Pulling a vacuum on the gap rather than pressurizing the area behind the cap intuitively seems like it would produce less offset, but there still would be a pressure differential.  My solution was to allow airflow through the gap to move back and forth, but then to force the air passing through into a heat exchanger.  This was done by a device that produced unidirectional flow in the cooling path after the voice coil so no pressure differential existed across the cone.

As for the temperature of the pole piece and magnet, one definitely wants to keep it cool.  Radiated heat transfer is just as legitimate a method of cooling as any other.  It is the one that has been most overlooked by speaker manufacturers and so arguably where the most improvements can be found.  Cooling vents already do a lot to help where forced air is concerned, but when the magnet is 200
Title: Re: Cooling fan (Patent 4757547)
Post by: Michael Hedden Jr. on September 17, 2006, 07:53:55 pm
Didn't I see this in a Jason or Freddy movie or maybe it was that classic "Race with the Devil" where everyone screams run, run for your lives, it's back, the attack of the cooling plug thread!  

Mike Hedden
Title: Re: Cooling fan (Patent 4757547)
Post by: Phil Pope on September 18, 2006, 11:12:19 am
i have tried this cooling fan on a woofer I have

http://www.eighteensound.it/index.aspx?mainMenu=view_product &pid=203

which already has a ring of vent holes behind the voice coil.

I got a cheap rectifier and some capacitors to make the circuit and tried using a cheap 12V computer fan. cut a tube from some 1 3/4" aquarium pipe that fits snug into the pole vent and made a little wooden box to house the fan, components and duct to the pipe.

I have as yet been unable to demonstrate a detectable reduction in power compression.

this is either because

I am not running at a high enough level for fear of damaging my fan as I don't have proper means to power it. Danley's patent shows a resistor to limit the power taken by the fan circuit. this would need to be something like a 50W lightbulb. I am thinking of using a transformer but am having difficulty finding a suitable one locally.

the driver I am using is designed to pump air through itself so it is hard to achieve additional air cooling

the fan I am using is too small 56 cubic feet per min. where do I get high powered fans? all the vacuum cleaners I have looked in are AC motors. I tried looking in my girlfriends hair-dryer and it LOOKS like an AC motor but I think if i take it to bits to check I might get hurt Rolling Eyes

I have tried sticking my shop vac onto the backs of the 18 sound driver which gave relatively little diaphragm movement and a Ciare 12.00SW which moved to about half of xmax. will try on a Lab12 when I open one of the boxes up.

will let you know of any success.

cheers
Phil
Title: Re: Cooling fan (Patent 4757547)
Post by: Tom Danley on September 18, 2006, 02:18:52 pm
Hi Phil

Nice driver, I don’t have one of those but you might check to see “IF” that driver also has vents on the top plate (under the spider).  These holes (that usually accompany the ones on the rear)  allow air to by pass the critical flow area, the gap.
Remember the key to this is making the air path involve passing between the magnet and Voice coil, as in the drawing, it does so on the inside and outside.

If you have an “old” or sacrificial driver available, you can drill into that rear cavity area like I did using a vacuum cleaner running, next to the bit while drilling.
If you do this to a LAB-12, I would find out the magnet ID and drill through as far from the center pole as possible (as this area is likely very High B already).
I would not use hose smaller that say 1 /2 inch ID as the pressures are low and there is net flow.

I would set up a test like I described before.
With air flow, like voltage or sound radiation, there is an “impedance” issue. With DC it is the ratio of voltage to current, with air or “what is the right fan?”, it is Volume to Pressure ratio.
Since you don’t know what you actually need in the beginning, you also need some simple “air” tools, an air meter and variable air power supply.

