If you mean whether or not the woofer in the 12Pi basshorn reaches xmax, no, it does not reach xmax within the passband. Sorry I did not respond - I had already answered that and thought you probably understood my meaning.
Not sure what you are saying, exactly. Better cooling than what?
Last edited:
No problem, I just wanted to make sure you meant Xmax since you used the word ‘mechanical limit’ which often is used for Xmech. Anyway, that does clear the misunderstanding I seem to have about the Lab12 drivers.
That does raise new questions. If the Lab12’s still can’t reach their Xmax even with the ‘cooling-plug’ attached it does mean it suffers from severe power compression. I think that is the same thing what happens in another Eminence driver, the 3015lf, which seems to have similarities in construction besides from the Neo magnets. Have you any idea what the excursion is at max power, with and without your cooling plugs?
Last edited:
No, it's just that horns reduce excursion within the passband, making excursion very small even at high power levels. The driver will overheat before it reaches xmax (and long before xmech).
Hi Wayne,
Post #183: "...Better cooling than what?"
The hypothetical TH probably has better cooling than the hypothetical FLH (see: Posts #179/182).
Post #182 was in response to your Post #180: "...would be interesting would be to see what groups of horns did."
Well, in #179 you have singles, in #182 groups of four.
Post #185: "... horns reduce excursion within the passband, making excursion very small even at high power levels."
In the hypothetical TH and FLH there is very little difference in excursion within the passband. The major difference is the lack of control that is typical for the TH below the passband.
Regards,
Post #183: "...Better cooling than what?"
The hypothetical TH probably has better cooling than the hypothetical FLH (see: Posts #179/182).
Post #182 was in response to your Post #180: "...would be interesting would be to see what groups of horns did."
Well, in #179 you have singles, in #182 groups of four.
Post #185: "... horns reduce excursion within the passband, making excursion very small even at high power levels."
In the hypothetical TH and FLH there is very little difference in excursion within the passband. The major difference is the lack of control that is typical for the TH below the passband.
Regards,
By this, I assume you think cooling might be better if the speaker magnet is exposed to open air. This is a common misconception. I thought the same way - that the air in the rear chamber heated up and reduced the vent's ability to cool the coil. But I discovered this is not true.
If the rear chamber were heated a few hundred degrees, then I'm certain it would make a difference. But that's not the case. The air temperature rise is pretty small compared to the temperature of the motor core. In fact, that's what heats the air, not the voice coil. The biggest contributor is the inductive heating of the motor core which eventually conducts to the back plate and other exposed surfaces, heating the surrounding air. So the best thing you can do is to use a cooling plug to wick the heat out of the core.
The main thing forced air convection needs is velocity - which is a function of excursion. Temperature difference matters too, but the difference between the (300°-400°) hot coil and the ambient (~70°) vented air is great. It almost doen't matter if the ambient air rises 20° or 30°, it's relatively insignificant. You can have the magnet trapped inside a sealed chamber, exposed to outside ambient air, or even put it in an air conditioner and cool it off. The difference in the air temperature doesn't do much to change the effectiveness of the cooling vent.
I know that's sort of counter-intuitive, but that's the way it is.
Without the cooling plug, the maximum is 45vrms, and even at that, the driver will fail after a while. It is pretty vulnerable to heat. If the test frequencies are at places where there is some excursion, the driver will last maybe 1.5 hours. If the test frequencies are anywhere else, the the driver can fail in just a few minutes. A sweep run up through the lower midrange, for example, will kill the driver in short order, sometimes during the sweep.
At 45vrms, the excursion maximum happens around 60Hz, and it is 10mm. It falls below that, but then rises again as the frequency drops through cutoff, below the passband. Excursion through most of the frequency range is under 5mm, at some points under 2mm. Those are places where the driver is most vulnerable to thermal failure.
When the cooling plug is installed, the maximum voltage rises to 60vrms+. The driver will not fail even when run at this level indefinitely, even at places where excursion is at its minimum. As above, the excursion is below 5mm through most of the passband, and under 2mm at some points. At 60Hz, the excursion maximum, it just kisses xmax at 13mm. As before, it falls as frequency drops below that and rises again as the frequency drops through cutoff.
