Hi
This makes me think about:
-First obviously inductors are not recommended for high power subs for professional use. Active is the way for it.
-What say about the voice coils of the loudspeakers. Or the formula does not functions for the voice coil wire also?
-When used that way (2000W) what are the numbers of the power compression. In exceed of 8?
-So I've to agree with BF. Need more? Double the boxes and power you will get 6dB and less PC instead of much less dBs aplying excessive power.
Regards,
This makes me think about:
-First obviously inductors are not recommended for high power subs for professional use. Active is the way for it.
-What say about the voice coils of the loudspeakers. Or the formula does not functions for the voice coil wire also?
-When used that way (2000W) what are the numbers of the power compression. In exceed of 8?
-So I've to agree with BF. Need more? Double the boxes and power you will get 6dB and less PC instead of much less dBs aplying excessive power.
Regards,
GM said:
I did exactly that with a 10kg reel of 14g enamelled copper wire I saved from a skip in my previous job. But high-value low-resistance air-cored inductors need a *ludicrous* amount of copper wire to get low enough resistance, transformer cored ones are probably better so long as you can stop them saturating.
Ian
iand said:
I have all the magazines here. The article is a very big one, it spreads over 5 magazines. I´l try to derive the essentials and post them here.
Exactly 🙂 In most reflex-Boxes the impedance of the system in interaction with the coil also changes the response.
Sabbelbacke said:
The Erse XQ air-cored coils certainly have enough power handling and low enough resistance -- the EAV79-10-4700 is 0.27ohms for 4.7mH, but at 150mm diameter x 38mm high I hate to think how much it weighs (3kg?) and costs (?????)
"-What say about the voice coils of the loudspeakers. Or the formula does not functions for the voice coil wire also?"
No, it most certainly does. If you run 10A through a typical 8R speaker (DCR of 6R), the heat will be 600W (for a short period of time). The wire heats up, the resistance goes up, power compression sets in.
If we maintain the initial 60V into our 6R example, and the voice-coil impedance rises to 12R, the speaker is now only consuming 300W, or about 3dB power compression (ignoring mechanical loses).
This would be typical of a 4" voice coil driver with a vented gap. Newer designs with under spider vents and heatsink fins may push out a few more watts. Note that this data is for the AES standard 5 minute test, long enough to make the voice-coil double in impedance, and go 3.9dB into compression (when it is actually consuming only 300W at that point).
http://www.jblpro.com/pub/components/2206.pdf
No, it most certainly does. If you run 10A through a typical 8R speaker (DCR of 6R), the heat will be 600W (for a short period of time). The wire heats up, the resistance goes up, power compression sets in.
If we maintain the initial 60V into our 6R example, and the voice-coil impedance rises to 12R, the speaker is now only consuming 300W, or about 3dB power compression (ignoring mechanical loses).
This would be typical of a 4" voice coil driver with a vented gap. Newer designs with under spider vents and heatsink fins may push out a few more watts. Note that this data is for the AES standard 5 minute test, long enough to make the voice-coil double in impedance, and go 3.9dB into compression (when it is actually consuming only 300W at that point).
http://www.jblpro.com/pub/components/2206.pdf
djk said:
All absolutely correct, but this isn't the real problem with inductors because they're usually less than 1/10 the the resistance of the voice coil and don't get anywhere near as hot.
The problem with cored inductors is nonlinearity, especially for high power PA applications (which is what the question was about I think). Air cored inductors don't have this problem but need to be absolutely massive to keep the resistance down -- about 500cm3 of copper for the one I quoted.
Ian
"All absolutely correct, but this isn't the real problem with inductors because they're usually less than 1/10 the the resistance of the voice coil and don't get anywhere near as hot."
None the less, the inductors do burn up before the drivers do.
The speakers in question were only 200W rated Eminence, the claimed power handling on the inductors was '500W'.
Dance club use can be brutal on speakers.
I've had good luck with the laminated steel ones Madisound sells, but they will eventually start smoking if you run high enough current through them as well. Compounding the problem is the lower temperature insulation on the inductors vs the voice-coil wire.
None the less, the inductors do burn up before the drivers do.
The speakers in question were only 200W rated Eminence, the claimed power handling on the inductors was '500W'.
Dance club use can be brutal on speakers.