Make up a water tube manometer so you can keep an eye on what pressure is / does.
This can be as  simple as a  “U” clear plastic tube with some water in it, taped to a yard stick to read from, the measure is in “inches of water” pressure.    
One of the tube is open (or connected somewhere if you want to measure a differential pressure) and the other end goes to your pressure measurement point (at the duct coming out of the speaker).
To start with, use a shop Vac or vacuum cleaner, these nearly always have a universal motor and are perfectly happy running on a variac or variable DC supply.  
This (a variac on a shop vac) gives you an easy to control source of vacuum from none up to about 40 – 60 inches of water depending on the vacuum motor.
You are not dealing with high pressures here, 26.9 inches of water pressure is only 1 psi, you will get good results at less than 1 / 10 of that.
At least in the speakers that were in production using this approach, none had problems with offset.
If I am not mistaken, they used something like a hand held vacuum motor (small DC) and with pink noise had about a 3X increase in capacity over “un cooled”.
Have fun, keep me posted on your results.
Best,

Tom
Title: Re: Cooling fan (Patent 4757547)
Post by: robwells on September 18, 2006, 08:01:49 pm
Michael Hedden Jr. wrote on Mon, 18 September 2006 00:53

Didn't I see this in a Jason or Freddy movie or maybe it was that classic "Race with the Devil" where everyone screams run, run for your lives, it's back, the attack of the cooling plug thread!  

Mike Hedden




surely 'Grant Rider' = 'Wayne Parham' ?

Laughing
Title: Re: Cooling fan (Patent 4757547)
Post by: Wayne Parham on September 19, 2006, 05:19:01 pm

Rob, I don't know that I need to bother with a reply but I have no trouble making my own points.  I don't know if PSW has trouble with sockpuppets, but they're easy enough to identify by IP or even better by MAC ID.  For what it's worth, it looks to me like Grant is pretty open minded.
Title: Re: Cooling fan (Patent 4757547)
Post by: Kerry Stansbury on October 10, 2006, 08:35:03 pm
The Eighteen Sound 15LW1401 does have vents around the spider, a pole vent and six vent holes direct from the voice coil gap. I don't know how much additional forced air would improve it's performance. It is rated at 7000 watts peak! I am currently using a pair of Eighteen Sound 18LW1400's 18" drivers, and they seem to be able to withstand a lot of abuse. I performed some spl comparisons using a frequency generator and an amplifier rated at 1500 watts into 8 ohms operating it just under clipping. With this continuos signal, The 18LW1400 performed fine as did an Aura NRT18-8, Which has a vented spider and pole. The JBL 2242 I tested had only 3 vent holes going directly to the gap which also have some pretty restrictive looking screens on them and no other vents. The JBL smelled like an overheated electrical component within 10 seconds into the test. I hope it didn't damage it, I haven't noticed any after effects "YET". I think with all of the ventilation on the eighteen sound and some other company's drivers like RCF, it may not be nessesary for additional air. I am in the process of trying out a Eighteen sound 15LW1401 which I will be operating at up to 2000 watts. If it fails on me, I will do some forced air testing. I think If you filtered air into the 6 gap vents with just enough to cause the air to move through the gap and out the spider and pole vents, you wouldn't have any problems with pressurization or distortion. Any thoughts on this?
Title: Re: Cooling fan (Patent 4757547)
Post by: Wayne Parham on October 11, 2006, 10:56:45 am

Kerry Stansbury wrote on Tue, 10 October 2006 19:35

The Eighteen Sound 15LW1401 does have vents around the spider, a pole vent and six vent holes direct from the voice coil gap. I don't know how much additional forced air would improve it's performance.

That's the best vent arrangement, in my opinion.  The center pole vent has passage to the gap in the front plate, and air flows through the gap to radial vents in the back plate.  Cone motion produces a lot of airflow through vents arranged like this.  I'm not sure that adding a fan makes sense, especially with the added complexity and cone offset from the pressure differential it produces.
Title: Re: Cooling fan (Patent 4757547)
Post by: Tom Danley on October 11, 2006, 05:30:12 pm
Hi Kerry, Wayne, all

The idea with forced air cooling is that the air that is swept away after being heated carries away energy which would other wise accumulate, causing a temperature rise.
If you look around, you will see air transfer is the most common method available, everything from your car’s radiator to the heat sinks in your amplifiers use this approach, the air carries away the heat energy.  
A typical V-8 car radiator can dump about 50KW of heat and its not that big, it just takes airflow..

Look at the link below which shows the effect of forced air cooling on a DC servomotor.  This motor is similar to that used in a Servodrive and has stationary magnets with a rotating hollow wire armature (very similar to a Voice coil). These motors only have a cooling air path on the outside of the armature where the air path in my cooling Patent is on both the inside and outside of the VC.
In the Servodrives with power cooling (using this patent), cold air was taken in at one edge of the enclosure and hot air expelled at the other, the motor was not in the driver compression chamber.
On the VC drivers licensed under this Patent, the heat once removed from the VC, heated the enclosure and its aluminum back panel.
In the loudspeakers using the approach shown in the patent, they were able to get more than a factor of two in steady state power handing.