Without the cooling plugs, if the maximum voltage is 45Vrms, then how does the original LABSub handle a couple of thousand watts in actual use. There are many on prosoundweb who will testify that they have been driving each LABSub cab that way for some years now, with minimal or nil failure. Is it because excursion is different in the 12Pi compared to the original?
Quote from Wayne’s post # 165:
“That's sort of what the cooling plug does - it makes a fairly mediocre 400 watt part a little more robust, giving it 800 watt RMS capacity. I guess it's probably more like 1000 watts, when measured by EIA RS-426A, which is a 6dB crest signal. The woofer+plug handles 840 watts continuous sine, which is much harder to handle than EIA-426A. The 12Pi basshorn uses two of those woofers, so in Djim's example, it would be 2000 watts continuous (EIA-426A), 4000 watts program material.”
Now suddenly, in my example your Labs would be able to handle 2000 watts continues and 4000 watts program material for two drivers. Just a quick comparison of the Voice coils of the Lab12 versus B&C 18SW115, I was mentioning before;
Lab12
VC height = 35,53mm
VC diameter = 63,5mm
B&C 18SW115
VC height = 34mm
VC diameter = 116mm
I understand that in the enthusiasm of the moment you needed some ‘extra’ but I don’t think your drivers share your enthusiasm knowing that the VC of the Lab is 43% smaller in relation to the B&C’s VC.
Quote from Wayne’s website:
“Each woofer in the 12π has been tested to 840 watts continuous RMS power with the heat exchanger inserted. So since the 12π bashorn sub has two woofers, it is safely rated at 1600 watts, continuous. That's also the point where compression sets in and power increases don't result in SPL increases. In testing, I ran at levels above 1600 watts and at frequencies well below horn cutoff specifically to test the mode of failure, to know whether it was mechanical or thermal. The failure mode is thermal, and that's why the heat exchanger is so important to the design.”
Quote from Wayne’s post # 188:
“When the cooling plug is installed, the maximum voltage rises to 60vrms+. The driver will not fail even when run at this level indefinitely, even at places where excursion is at its minimum. As above, the excursion is below 5mm through most of the passband, and under 2mm at some points. At 60Hz, the excursion maximum, it just kisses xmax at 13mm. As before, it falls as frequency drops below that and rises again as the frequency drops through cutoff.”
In both, on your website as well in post # 188, you seem to give a more realistic number, which I assume are all based on 60V with an Re of 4.29Ohm = 839Watt and “kisses Xmax at 13mm”.
As you state on your website that after Xmax power compression begins, suggest there is no power compression before 1600 watts. Maybe you found a unique feature, since I never seen it, but I'll ask you again so you have the possibility to respond again;
1.) Does or does it not reaches the Xmax with 839Watt in your 12Pi?
2.) What is the power compression figure at Xmax?
3.) Doesn’t the THD increase as well in relation to power compression?
4.) Isn't it true that after 839 watt at 60Hz your Horn start to drop its LF content while upband, where the excursion is fairly low, you only will see an increase after 839Watt per driver + a rapid rise in THD figures?
Of course I understand Music signals allow higher pwr ratings and have lower excursions but you can’t scratch away the fact in numbers between the voice coils of the Lab’s and the B&C. Besides the 43% size deference the B&C also has a more advanced internal airflow that cools the much larger VC surface that will cool much better in all circumstances compared to the 'tiny' Lab12's VC and motor compartment.
If the horn suppresses excursion so much as you stated, by the high compression ratio of the system, and the input power will be heat up the VC to its thermal limit, that is a sign of power compression and it happens way before Xmax will be reached. That counts for all horns and TH's as well. So denying there isn't any power compression below Xmax says more about the way you measured than the reliability of your figures. Just the way you question somebody else his measurements.
In my view, and that of all modern standards used by companies who manufacture extended LF drivers of the latest generation, your ‘toast-tests’ as you call them would only give a realistic figures in combination with complex signals (noise type with crest factor 6 for instance) that represent dynamic music signals. Of course your 'cooling-plug' shows admirable figures under continues clean sine-wave signals, maybe even better than extended LF drivers of the latest generation. But these' new' style drivers are all designed for maximum performance around Xmax in relation to complex dynamic signals and linear movement.