I've had good luck with the laminated steel ones Madisound sells, but they will eventually start smoking if you run high enough current through them as well. Compounding the problem is the lower temperature insulation on the inductors vs the voice-coil wire.
djk said:
If the inductors are burning up they've got too much resistance, regardless of what the claimed "power handling" is (which is often understated).
The massive air-cored inductor I quoted is 0.27ohms DC so will dissipate 3.4% of the power of the driver with 8ohms average impedance, or about 70W at 2000W into the driver. Given that it's a coil 6" diameter and 1.5"x1.5" cross-section I expect it's perfectly capable of dissipating this much heat.
But very few (more likely, no) speaker manufacturers or builders are willing to cough up $100-$200 (my guess -- can anyone confirm this?) for such an inductor, so they penny-pinch and the inductor fries. This is a specification problem, not an inherent inductor problem.
On the other hand almost no metal-cored inductors will still be linear at these power levels -- though of course neither are the drivers, so there's no point worrying about 1% distortion due to the inductors. But if the inductor gets close to saturation -- which many will at these power levels -- then big problems arise, not just a bit of distortion...
So you can build high-quality high-power passive crossovers, but it isn't half expensive :-(
Ian
I bought some 700w 4mh inductors for my 3015lf's.
At 100w with sine waves they got pretty darn hot.
I've decided not to use them at all. Too risky.
At 100w with sine waves they got pretty darn hot.
I've decided not to use them at all. Too risky.
To tap or not to tap?
Several months ago I posted here the proposed plans to convert my (already built) tl into a tapped horn. After posting but before cutting I compared the tapped proposal to a ported box, same driver and tuning, and discovered that the ported box matched the tapped horn in spl and also offered far greater bandwidth on the top end (obviously), in a box about 1/3 smaller. Due to the fact that it's already built I wasn't able to optomize it much so I abandoned the idea until further notice.
Meanwhile, although I promised not to buy any new drivers I found a deal that was too good to pass up for a few of the partsexpress tang 6.5 inch drivers with 13 mil xmax. While I'm waiting for them to arrive I'm playing with ideas. First reflex instinct is to model them in a ported box. Next I sought out Volvotreter's tapped horn for these drivers (not the same model number but they model the same). Finally, I decided to mock up a more "Danley"esque version with wider bandwidth.
PORTED - 16.6L, 215w
VOLVOTRETER TAPPED HORN - 45L, 52/212 (throat/mouth area in cm), 165w
LARGE TAPPED HORN - 64L, 30/300, 130w
Both tapped horns have 4mH total Le, .25 rg and all 3 are modelled as corner loaded. Subtract a few db for 2 pi, subtract a few more for halfspace. All boxes shown with enough power to reach xmax.
Very interesting stuff. First off, the ported box is shown with 215w applied to reach xmax. I would not expect this driver to actually be able to thermally handle that kind of power, so that graph is probably just theoretical best case wishful thinking. Add in driver and port compression under these conditions and you can subtract a few more db for good measure. At only 16.6L (including port volume), this box is TINY with fairly large (long)ports, which interestingly enough cause almost as much trouble up higher with it's single spike as the dreaded tapped horn dual spikes as seen in attached tapped horn pics.
Next up, Volvotreter's excellent 30 hz tapped horn, to which I added a coil to flatten response a bit. Here is an obvious gain in spl and a huge reduction in bandwidth. Pick your poison. Both of these tapped horns will still suffer driver thermal compression (although not as much - or not as soon - with "only" 165w this time) but won't suffer from port compression. 3x larger than the ported box.
And finally a wider bandwidth tapped horn. In my very limited experience (I have not modelled many tapped horns and could be wrong about this assumption) this requires a larger mouth:throat ratio, which in turn requires a larger box for the same tuning. Hence the 10:1 ratio of mouth to throat and the size (4x the ported box). Is the cone strong enough for that kind of load? I don't know, but I do know that this tapped horn needs a couple of Danley style resonators inside to iron out the 2 spikes at the top end but offers 30 - 150 hz bandwidth. "Only" 130w to reach xmax. BTW, the flat response was not by design and the design was not really optomized at all (5 minutes invested so far). I could make a slightly rising response to fit room gain better but I would have to tune the box lower. Or possibly I might be able to get the slightly rising response from making the box a bit smaller.
Does anybody have any ideas about the "best" tapped horn for this driver? Is my design good? Can it be improved?
120 db (theoretically) from a single corner loaded 6.5 inch driver from 30 - 150 hz in 64L. Can we do better?