For this to work, it is critical for the air to flow thru the magnetic gap (intimate turbulent contact with the VC) as radiation is a poor mechanism at safe VC temps.  Holes that allow air to bypass this path truly defeat the function.   Also, doing it this way, one avoids any significant offset, which is a problem the other ways it could be done.  As I have explained, this way avoids the radiator area feeling any pressure from the cooling system..

Scroll to the bottom of this link and look at the permissible dissipated power and thermal resistance as a function of air flow.  Keep in mind, these are real Watts and not Loudspeaker Watts but the change would be similar (approaching a factor of four improvement). Also keep in mind, the air path is only on one side of the cup armature.
http://www.pacsci.com/support/low_inertia_pmdc/lowinertia55n msupport.html

Hope this helps.

Tom Danley
Title: Re: Cooling fan (Patent 4757547)
Post by: Wayne Parham on October 14, 2006, 09:56:28 pm

Tom, I'm curious:  Has any company you've worked for ever manufactured a voice coil driven speaker with a cooling system as shown in this patent?

Tom Danley wrote on Wed, 11 October 2006 16:30

If you look around, you will see air transfer is the most common method available, everything from your car’s radiator to the heat sinks in your amplifiers use this approach, the air carries away the heat energy.


One of the key things in both of these examples is surface area.  Both are designed to maximize surface area using fins.  They also both use conduction or radiation to carry heat to the fins.

One of the things that reduces the effectiveness of air cooling on loudspeakers is the fact that voice coil surface area is so small.  Some have used conductive formers and other things in an attempt to improve this, but the surface area still remains small.  Cooling vents are certainly worth using, but the fact remains that the relatively small surface area of the voice coil limits the amount of heat that can be removed.  That's why I've suggested that air cooling be augmented by other cooling mechanisms.

Tom Danley wrote on Wed, 11 October 2006 16:30

A typical V-8 car radiator can dump about 50KW of heat and its not that big, it just takes airflow.


Actually, the radiator is huge in terms of surface area.  The whole structure is designed to maximize surface area.

Tom Danley wrote on Wed, 11 October 2006 16:30

Look at the link below which shows the effect of forced air cooling on a DC servomotor.  This motor is similar to that used in a Servodrive and has stationary magnets with a rotating hollow wire armature (very similar to a Voice coil). These motors only have a cooling air path on the outside of the armature where the air path in my cooling Patent is on both the inside and outside of the VC.


Forced air cooling makes sense in a loudspeaker with a motor that is isolated from the diaphragm.  As long as the motor is pneumatically sealed so the pressure from the forced air lines doesn't offset the cone, that's fine.  But in a speaker with a voice coil, the cone is in the same area as the gap.  Unless the speaker is designed with some kind of seal, whatever pressure is placed on the gap is also placed on the cone.  Whether this pressure is positive or negative, on the front or the back, there is still a pressure and that causes offset.

Patents filed after yours have addressed this problem.
Title: Re: Cooling fan (Patent 4757547)
Post by: Tom Danley on October 15, 2006, 12:17:59 pm
Hi Wayne

When I came up with this idea, some 20 years ago now, Intersonics had allowed me to pursue the Servodrive woofer concept but for most of the time I was specifically prohibited from using “conventional” drivers in any design.  
Keep in mind they were a NASA flight hardware contractor, not a speaker company, my real job was working on that stuff much of the time..
They did license the technology to a company in Canada who’s principal in now head of R&D at Harmon Motive.  Also, they did talk to several big companies about licensing but the high license fee and rate the owner asked for stopped that.
I did however build a couple conventional drivers using this and the company that licensed it sold a number of products using it.

I suppose you would have had to have been there to see the setup / measured results to see how effective it is.

You say “They also both use conduction or radiation to carry heat to the fins.”
Actually radiation is not a significant part of transferring heat from either a transistor or engine block to the radiator, it is conduction.  Even amplifier heatsink convection is a conduction driven mechanism, when the air in contact is heated, it rises.

