Chris8sirhC said:
Almost 4000 L actually due to its relatively high Qes, Vas, though using two drivers in clamshell isobaric will cut it in half to ~1972.5 L with a ~3 dB reduction in LF output, but still too large for all but a built-in.
Hmm, there's nothing 'incorrect' about the sim AFAIK:
GM
Attachments
jnb said:GM, we may have been at crossed purposes back there
Do you know why hornresp shows a falling top end where there is a direct radiating component? The horn should behave as an infinite baffle at high frequencies. I'm expecting the response should plateaux at the rated sensitivity.
Not sure I follow..........
Hmm, it can't behave as an IB at high frequencies due to the terminus' high inertance damping it and the comb filtering of the increasingly out of phase output from both sides of the driver with increasing frequency.
GM
OK. I've noticed this HF response in general as I've asked elsewhere as I think it's OT.
http://www.diyaudio.com/forums/showthread.php?postid=1417162#post1417162
http://www.diyaudio.com/forums/showthread.php?postid=1417162#post1417162
Re: Recent Sims
With today's DVD releases you just about need flat to ~1 Hz, but unless you have the space/budget for an IB woofer wall this isn't practical and with rare exception, ~flat to 16 Hz is good enough, so in theory you would measure your room's gain curve and design a complementary roll off response curve to sum ~flat with it.
Some folks like me prefer to design for the max gain and attenuate any excessive 'boom' for music and leave it as is for movies for a bit more thrilling movie experience. As I previously noted though, a falling response will blend better to most rooms and it allows the cab to be smaller, so probably a better choice for most folks.
GM
Killjoy said:
With today's DVD releases you just about need flat to ~1 Hz, but unless you have the space/budget for an IB woofer wall this isn't practical and with rare exception, ~flat to 16 Hz is good enough, so in theory you would measure your room's gain curve and design a complementary roll off response curve to sum ~flat with it.
Some folks like me prefer to design for the max gain and attenuate any excessive 'boom' for music and leave it as is for movies for a bit more thrilling movie experience. As I previously noted though, a falling response will blend better to most rooms and it allows the cab to be smaller, so probably a better choice for most folks.
GM
Sorry to ask GM (as I consider you better at horn theory than I am), but isn't this compression rate a bit high?
It is interesting how you get to 170Hz useful BW. My gut feeling would tell me to stop earlier, but if you use the +-3dB as a criterium it might work. Is this a visual guess or is there more to it, like some sort of calculation?
P.S. to tb46:
This weekend is a bit busy, but I'll try to post my latest akabak script. It has been dormant for more than a week, but I haven't forgot it. Just haven't got the time for it.
It is interesting how you get to 170Hz useful BW. My gut feeling would tell me to stop earlier, but if you use the +-3dB as a criterium it might work. Is this a visual guess or is there more to it, like some sort of calculation?
P.S. to tb46:
This weekend is a bit busy, but I'll try to post my latest akabak script. It has been dormant for more than a week, but I haven't forgot it. Just haven't got the time for it.
Cordraconis said:
No need to be sorry over asking a question and yes, the CR is very high, but it runs out of Xmax around 42W and neither the driver nor TH is intended for a prosound app, so I seriously doubt there would be a problem WRT damaging a driver rated 300 W peak. I figure the worst case scenario is the surround will deform if driven much past Xmax, which may/may not be audible. Not having actually built a TH though or have any experience with this driver, YMMV.
I just use standard T/S BP4 math to find Fh, Fc based on the desired Fl and Prof. Leach's math for the rest. Indeed, some of the best looking sims I've seen posted defy any of the rules-of-thumb/math I know, so without a computer program to keep trying different dims till it looks good I wouldn't have suggested/considered the drivers used as being suited for the desired BW.
Again, with no TH experience beyond auditioning DSL's DTS-20 that was XO'd below 100 Hz AFAIK, I don't know how high a TH can be XO'd before its HF phasing issues become unacceptable, but it's moot in this particular case since its corner loading will dictate the usable XO point/slope.
Anyway, appreciate yours and other's posting info on Akabak, once I get serious about learning it I'm going to need all the help I can find as it's getting tougher by the day for this 'old dog' to learn new 'tricks'.