The driver - tang band w61139si - http://www.parts-express.com/pe/showdetl.cfm?Partnumber=264-832
Volvotreter's tapped horn page - http://volvotreter.de/th.htm
Several months ago I posted here the proposed plans to convert my (already built) tl into a tapped horn. After posting but before cutting I compared the tapped proposal to a ported box, same driver and tuning, and discovered that the ported box matched the tapped horn in spl and also offered far greater bandwidth on the top end (obviously), in a box about 1/3 smaller. Due to the fact that it's already built I wasn't able to optomize it much so I abandoned the idea until further notice.
Meanwhile, although I promised not to buy any new drivers I found a deal that was too good to pass up for a few of the partsexpress tang 6.5 inch drivers with 13 mil xmax. While I'm waiting for them to arrive I'm playing with ideas. First reflex instinct is to model them in a ported box. Next I sought out Volvotreter's tapped horn for these drivers (not the same model number but they model the same). Finally, I decided to mock up a more "Danley"esque version with wider bandwidth.
PORTED - 16.6L, 215w
An externally hosted image should be here but it was not working when we last tested it.
VOLVOTRETER TAPPED HORN - 45L, 52/212 (throat/mouth area in cm), 165w
An externally hosted image should be here but it was not working when we last tested it.
LARGE TAPPED HORN - 64L, 30/300, 130w
An externally hosted image should be here but it was not working when we last tested it.
Both tapped horns have 4mH total Le, .25 rg and all 3 are modelled as corner loaded. Subtract a few db for 2 pi, subtract a few more for halfspace. All boxes shown with enough power to reach xmax.
Very interesting stuff. First off, the ported box is shown with 215w applied to reach xmax. I would not expect this driver to actually be able to thermally handle that kind of power, so that graph is probably just theoretical best case wishful thinking. Add in driver and port compression under these conditions and you can subtract a few more db for good measure. At only 16.6L (including port volume), this box is TINY with fairly large (long)ports, which interestingly enough cause almost as much trouble up higher with it's single spike as the dreaded tapped horn dual spikes as seen in attached tapped horn pics.
Next up, Volvotreter's excellent 30 hz tapped horn, to which I added a coil to flatten response a bit. Here is an obvious gain in spl and a huge reduction in bandwidth. Pick your poison. Both of these tapped horns will still suffer driver thermal compression (although not as much - or not as soon - with "only" 165w this time) but won't suffer from port compression. 3x larger than the ported box.
And finally a wider bandwidth tapped horn. In my very limited experience (I have not modelled many tapped horns and could be wrong about this assumption) this requires a larger mouth:throat ratio, which in turn requires a larger box for the same tuning. Hence the 10:1 ratio of mouth to throat and the size (4x the ported box). Is the cone strong enough for that kind of load? I don't know, but I do know that this tapped horn needs a couple of Danley style resonators inside to iron out the 2 spikes at the top end but offers 30 - 150 hz bandwidth. "Only" 130w to reach xmax. BTW, the flat response was not by design and the design was not really optomized at all (5 minutes invested so far). I could make a slightly rising response to fit room gain better but I would have to tune the box lower. Or possibly I might be able to get the slightly rising response from making the box a bit smaller.
Does anybody have any ideas about the "best" tapped horn for this driver? Is my design good? Can it be improved?
120 db (theoretically) from a single corner loaded 6.5 inch driver from 30 - 150 hz in 64L. Can we do better?
The driver - tang band w61139si - http://www.parts-express.com/pe/showdetl.cfm?Partnumber=264-832
Volvotreter's tapped horn page - http://volvotreter.de/th.htm
Your 64L version looks similar to some things I've modeled up for the TB W8-740c, and in those versions the compression ratio (ratio between Sd and throat area) typically was very high. Conventional wisdom is that you can't really push this ratio up over about 4 for most drivers. If you post your Hornresp input screens, it would make concrete feedback a bit easier.
One thing you definitely should do is look back a few pages (page 95 according to my bookmark) at the double-driver W8 horn that JLH built, and the preliminary design of a similar unit for the W6 that someone posted. If I ever find the second W6 of my pair, I'd be very tempted to try something like that.
One thing you definitely should do is look back a few pages (page 95 according to my bookmark) at the double-driver W8 horn that JLH built, and the preliminary design of a similar unit for the W6 that someone posted. If I ever find the second W6 of my pair, I'd be very tempted to try something like that.