You say “One of the things that reduces the effectiveness of air cooling on loudspeakers is the fact that voice coil surface area is so small. Some have used conductive formers and other things in an attempt to improve this, but the surface area still remains small. Cooling vents are certainly worth using, but the fact remains that the relatively small surface area of the voice coil limits the amount of heat that can be removed. That's why I've suggested that air cooling be augmented by other cooling mechanisms”

Your still missing a major concept here.
Consider your surface area argument and then consider that the motor that I linked has less exposed area than one could have with a modest sized VC.  Even so, a vary large increase of heat capacity is demonstrated in its measurements via passing air directly across the hot conductors.
Like the loudspeaker, the problem is not cooling off the case but cooling off the part producing heat which is much much hotter.   

You keep dwelling on cone offset, perhaps your also missing part of the concept here.
Off set force is equal to area times pressure.
Lets say you had a peak pressure of  2 inches H2O, (this is also a pressure of .074 PSIG).
This pressure if applied as I had shown in the Patent, is applied at the VC pocket.
Lets say you had a 2.5 inch ID VC with a 2.3in ID which yields a piston area of  .753 sq/in on which that pressure is applied.
This results in a peak off set force of  about  25 Gms which to put in perspective for a LAB-12, is about 1/6 the offset force on the system “IF” it were simply mounted facing up or down.
As you can see from the shape of the thermal resistance graph for the DC Servomotor, the most rapid increase in PHC happens when you start moving the air not at its highest flows.

You say “Patents filed after yours have addressed this problem.”
So far as I know, we (Intersonics inc) were the first company to “eliminate” power compression,  when I presented my AES paper on “eliminating power compression”,  Doug B from JBL presented right after one of the first papers to acknowledge its existence.
Later Patents may mention this first one seeing “problems” but I am not away of any of these other approaches resulting in reduction in power compression this one delivered on the Servodrives or in the VC versions Bond Acoustics sold back then.
You can bet you will see the idea of directly cooling the vc with forced air again, it works.

Tom Danley

Title: Re: Cooling fan (Patent 4757547)
Post by: Kerry Stansbury on October 15, 2006, 12:32:18 pm
Tom's design is a start, but forcing air into the pole vent will always cause pressurization against the dust cap and effect the speakers performance. Also, there is no positive air exchange on the voice coil itself, only on the inside of the former. The voice coil will have air pumped back and forth through it, but some of the air will be hot air trapped between the voice coil and the spider. there will be some air drawn away by the fan at the bottom of the voice coil, but I feel it would not be that much of an improvement considering the downside of pressurization.
On the other hand, If you have vents like those that go directly to the voice coil gap, a pole vent and vents around the spider (Like those in "18 Sound" loudspeakers design) You could filter and force air through the gap vents, it would pass directly by the voice coil and the inside of the voice coil former, providing maximum air exchange. The air would the exit the pole vent and the spider vents causing no pressurization as long as you didn't try to pump to much air through.
Title: Re: Cooling fan (Patent 4757547)
Post by: Tom Danley on October 15, 2006, 02:55:39 pm
Hi Kerry

Look again at the drawing, the air connection is at the rear VC pocket, NOT the center pole vent.
In most speakers, one has to drill to this spot (as in the ancient photo earlier in this thread).
If a vacuum is used, then the air path is “in” the center pole vent AND through the spider, into the magnetic gap on the inside and outside. The air flows between the iron and the copper wire on the outside and the iron and former on the inside, then, out the back.

There is no resultant air pressure on the dust cap or cone, the only net force on the system is caused by the cooling pressure and end area of the VC (OD area – ID area).
Some drivers now are made with a vent to this location, like some B&C drivers but the ones I have seen also have a matching set of holes on the front plate.
These defeat the idea when a fan is used as the front holes give a much easier air flow path which doesn’t involve scrubbing up against the hot VC in the gap.
To reach the same gap pressure with these vents, a much larger fan would be needed.

