oosting Posted March 30, 2001 Share Posted March 30, 2001 The speed of sound statement was in reply to : "f the throat were larger and the length the same (with a small increase in mouth size, perhaps) would the cut-off rise to an unacceptable frequency? Specifically, what does the small throat do to sound (velocity too high?)? " Is this non-linear effect larger than the non-linearity of the compression driver at high power levels? If not, you wouldn't hear it. Ed Quote Link to comment Share on other sites More sharing options...
Al Klappenberger Posted March 30, 2001 Author Share Posted March 30, 2001 John, Your distortion equation is quite interesting but, like Ray, I can't get it to give me reasonable results. The numbers come out extremely small for me also. I assumed numbers based on "dope for Hope" information about the K400 horn in the Klipschorn. Throat diameter = .9 Inches Cutoff = 230 Hz. I figure that 2000 Hz would be the point where the human ear is most sensitive to distortion, So: F / Fc would be 8.70. I have also assumed 1 acoustic Watt which should yield 114 dB at 4 feet. That, of course, is EXTREMELY loud! Could you work the equation and see what you get? Al K. Quote Link to comment Share on other sites More sharing options...
Al Klappenberger Posted March 30, 2001 Author Share Posted March 30, 2001 John, I think I solved the confusion. I found the formula and it is wrong in the book! It's by Alexis Badmaieff and it should be: distortion = <1.73 (F/Fc) I^.5> I think the square root of I is the error. It could be written as: I^(1/2) Here's a quick program written in old GWBASIC to do it: 10 CLS 20 INPUT "(<Enter> to end) Throat dia (In)"; DIA 30 IF DIA = 0 THEN END 40 INPUT "Horn cutoff (Hz)"; FC 50 INPUT "At frequency (Hz)"; F 60 INPUT "Power level (W)"; W 70 CM = DIA * 2.54 '* Convert Inches to Cm 80 R = CM / 2 '* Convert diameter to radious in Cm 90 AREA = 3.14159 * R * R '* area of circle = pi times radious squared 100 I = SQR(W / AREA) 110 DIST = (1.73 * (F / FC) * I) 120 PRINT USING "##.####% distortion"; DIST 130 GOTO 20 By this program my example comes out to about 7% distortion. Al K. Quote Link to comment Share on other sites More sharing options...
oosting Posted March 31, 2001 Share Posted March 31, 2001 The equation: 1.73(F/Fc)I/100 implies that the distortion level is inversely proportional to the throat diameter. If a .9 inch throat produces 7% distortion, you would need a 6.3 inch throat to keep the distortion below 1%. A 2 inch throat will produce 3.3% distortion. Does anyone know if this is consistant with measurments of the K55V driver? Ed Quote Link to comment Share on other sites More sharing options...
John Warren Posted April 1, 2001 Share Posted April 1, 2001 Al-- The expression comes from L.L. Beranek's "Acoustics" (McGraw-Hill, 1954). I went from memory screwed up! The proper equation for % 2cd harmonic distortion with I in (Watts/cm. sq.) is 1.73 (F/Fc) SQUARE ROOT Since I is directly related to throat size we see immediately that a larger throat will reduce the second harmonic. The K-400 has 0.9 inch diameter throat (2.29 cm). One acoustic watt at the throat yields a power density of 0.243 W/cm^2. A 2" diameter throat is 0.049 W/cm^2 which is almost 5 times lower power density than a 0.9 in. K-400. Since SQRT<5>= 2.23, the 2" throat reduces 2cd harmonic by 54%. NOT insignificant since 2% mid-band total harmonic distortion is a design goal, 1% or less is were I want to be. If we look at the equation and use Fc=230Hz we can calculate the 2cd harmonic over a range of F's for the 0.9 inch throat K-400 and a 2 inch throat with same Fc (in parenthesis). Remember that anything >3% is noticable distortion! F=4000 14% (6%) both fail F=2000 7% (3%) K55M fails F=1000 4% (2%) K55M fails F= 400 1.5% (0.7%) both pass So based on this model two conclusions can be drawn about the K-400 1. It distorts appreciably above about 1000Hz when played at levels necessary to fill a large room and 2. the Fc for a throat this size is too low. No network mod will cure this. Quote Link to comment Share on other sites More sharing options...