Input screen.
Thanks for the comments guys.
That's true. But doesn't a higher ratio usually also mean higher spl/higher accoustic load/increased pressure against the cone? In this case the flare rate goes way up but spl (and I would assume cone pressure) remain the same. Maybe? Looking at the 2 graphs, one is 4:1 and the other is 10:1 but max spl remains the same, in fact the smaller one looks a bit louder.
I'll check into it and see what I can do, thanks.
An externally hosted image should be here but it was not working when we last tested it.
Thanks for the comments guys.
That's true. But doesn't a higher ratio usually also mean higher spl/higher accoustic load/increased pressure against the cone? In this case the flare rate goes way up but spl (and I would assume cone pressure) remain the same. Maybe? Looking at the 2 graphs, one is 4:1 and the other is 10:1 but max spl remains the same, in fact the smaller one looks a bit louder.
I'll check into it and see what I can do, thanks.
After muckin about in hornresp a bit I'm not seeing any advantage to using a compression chamber yet...
I'm having trouble beating the 64L design posted previously with a design using a compression chamber.
I'm having trouble beating the 64L design posted previously with a design using a compression chamber.
For the sake of comparison, I checked the design on page 96 and made a couple of changes. Changed tuning a couple hz higher to get the same 30 hz tuning as all previous examples. Changed the design to a single driver design for easy comparison. Changed the flare rate a bit just for fun, and a couple of other minor changes. As you can see the compression chamber helping much.
tapped horn with compression chamber, 65L, 130w
Excursion and max power (to reach xmax) are basically exactly the same, bandwidth is virtually identical, the box is the same size, it's basically the same thing done slightly differently. I played for almost 1/2 hour and can't optomize this much more.
Here's the input screen for this thing.
tapped horn with compression chamber, 65L, 130w
An externally hosted image should be here but it was not working when we last tested it.
Excursion and max power (to reach xmax) are basically exactly the same, bandwidth is virtually identical, the box is the same size, it's basically the same thing done slightly differently. I played for almost 1/2 hour and can't optomize this much more.
Here's the input screen for this thing.
An externally hosted image should be here but it was not working when we last tested it.
My buddy and I just finished (well, almost) an 18" driver version of this sub-horn. It's 8'-3" tall, has O.D. 21" walls. Material is 1-1/8" sub-floor decking plywood.
We ran the inner baffle corner to corner, rather than sidewall to sidewall. This way, there's no parallel walls (almost, just a sliver near the port).
We used the older Mach-5 18" driver.
Next weekend, we'll brace the inner baffle and taper the top of the cabinet with 45 degree reflectors to prevent standing waves top to bottom.
The sound is amazing! Just as advertised. This unit and driver combination has a natural roll off and requires no crossover. We fired it up in his garage to Pink Floyd's Breath. Everything in that garage rattled or squeaked. My, oh my. I actually found myself checking the rafters for objects that might fall and bonk me.
Regardless of its natural roll off, I'll be using the sub-out port on my receiver and internal x-over, but his receiver is 32 years old and doesn't have this feature.
I have 4 Mach-5 18" drivers in two large bass-reflex cabinets in my great-room and they don't possess the depth and authority of just one of these sub-horns.
I will be fabricating smaller versions for friends with smaller rooms.
I can't begin to express how important it is to get out there and make some sawdust. 😉 The payoff is enormous! 😀
We ran the inner baffle corner to corner, rather than sidewall to sidewall. This way, there's no parallel walls (almost, just a sliver near the port).
We used the older Mach-5 18" driver.
Next weekend, we'll brace the inner baffle and taper the top of the cabinet with 45 degree reflectors to prevent standing waves top to bottom.
The sound is amazing! Just as advertised. This unit and driver combination has a natural roll off and requires no crossover. We fired it up in his garage to Pink Floyd's Breath. Everything in that garage rattled or squeaked. My, oh my. I actually found myself checking the rafters for objects that might fall and bonk me.
Regardless of its natural roll off, I'll be using the sub-out port on my receiver and internal x-over, but his receiver is 32 years old and doesn't have this feature.
I have 4 Mach-5 18" drivers in two large bass-reflex cabinets in my great-room and they don't possess the depth and authority of just one of these sub-horns.
I will be fabricating smaller versions for friends with smaller rooms.
I can't begin to express how important it is to get out there and make some sawdust. 😉 The payoff is enormous! 😀