Lastly, one other advantage of the fan system is that its pressure is related to input Voltage where an excursion based (self cooling) system, one finds where heating is greatest, the motion is often least.
Best,

Tom Danley
Title: Re: Cooling fan (Patent 4757547)
Post by: Mike {AB} Butler on October 16, 2006, 11:07:51 am
Alright Tom I see the ILP/Gladstone module (hehe)  Very Happy  What was that doing as part of the testbed? (John H., feel free to chime in..)
I used Gladstone modules and different amps for all sorts of testbeds. They were fine when they worked.. but could fail spectacularly..
Cool to see some early years photos.. I wished I had a few of our first servo'd sub developments in 1980..
Regards,
Title: Re: Cooling fan (Patent 4757547)
Post by: Wayne Parham on October 17, 2006, 01:11:51 pm

Will we see this cooling system on any Danley Sound Lab products?  Since you are the patent holder, it seem natural to me that you would use it on your own products.

Tom Danley wrote on Sun, 15 October 2006 11:17

You say “They also both use conduction or radiation to carry heat to the fins.”
Actually radiation is not a significant part of transferring heat from either a transistor or engine block to the radiator, it is conduction.  Even amplifier heatsink convection is a conduction driven mechanism, when the air in contact is heated, it rises.


My point is that all three types of heat transfer are happening.  In a car engine, the combustion chamber produces heat which is radiated into the heads and cylinder block.  This plus the heat of friction is conducted into the water (if water cooled) or out to the fins (if air cooled).  If water cooled, the heat is carried out to the radiator and heat is exchanged there, convection carries it away from the fins.  If air cooled, the fins are on the block itself.  All three types of heat transfer are in play.

Same thing with a transistor.  Heat is generated as current flows through the junction.  Heat is conducted and radiated to the metal case or tab where it is conducted to a heat sink.  The fins of the heat sink provide convection cooling.

There is no reason to discount any method of heat transfer.  In the case of the loudspeaker, the voice coil is surrounded by a large pieces of metal and ceramic.  Those work as a radiant cooler until they become hot.  That's why removing the heat from the pole piece and magnet structure is important, in addition to air cooling through the vents.  It is counterproductive to allow the magnet and pole piece to reach a high temperature.

Tom Danley wrote on Sun, 15 October 2006 11:17


Your still missing a major concept here.
Consider your surface area argument and then consider that the motor that I linked has less exposed area than one could have with a modest sized VC.  Even so, a vary large increase of heat capacity is demonstrated in its measurements via passing air directly across the hot conductors.
Like the loudspeaker, the problem is not cooling off the case but cooling off the part producing heat which is much much hotter.


No, Tom, I am not missing anything.  I am not saying that air cooling should not be used.  I am stating its limitations, but I am not advocating removing the cooling vents.  What I am saying is that radiant cooling is in play also, if the pole piece and magnet are kept cool.  I am also saying that the amount of heat transferred by radiation is significant.  So what I am advocating is the use of pole piece cooling plugs and heat sinks in addition to cooling vents.
   
Tom Danley wrote on Sun, 15 October 2006 11:17


You keep dwelling on cone offset, perhaps your also missing part of the concept here.
Off set force is equal to area times pressure.
Lets say you had a peak pressure of  2 inches H2O, (this is also a pressure of .074 PSIG).
This pressure if applied as I had shown in the Patent, is applied at the VC pocket.


I get the concept, it's neat.  But if you put a vacuum on the voice coil gap, you'll pull the cone down.  You can increase suspension stiffness to reduce it, but it is still a cone offset and it creates a one-way force that tends to make driver movement asymmetrical.  You could probably use it to balance the asymmetry caused by flux modulation or something, that would be cool.  But the point is that no matter how you slice it, the (positive or negative) pressure required to create cooling air flow also causes a cone offset and an asymmetrical force.
Title: Re: Cooling fan (Patent 4757547)
Post by: John Roberts {JR} on October 17, 2006, 02:04:32 pm
Tom is the inventor but the patent shows as assigned to Intersonics. Tom has posted previously that Intersonics charged a high license fee that discouraged use by others.

Just because you get your name in the inventor box, you don't automatically control that technology. Perhaps ironic when an inventor can't use his own inventions but I know a little about that. Confused

I don't know for a fact that Tom can't, and he can answer for himself. Just thought I'd share the obvious assignment information.

JR  
Title: Re: Cooling fan (Patent 4757547)
Post by: Wayne Parham on October 17, 2006, 02:32:05 pm

Very good point, John.  Is Intersonics still around?  Might be worth a call to his old boss at Intersonics, just to see how he felt about using this technology.  Seems underutilized, nobody is using it as far as I can tell.