Al Klappenberger Posted April 1, 2001 Author Share Posted April 1, 2001 John, Yes, my mistake, Alexis Badmaieff got the equation from the source you sited.. I agree, a biger throat is better, but I am not willing to do surgery on the K500 in my Belles. You have to show me a commercially available alternitive. Al K. Quote Link to comment Share on other sites More sharing options...
djk Posted April 1, 2001 Share Posted April 1, 2001 A couple of notes here.A speaker with an No=1% is only 92dB/W/1M in half space.The size of the entrance hole of a horn is not usually the throat size.The area of the slots of the phase plug directly adjacent to the diaphragm constitute the throat for pressure distortion calculations.Doing this for a K55 vs an M200 shows that an M200 will have 10dB less distortion at any given frequency and level with respect to the K55.I think I can hear 10dB.As far as SPL goes the worst case situation for the 6Khz crossover on a K55 vs the 3.5Khz crossover on an M200 gives a 6dB lower distortion spec due to reduction in bandwidth.Total worst case distortion would be a total of 16dB higher with the K55.A K400 cut off at 1.375" internal height will mate up with a JBL 2328 2" throat adapter.The 2328 will need to have about .125" filled with bondo and hand rasped to mate with the K400 internal width.To mate with the shorter K500 about .0625" will need to be removed from the 2328.So I guess the question is: How much time and money are you willing to spend for 16dB of distortion reduction in the octave your hearing is most sensitive in? A pair of M200 and a pair of 2328 cost less than $500. Quote Link to comment Share on other sites More sharing options...
John Warren Posted April 1, 2001 Share Posted April 1, 2001 $500 for 16db, I'd sell my mother for a 6db drop!!!!!!! Quote Link to comment Share on other sites More sharing options...
Al Klappenberger Posted April 1, 2001 Author Share Posted April 1, 2001 John and Dennis, In case you were not following the eBay auction on this set of 260 horns, I won the auction and will be getting these to evaluate. I hope to actually take harmonic distortion measurements on them as well as sone frequency plots. My distortion measurement equipment looks like it will have about 1% residual distortion at this point unless I can come up with a better way to do it than I am planning. I will post the results when I get them. I will also try to make comparison measurents on the Belles. AL K. Quote Link to comment Share on other sites More sharing options...
Q-Man Posted April 5, 2001 Share Posted April 5, 2001 DJK, Where can I find this Community M200? I just bought a pair of Altec 288-16K drivers & Altec 805-B Horns to try. They have 1 1/4" throats. I may want to try 2" throuths after reading this stuff. Q. Quote Link to comment Share on other sites More sharing options...
Allan Songer Posted April 5, 2001 Share Posted April 5, 2001 I think you can get just about any driver in production at Filament Audio (www.filamentaudio.com, I think). I don't know the price, but I'll bet they're about $150-$175 each. I just don't have the heart to hack up my pristine old Corwalls--to me it would be akin to dropping a small block Chevy into my '55 Studebaker pickup. The idea of it just makes my skin crawl. For most others it would be the obvious thing to do. Plus, I LIKE the way my Cornwalls sound! But I've got these plans for bulding Klipschorns using state of the art drivers, etc. . . Someday. Quote Link to comment Share on other sites More sharing options...
Q-Man Posted April 5, 2001 Share Posted April 5, 2001 Allan, Thanks for the tip. I just checked Filament Audio. They have diaphragms for community drivers ($149.00), but they don't have the drivers. I'm not modifying my Klipschorns and LaScalas. This is a new build project. Q. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.