Like I said before with drivers it is all about their balance. Spocks logic 😉: "If you start lifting the floor, sooner or later you will bump your head"
“That's sort of what the cooling plug does - it makes a fairly mediocre 400 watt part a little more robust, giving it 800 watt RMS capacity. I guess it's probably more like 1000 watts, when measured by EIA RS-426A, which is a 6dB crest signal. The woofer+plug handles 840 watts continuous sine, which is much harder to handle than EIA-426A. The 12Pi basshorn uses two of those woofers, so in Djim's example, it would be 2000 watts continuous (EIA-426A), 4000 watts program material.”
Now suddenly, in my example your Labs would be able to handle 2000 watts continues and 4000 watts program material for two drivers. Just a quick comparison of the Voice coils of the Lab12 versus B&C 18SW115, I was mentioning before;
Lab12
VC height = 35,53mm
VC diameter = 63,5mm
B&C 18SW115
VC height = 34mm
VC diameter = 116mm
I understand that in the enthusiasm of the moment you needed some ‘extra’ but I don’t think your drivers share your enthusiasm knowing that the VC of the Lab is 43% smaller in relation to the B&C’s VC.
Quote from Wayne’s website:
“Each woofer in the 12π has been tested to 840 watts continuous RMS power with the heat exchanger inserted. So since the 12π bashorn sub has two woofers, it is safely rated at 1600 watts, continuous. That's also the point where compression sets in and power increases don't result in SPL increases. In testing, I ran at levels above 1600 watts and at frequencies well below horn cutoff specifically to test the mode of failure, to know whether it was mechanical or thermal. The failure mode is thermal, and that's why the heat exchanger is so important to the design.”
Quote from Wayne’s post # 188:
“When the cooling plug is installed, the maximum voltage rises to 60vrms+. The driver will not fail even when run at this level indefinitely, even at places where excursion is at its minimum. As above, the excursion is below 5mm through most of the passband, and under 2mm at some points. At 60Hz, the excursion maximum, it just kisses xmax at 13mm. As before, it falls as frequency drops below that and rises again as the frequency drops through cutoff.”
In both, on your website as well in post # 188, you seem to give a more realistic number, which I assume are all based on 60V with an Re of 4.29Ohm = 839Watt and “kisses Xmax at 13mm”.
As you state on your website that after Xmax power compression begins, suggest there is no power compression before 1600 watts. Maybe you found a unique feature, since I never seen it, but I'll ask you again so you have the possibility to respond again;
1.) Does or does it not reaches the Xmax with 839Watt in your 12Pi?
2.) What is the power compression figure at Xmax?
3.) Doesn’t the THD increase as well in relation to power compression?
4.) Isn't it true that after 839 watt at 60Hz your Horn start to drop its LF content while upband, where the excursion is fairly low, you only will see an increase after 839Watt per driver + a rapid rise in THD figures?
Of course I understand Music signals allow higher pwr ratings and have lower excursions but you can’t scratch away the fact in numbers between the voice coils of the Lab’s and the B&C. Besides the 43% size deference the B&C also has a more advanced internal airflow that cools the much larger VC surface that will cool much better in all circumstances compared to the 'tiny' Lab12's VC and motor compartment.
If the horn suppresses excursion so much as you stated, by the high compression ratio of the system, and the input power will be heat up the VC to its thermal limit, that is a sign of power compression and it happens way before Xmax will be reached. That counts for all horns and TH's as well. So denying there isn't any power compression below Xmax says more about the way you measured than the reliability of your figures. Just the way you question somebody else his measurements.
In my view, and that of all modern standards used by companies who manufacture extended LF drivers of the latest generation, your ‘toast-tests’ as you call them would only give a realistic figures in combination with complex signals (noise type with crest factor 6 for instance) that represent dynamic music signals. Of course your 'cooling-plug' shows admirable figures under continues clean sine-wave signals, maybe even better than extended LF drivers of the latest generation. But these' new' style drivers are all designed for maximum performance around Xmax in relation to complex dynamic signals and linear movement.