GM
Re: Tang Band Comments
Yes, I've often changed a driver's effective specs to make it a better 'fit' to the app and in the good ol' days of field coil drivers the field coil voltage was varied to 'dial in' the driver to the cab/room. Later it was done using the bass boost (aka adjustable DF) on tube amps. Never done it digitally though. More work/parts than just adding a series power resistor, but, then there's no loss of efficiency.
GM
Killjoy said:
Yes, I've often changed a driver's effective specs to make it a better 'fit' to the app and in the good ol' days of field coil drivers the field coil voltage was varied to 'dial in' the driver to the cab/room. Later it was done using the bass boost (aka adjustable DF) on tube amps. Never done it digitally though. More work/parts than just adding a series power resistor, but, then there's no loss of efficiency.
GM
Re: Ciare 15 W2 TH
Well, about as good as any of the acoustically too small, highly resonant TPs and THs so far published. Ideally we want a smoother roll off like the expo version, so the goal is find a driver that's fairly flat over a wide enough BW that a 1/4 order XO is sufficient and fits in a reasonable bulk.
Anyway, another one that when I try different conic sims 'by the numbers' they're all unusable IMO, though looks good as an expo, but this isn't surprising considering its ~1717 L net Vb.
FYI, yours reaches Xmax/20 Hz with ~15 V.
GM
Killjoy said:
Well, about as good as any of the acoustically too small, highly resonant TPs and THs so far published. Ideally we want a smoother roll off like the expo version, so the goal is find a driver that's fairly flat over a wide enough BW that a 1/4 order XO is sufficient and fits in a reasonable bulk.
Anyway, another one that when I try different conic sims 'by the numbers' they're all unusable IMO, though looks good as an expo, but this isn't surprising considering its ~1717 L net Vb.
FYI, yours reaches Xmax/20 Hz with ~15 V.
GM
Attachments
I am thinking of trying a tapped horn for a salvaged 15" MI speaker that I am trying to set up as a sub for my keyboard. I tried a ripole but it really isn't looking too promising. I have noticeable drop off by 50 or 60 Hz. Part of the problem is probably the deep magnet and chassis (can't get very small slot out of it) and very strong magnet.
I have modeled sealed enclosures and the response is pathetic. BR requires a box as big as a horn with less extension and less efficiency.
I don't have the params on me but as best I can remember it measured about .30 Qts, fs 45Hz, Vas about 11cubic feet, Re 6.5 Qme was pretty large too as I recall (over 10). Though not as small as a ripole I think I can get a workable size out of it and get solid down to 40. I am thinking about 4 feet deep by 2 or 2.5 feet tall by 19 inches wide give or take.
I will probably remeasure the driver at some point to be sure that I got the params right but my sense is that TH is fairly forgiving and super flat response is not really necessary for my application.
The odd dimensions are a result of its need to be used on stage where height is much more problematic than depth. Because of the orientation I intend for it to fire out the end rather than the side like many I have seen.
mike
I have modeled sealed enclosures and the response is pathetic. BR requires a box as big as a horn with less extension and less efficiency.
I don't have the params on me but as best I can remember it measured about .30 Qts, fs 45Hz, Vas about 11cubic feet, Re 6.5 Qme was pretty large too as I recall (over 10). Though not as small as a ripole I think I can get a workable size out of it and get solid down to 40. I am thinking about 4 feet deep by 2 or 2.5 feet tall by 19 inches wide give or take.
I will probably remeasure the driver at some point to be sure that I got the params right but my sense is that TH is fairly forgiving and super flat response is not really necessary for my application.
The odd dimensions are a result of its need to be used on stage where height is much more problematic than depth. Because of the orientation I intend for it to fire out the end rather than the side like many I have seen.
mike
15 inch Tapped Horn
zlast 15 inch TH Post #1057
Hi zlast: The fields S1 to S5 are in cm^2 and the lengths are in cm. Your are looking at a throat area of 450cm^2 and a mouth of 750cm^2. In your drawing: The speaker would go into the slanted divider board 50cm from the center of the speaker to the throat (Z). The length 350cm is from the front of the speaker to the back of the speaker, and the remaining length of 50 cm is from the back of the speaker to the mouth (M). This is just one design example and the possibilities are (almost) endless. I would recommend downloading David McBean's Hornresp, it's free, and working through the wiki on how to use it is very informative. Hope this helps.
