Modeling MEHs using HornResp
Hi David,
I am trying to model a two-way MEH using HornResp. I intend to use a large format compression driver and two 15” woofers wired in parallel and mounted on each side of a large horn (Klipsch K-402). I will use two entry ports per woofer (i.e., 4 in total on each horn).
In the MEH input file, I shall specify Ap1 = the 'throat port cross-sectional area' [for the off-set woofers].
Q1: Is the Ap1 value to be inserted, the TOTAL throat port cross-sectional area for all 4 entry ports (or just for the two ports per driver)?
Q2: I have to specify the number of drivers for the MEH1. In my case, would that then be ‘2P’? [since I use two woofers wired in parallel per horn]
Thanks!
Best regards
Peter
Hi David,
I am trying to model a two-way MEH using HornResp. I intend to use a large format compression driver and two 15” woofers wired in parallel and mounted on each side of a large horn (Klipsch K-402). I will use two entry ports per woofer (i.e., 4 in total on each horn).
In the MEH input file, I shall specify Ap1 = the 'throat port cross-sectional area' [for the off-set woofers].
Q1: Is the Ap1 value to be inserted, the TOTAL throat port cross-sectional area for all 4 entry ports (or just for the two ports per driver)?
Q2: I have to specify the number of drivers for the MEH1. In my case, would that then be ‘2P’? [since I use two woofers wired in parallel per horn]
Thanks!
Best regards
Peter
Last edited:
Hi
First i want to thank you for this site and the huge knowledge we can find in it, a escuse myself for my english
I have a question: Can Hornresp calculate PPSL and how ?
Sorry the question has been already answered, but the 963 pages a quite prohibitive for a complete look up
Thanks from France
First i want to thank you for this site and the huge knowledge we can find in it, a escuse myself for my english
I have a question: Can Hornresp calculate PPSL and how ?
Sorry the question has been already answered, but the 963 pages a quite prohibitive for a complete look up
Thanks from France
Hi Peter,
If the mid frequency drivers are located in the S2 plane, then the distance is given by L12 in the Nd record. If the drivers are located in the S3 plane, then the distance is given by L12 + L23, as shown by the red line in the attachment.
It is the distance to the centre of the entry port.
(If a driver has 2 ports each of area A, then Hornresp assumes a single composite port of area 2 x A, centred on the driver axis).
Kind regards,
David
If the mid frequency drivers are located in the S2 plane, then the distance is given by L12 in the Nd record. If the drivers are located in the S3 plane, then the distance is given by L12 + L23, as shown by the red line in the attachment.
It is the distance to the centre of the entry port.
(If a driver has 2 ports each of area A, then Hornresp assumes a single composite port of area 2 x A, centred on the driver axis).
Kind regards,
David
Attachments
Hi fouinies,
Not that I am aware of.
I know that a number of users have attempted to use Hornresp to simulate PPSL type loudspeakers, but I am not sure how successful they have been.
https://www.diyaudio.com/forums/subwoofers/177905-thread-ppsl-enclosures.html#post2377161
Kind regards,
David
Karlson Enclosure Simulation Model
I didn't think that Ivo was seriously suggesting that a Karlson enclosure option be included in Hornresp, which is why I haven't responded before now, but just in case anyone is under the impression that it might be a possibility - sorry to disappoint you, but it is not going to happen
.
I understand that some users have attempted to approximate the Karlson design using Hornresp, but I don't know how successful they have been, and if they have compared their simulation results to the measured performance of an actual loudspeaker. I would be surprised if an accurate model of a Karlson system could be specified using the available Hornresp inputs.
So after all this bandpass stuff, isn't it time to add the karlson-slot?
And I want this too!
I didn't think that Ivo was seriously suggesting that a Karlson enclosure option be included in Hornresp, which is why I haven't responded before now, but just in case anyone is under the impression that it might be a possibility - sorry to disappoint you, but it is not going to happen
.I understand that some users have attempted to approximate the Karlson design using Hornresp, but I don't know how successful they have been, and if they have compared their simulation results to the measured performance of an actual loudspeaker. I would be surprised if an accurate model of a Karlson system could be specified using the available Hornresp inputs.