Tom said Intersonics had no interest in the voice coil version of the patent.  I'm sure they would license it for a song to Tom if licensing were needed.  Might be a mute point because I think the patent expires next year, but I imagine Tom would contact his old boss anyway, out of courtesy.

I think if Tom is really sold on this technology he could probably use it if he wanted.
Title: Re: Cooling fan (Patent 4757547)
Post by: John Roberts {JR} on October 17, 2006, 02:46:05 pm
Wayne Parham wrote on Tue, 17 October 2006 13:32


Very good point, John.  Is Intersonics still around?  Might be worth a call to his old boss at Intersonics, just to see how he felt about using this technology.  Seems underutilized, nobody is using it as far as I can tell.

Tom said Intersonics had no interest in the voice coil version of the patent.  I'm sure they would license it for a song to Tom if licensing were needed.  Might be a mute point because I think the patent expires next year, but I imagine Tom would contact his old boss anyway, out of courtesy.

I think if Tom is really sold on this technology he could probably use it if he wanted.



They weren't interested before when they also charged too much.

If the patent becomes moot due to expiration in a year, one could start pedalling now because it takes time to tool etc, were one so inclined. It also takes capital and/or a driver mfr with deep pockets willing to pursue an idea that BTW is no longer protected. I suspect Tom has a full plate already but we can always stack on a few more projects.  Laughing

He is probably swiftly moving from the fun part of a new company startup when everything is new and engineering is a blank sheet of paper, to the more work like support of a stable of products in production that need lovin too, but he's got some good crew working with him.

JR
Title: Re: Cooling fan (Patent 4757547)
Post by: Wayne Parham on October 17, 2006, 03:03:32 pm


John Roberts  {JR} wrote on Tue, 17 October 2006 13:46


If the patent becomes moot due to expiration in a year, one could start pedalling now because it takes time to tool etc, were one so inclined.


The best part for Tom is that, as shown in the photo he posted, all he really has to do is hook up a fan and minimal ducting.  The use of his cooling system doesn't require a lot of tooling to be done;  The biggest thing is pressing or threading fittings into the radial vents on the back plate.  The rest is duct work, and can be done with off-the-shelf parts.

Since his tapped horns have both sides of the driver exposed to the mouth, the ducting doesn't have to even be run to the outside.  The rear plate already is outside, so all that is required is a pump to scavenge air from the radial vents.
Title: Re: Cooling fan (Patent 4757547)
Post by: Tom Danley on October 17, 2006, 06:00:54 pm
Hi Wayne, All

Because of the NASA work Intersonics did, the first 13 of my Patents were assigned to that company and then its patent holding company.  
While I am the inventor, I do not own any of them and don’t use any of them currently.

Sounds like John has worn these shoes too, spot on, sorry to hear it.
Well the up side is it was a “learning experience”.

A couple things, radiation is how the heat from a light bulb filament heats the glass or how the heat from combustion is transferred in the “flash” to the metal around it.
Heat does not radiate through metal, think like light, photons that is radiation and it isn’t comparatively significant until something is real pretty hot.  Look up Doug Button’s paper, I think he had the breakdown.
You say” Those work as a radiant cooler until they become hot.” Yet, like light, one can’t radiate cool or darkness, they are simply lower energy states.

My point is that in order for the magnet to become heated, the air in the gap must conduct most of that energy, since the air is already hot, just remove it.
Of the possible things one could do, it offers the largest increase in phc.
If done as in the Servodrive speakers, the cooling ejects the heat from the cabinet altogether so the increase is long term.

Yes there is a tiny asymmetrical force proportional to cooling flow for a VC, but it is acoustically insignificant in practice and in the example above.

Tom
Title: Re: Cooling fan (Patent 4757547)
Post by: Michael 'Bink' Knowles on October 17, 2006, 11:48:24 pm
Quote:

...when I presented my AES paper on “eliminating power compression”,  Doug B from JBL presented right after one of the first papers to acknowledge its existence...