Like I said before with drivers it is all about their balance. Spocks logic 😉: "If you start lifting the floor, sooner or later you will bump your head"
Last edited:
Cooling etc
Re - Excursion at minima
Minimum excursion at certain frequencies/impedances, resulting in low cooling due to being driven at selected frequencies, is IMO not very realistic. Sure with test tones, but with music playing bass sounds that vary in pitch and/or level, the bandspread of frequencies will move the cone at more than just the minima, and therefore help to provide cooling.
I do believe that Wayne Parham should be congratulated for Actually putting into practice some very good ideas on how to maximise output and long term survivability via heatsinking 🙂
I'm surprised that most driver manufacter havn't addressed this issue more thoroghly before now ? One that did identify the problem, and then went some way in attempting to allieviate it was, Mckenzie Acoustics of Yorkshire UK, back in the late 1970's early 1980's. Several of their higher powered drivers came with a black Finned Heatsink attached around the magnet area & part way around the back. I asked Brian Mckenzie if it was just cosmetic, or did it really do much, and was told that it most definately made a difference. They were also one of the first, if not first, companies to introduce double rear suspensions.
Re - Excursion at minima
Minimum excursion at certain frequencies/impedances, resulting in low cooling due to being driven at selected frequencies, is IMO not very realistic. Sure with test tones, but with music playing bass sounds that vary in pitch and/or level, the bandspread of frequencies will move the cone at more than just the minima, and therefore help to provide cooling.
I do believe that Wayne Parham should be congratulated for Actually putting into practice some very good ideas on how to maximise output and long term survivability via heatsinking 🙂
I'm surprised that most driver manufacter havn't addressed this issue more thoroghly before now ? One that did identify the problem, and then went some way in attempting to allieviate it was, Mckenzie Acoustics of Yorkshire UK, back in the late 1970's early 1980's. Several of their higher powered drivers came with a black Finned Heatsink attached around the magnet area & part way around the back. I asked Brian Mckenzie if it was just cosmetic, or did it really do much, and was told that it most definately made a difference. They were also one of the first, if not first, companies to introduce double rear suspensions.
The difference is the 45vrms figure is the maximum voltage the LAB12 will safely handle as a continuous sine. More specifically, that's what it will handle when excursion is allowed to rise to ~10mm, as at relatively low frequency in a cabinet that doesn't reduce excursion. It will handle less when used at higher frequency or in a cabinet that limits excursion.
A continuous sine is much more difficult for a speaker to handle than most music, which has a variety of frequencies and what not. Music usually has some "pauses" in between high power signals to allow cooldown time.
This is also why the EIA-426A spec (which uses a pink noise signal with a crest factor of 6dB) is used to rate speakers. It is more like music. A continuous sine has a crest factor of 3dB, and is much harder on them. The LAB12 will handle 400 watts of pink noise for over 8 hours, which is what the EIA-426A rating requires. But it will fail after about an hour and a half when given a low frequency sine wave of 400 watts, and will fail after just a few minutes if given a 400 watt sine wave if excursion is limited. These are just figures, things to know when designing or using the speaker, things that help us better know how it will act and what to do with it.
If your music is not bass-heavy, then the LABhorn will easily handly "2000 watts", because it is never really seeing this power level, at least, not anywhere close to a continuous level. However, if it has throbbing bass that is always there, especially if it is at certain frequencies, then the LABhorn will fail at less power. Most all failures are thermal. Add cooling plugs and the likelihood of failure is greatly diminished.
Hi,
I second Zero D's motion that: "...Wayne Parham should be congratulated for Actually putting into practice some very good ideas on how to maximise output and long term survivability via heatsinking." He should also be congratulated for an extensive effort in documentation, and making a fine product available to anyone who wants to build it.
I also remember companies trying to fight heatrise in the pole pieces (just don't remember the names) through using laminated poles, and through attaching forward pointing heatsinks (Eminence). I have no idea how successful these attempts finally were.