I downloaded HORNRESPONSE. TO make sure I am applying the values correctly, from your post:
S1= 450
S2 = 479
S3 = 712
CON = 50
CON = 350
CON = 50
I would end up with an enclosure like this? THanks again for the help
zlast 15 inch TH Post #1057
Hi zlast: The fields S1 to S5 are in cm^2 and the lengths are in cm. Your are looking at a throat area of 450cm^2 and a mouth of 750cm^2. In your drawing: The speaker would go into the slanted divider board 50cm from the center of the speaker to the throat (Z). The length 350cm is from the front of the speaker to the back of the speaker, and the remaining length of 50 cm is from the back of the speaker to the mouth (M). This is just one design example and the possibilities are (almost) endless. I would recommend downloading David McBean's Hornresp, it's free, and working through the wiki on how to use it is very informative. Hope this helps.
I downloaded HORNRESPONSE. TO make sure I am applying the values correctly, from your post:
S1= 450
S2 = 479
S3 = 712
CON = 50
CON = 350
CON = 50
I would end up with an enclosure like this? THanks again for the help
Attachments
zlast_15inch_TH
Hi zlast: You have the numbers correct, but the length and areas do not agree with you drawing, also there is only one opening: the mouth. The total length (from throat to mouth) would be 450cm, the box would have to be approximately 38cm longer to accomodate that length. Also, for a box width of 48cm the depth dimension (e.g: 5cm @ the throat) are incorrect (divide 450 by 48). This is also not a finished design, but just a starting point to learn about the design process (I hope). I'll try to find some time tonight to make you a rough drawing. Take a look at Volvotreter's spreadsheet (Post #1058) if you have Excel.
Hi zlast: You have the numbers correct, but the length and areas do not agree with you drawing, also there is only one opening: the mouth. The total length (from throat to mouth) would be 450cm, the box would have to be approximately 38cm longer to accomodate that length. Also, for a box width of 48cm the depth dimension (e.g: 5cm @ the throat) are incorrect (divide 450 by 48). This is also not a finished design, but just a starting point to learn about the design process (I hope
). I'll try to find some time tonight to make you a rough drawing. Take a look at Volvotreter's spreadsheet (Post #1058) if you have Excel.Woud one of you computer savy guys...
consider taking the time to run some simulations (along with a cab drawing) for a Peerless SLS 12 inch driver that I think would be a good candidate for a TH? The unit is a Peerless #830669 and all the T/S data can be found here http://www.tymphany.com/830669 Thanks.
consider taking the time to run some simulations (along with a cab drawing) for a Peerless SLS 12 inch driver that I think would be a good candidate for a TH? The unit is a Peerless #830669 and all the T/S data can be found here http://www.tymphany.com/830669 Thanks.
Hmm, relatively high Vas and Qes + low Xmax = big box for modest output (as TH goes), so best bang/buck appears to be a tapped pipe (TP), though someone willing to trial n' error it might can find a smaller TH for a given BW than ML's math calcs. 311.371 L TP:
GM
GM
Attachments
zlast_15inch_TH
Hi zlast: ref.: Posts #1046, #1048, 1049, 1057 and #1075: Here is a drawing of the enclosure with a single fold. As I said before this is just an example, and it does not quite meet your desired frequency range. Do you have a name for the driver for which you provided the parameters in Post #1046? Obviously one would have to have the physical dimensions for a correct drawing. If you build something like this you may consider screw mounting the front and back panels to give you access to the inside, so that you can make modifications more easily.
Hi zlast: ref.: Posts #1046, #1048, 1049, 1057 and #1075: Here is a drawing of the enclosure with a single fold. As I said before this is just an example, and it does not quite meet your desired frequency range. Do you have a name for the driver for which you provided the parameters in Post #1046? Obviously one would have to have the physical dimensions for a correct drawing. If you build something like this you may consider screw mounting the front and back panels to give you access to the inside, so that you can make modifications more easily.