Wow, there's a story there. Doug Button? Is this the mid- to late-'80s? Did Doug ever speak to you again after this embarassment?  Twisted Evil

-Bink
Title: Re: Cooling fan (Patent 4757547)
Post by: Wayne Parham on October 18, 2006, 11:31:55 am

Tom Danley wrote on Tue, 17 October 2006 17:00

A couple things, radiation is how the heat from a light bulb filament heats the glass or how the heat from combustion is transferred in the “flash” to the metal around it.
Heat does not radiate through metal, think like light, photons that is radiation and it isn’t comparatively significant until something is real pretty hot.  Look up Doug Button’s paper, I think he had the breakdown.


Actually, radiated heat is significant.  Incandescence is not required for radiant heat.  Incandescence (light from a hot body) or luminescence (light from a cold body) just means that radiation is in the visible spectrum.  Radiation happens anytime an object is above absolute zero, which is to say all objects are radiant.  When you feel heat from something, it is almost always radiant heat that you feel.  Stand by a fire, a warm car, a hot pavement, even next to a person.  Radiant heat is what you feel, not convected heat from the air.

Tom Danley wrote on Tue, 17 October 2006 17:00

You say” Those work as a radiant cooler until they become hot.” Yet, like light, one can’t radiate cool or darkness, they are simply lower energy states.


Look up radiant coolers.  It's a simple concept.  Any time an object that is cool is placed near an object that is warm, there is radiant heat transfer.  The amount of heat transferred is determined by the temperature difference between the objects, their size and their distance.

Tom Danley wrote on Tue, 17 October 2006 17:00

My point is that in order for the magnet to become heated, the air in the gap must conduct most of that energy, since the air is already hot, just remove it.


That's not true.  I think you misunderstand radiant heat transfer and that's why you are misunderstanding this concept.  The heat that passes into the magnet and pole piece is almost entirely radiated, not convected or conducted.

Tom Danley wrote on Tue, 17 October 2006 17:00

Yes there is a tiny asymmetrical force proportional to cooling flow for a VC, but it is acoustically insignificant in practice and in the example above.


The asymmetrical force is obviously directly related to the air pressure, and therefore cooling air flow.  The more flow, the more force.  That's the main thing that concerns me about this cooling system.

Fortunately for you, it appears that the patent will expire in a year, so you can use it in Danley Sound Labs products very soon if you want.
Title: Re: Cooling fan (Patent 4757547)
Post by: Phil Pope on October 18, 2006, 02:51:58 pm
Wayne Parham wrote on Wed, 18 October 2006 16:31


The asymmetrical force is obviously directly related to the air pressure, and therefore cooling air flow.  The more flow, the more force.  That's the main thing that concerns me about this cooling system.



if you draw air through the small holes round the voice coil then the pressure differential is only acting on the area of the end of the voice coil, right? assume 10 cm diameter coil, assume 3mm thick, gives an area of the order of 10cm^2. assume a pressure of 10cm water gives a force of about 1 Newton, right? as the suspension compliance is usually of the order of 0.1mm/N this should be practically insignificant as long as my reasoning is right.

If I understand correctly there will be far greater problems with offset if you suck air out the big pole vent rather than blow into the small holes.

I haven't done any further tests for lack of suitable blower. my shop vac isn't suitable as it isn't variable and can't be switched to blow easily. can't find much choice of stuff that runs off 110 or 120 Vdc. 12V fans would be OK run off a step down transformer but those I have seen don't have nearly high enough static pressure ratings. can anyone point me in the right direction?

Phil
Title: Re: Cooling fan (Patent 4757547)
Post by: Phil Pope on October 18, 2006, 03:12:06 pm
sorry, the last point I made has already been answered.

Tom Danley wrote on Sun, 15 October 2006 19:55


There is no resultant air pressure on the dust cap or cone, the only net force on the system is caused by the cooling pressure and end area of the VC (OD area – ID area).
Some drivers now are made with a vent to this location, like some B&C drivers but the ones I have seen also have a matching set of holes on the front plate.
These defeat the idea when a fan is used as the front holes give a much easier air flow path which doesn’t involve scrubbing up against the hot VC in the gap.



taping over the holes on the front plate by the spider will make the air cooling work. I thought spiders were fairly porous so there is no reason this will cause any harm is there?

Phil
Title: Re: Cooling fan (Patent 4757547)
Post by: Michael 'Bink' Knowles on October 18, 2006, 05:43:14 pm
http://www.coffeegeek.com/images/6228/200x200/variac.jpg

You could run your shop vac off a variac.

-Bink