As to the excursions, Hornresp-which doesn't take into consideration losses or power compression-gives a pretty good picture of the drivers' excursion, so maybe we can go with that. There just isn't a big difference between properly designed TH and FLH in the passband. As to heating, there has to be very considerable voice coil heating at the impedance minima in addition to the inductive heating of the magnet structure, e.g.: @ about 40Hz the impedance drops below 2.3 Ohm for the 12Pi in Hornresp; @ 60Vrms that will not bode well for the drivers' survival when using constant frequency sine waves. The Hornresp excursion plots also indicate a definite need for low cut filtering (even with music).
Regards,
I second Zero D's motion that: "...Wayne Parham should be congratulated for Actually putting into practice some very good ideas on how to maximise output and long term survivability via heatsinking." He should also be congratulated for an extensive effort in documentation, and making a fine product available to anyone who wants to build it.
I also remember companies trying to fight heatrise in the pole pieces (just don't remember the names) through using laminated poles, and through attaching forward pointing heatsinks (Eminence). I have no idea how successful these attempts finally were.
As to the excursions, Hornresp-which doesn't take into consideration losses or power compression-gives a pretty good picture of the drivers' excursion, so maybe we can go with that. There just isn't a big difference between properly designed TH and FLH in the passband. As to heating, there has to be very considerable voice coil heating at the impedance minima in addition to the inductive heating of the magnet structure, e.g.: @ about 40Hz the impedance drops below 2.3 Ohm for the 12Pi in Hornresp; @ 60Vrms that will not bode well for the drivers' survival when using constant frequency sine waves. The Hornresp excursion plots also indicate a definite need for low cut filtering (even with music).
Regards,
Attachments
Last edited:
eminence kilomax... hmm... Read through the Pi patent... kinds seems similar?? no? In fact the statement "may be included as an integral part of the loudspeaker assembly" in the detailed description makes me think the kilomax is directly patent infringing.
different question -- wayne -- have you worked/tested with any other driver with cooling plug besides the lab12?
Hi Oliver, 18Sound also had an optional heatsink for their first line of Neo drivers, ND730 series , that were released around 2003. These heatsinks could be attached by four screws on the back of the driver.
If the kilomax is infringing, then the patent wouldn't likely hold water- off the top of my head I think the large-format phase plug cooling used in the eminence magnum series predates the cooling plug patent. It was specifically intended for cooling.
What I'd like to see is careful thermal coupling to a metal cone. Best would be a one-piece metal cone and VC former, but then you've got a shorted turn- I don't think splitting the former would do the trick if it were conducting to the cone.
What I'd like to see is careful thermal coupling to a metal cone. Best would be a one-piece metal cone and VC former, but then you've got a shorted turn- I don't think splitting the former would do the trick if it were conducting to the cone.
Hi jbell,
Yes the Eminence Kilomax. Thanks.
Hi Djim,
Interesting that they would have an additional detachable heatsink. Neodymium suffers somewhat from its relatively low curie temperature. I've look at it for some downhole applications, but it just will not get to our maximum temperature.
Regards,
Yes the Eminence Kilomax. Thanks.
Hi Djim,
Interesting that they would have an additional detachable heatsink. Neodymium suffers somewhat from its relatively low curie temperature. I've look at it for some downhole applications, but it just will not get to our maximum temperature.
Regards,
You mean you are going to drill illegally for oil in your backyard and use your swimming pool as camouflage for the activities😀?
McKenzie Heatsinks
I've managed to grab a photo of a heatsinked McKenzie driver to show how they looked. Not the sharpest picture, but i'm sure you get the idea 😉
Remember these were first introduced way back in the late 1970's early 1980's by McKenzie, so ahead of their time 🙂 Not many people "appear" to have realised the significance of this, at the time 🙁 Ah well it's only taken about 20 -30 years for "some" others to catch up 😉
I've managed to grab a photo of a heatsinked McKenzie driver to show how they looked. Not the sharpest picture, but i'm sure you get the idea 😉
Remember these were first introduced way back in the late 1970's early 1980's by McKenzie, so ahead of their time 🙂 Not many people "appear" to have realised the significance of this, at the time 🙁 Ah well it's only taken about 20 -30 years for "some" others to catch up 😉
Attachments
- Status
- Not open for further replies.