Matthew Morgan J's constricted transflex kluge for the Acoustic Control 115BK looked pretty close (115BK is approximately the same cabinet as KK Audio, Transylvania Power co. , Westwood, and more ! and all roughly the size of Karlson's "X15" system, introduced in 1965 - the "story" goes that Mrs. K sold out licences to multiple parties. One party says his company was approached by JBL saying to desist building or be sued . . . )
SIM
https://i.imgur.com/RND3rmR.png
SIM
https://i.imgur.com/RND3rmR.png
I didn't think that Ivo was seriously suggesting that a Karlson enclosure option be included in Hornresp, which is why I haven't responded before now, but just in case anyone is under the impression that it might be a possibility - sorry to disappoint you, but it is not going to happen
I understand that some users have attempted to approximate the Karlson design using Hornresp, but I don't know how successful they have been, and if they have compared their simulation results to the measured performance of an actual loudspeaker. I would be surprised if an accurate model of a Karlson system could be specified using the available Hornresp inputs.
Hello David, I actually was serious, but was trying not to come across too pushy.
But I didn't mean a karlson-enclosure, rather the slot itself. A karlson slot has, in the end, a similar goal and fuction to a horn expansion. There is a paper by Poppe I think (am I right, Freddy?) where a a horn and pipe-with-karlson-slot are made for the same driver and compared. Indeed, they performed similarly.
Everybody messes about with approximations, but I think it would be interesting and relevant to have it as an option in hornresp.
I see Marty C. Poppe passed away October 31 2016
Martin Poppe Obituary - Burlington, VT | The Burlington Free Press
"The K-Coupler"
http://home.planet.nl/~ulfman/files/poppe_kcoupler.pdf
some years back, Lars Moseholm did some testing and partial analysis of a little "klam" like the one in my old video
YouTube
Lars Moseholm's 01 report on the Weiss "Rocket"
Posted by freddyi (A) on February 14, 2010 at 14:19:22
Text of Lars report from Job Ulfman's Karlson Speaker Project forum
LAR’S REPORT #5 ON WEISS PS2000 K-COUPLER
"-Dear Group,
The PS2000 taper rate is 4.23 1/m corresponding to a frequency cut-off 116 Hz.
A horn for this rate suitable to the dimensions of the PS2000 (throat area St=500 cm^2
[77.5 in^2] and length L=30 cm [11.8 in]) would have a mouth area Sm=1760 cm^2 [273 in^2].
For comparison the area of the tapered opening of the PS2000 is 256 cm^2 [40 in^2]
and the area of the terminating PS2000 'mouth' is Smk=90 cm^2 [14 in^2].
The horn radiates sound determined by the mouth real impedance.
The PS2000 radiates through the real impedance of the tapered opening, and the end area Smk
(nose area). If the horn is short, or if a lot of energy is left behind in the wave front
before it reaches the nose area in the K-coupler, a lot of reflections take place.
This is well known to all of you. Some more information on the PS2000 speaker.
The mechanical resistance of the driver motor is 5.7 N*sec/m. The mechanical resistance of the driver
suspension losses is 1.6 N*sec/m. The mechanical resistance of the back chamber load (vol. = 5 liters)
on the speaker is 1.65 N*sec/m.
Those resistances determine the SPL together with the real mechanical impedance of the throat.
The optimal (horn) throat area for max. efficiency of the system is 85 cm^2 [13 in^2] which
would give a max. SPL 1W/1m close to 107 dB. The actual PS2000 St is 4 times larger, lowering the SPL.
Based on well known horn equations I estimated the complex input impedance of the above horn.
Then I applied the Poppe equations (which also are the well known horn equations if there
is no loss through the tapered opening), but adjusted for loss - and this is where the transmission
line theory comes in handy - , and from that I also estimated the throat complex impedance of the PS2000.
I had at this point to assume that the K-coupler was a tube
(I have not worked out the clam design yet - so it is a first approximation).
It should also be noted that the PS2000 K-coupler is very short. For all energy to emit through
the tapered opening the length should be about 90 cm [34 in] for that specific taper rate,
so one would expect a lot of reflections.
I found the k_coupler acoustical characteristic capacitance to be Co=9.9*10-^8 m^5/N
and inductance Lo=6.6 N*s^2/m^5.
Due to the tilt of the panel (and square form) Co would decrease and Lo would increase compared
to the tube form assumed. The overall effect would be a relative increase in
throat impedance in the clam type box compared to a tube. At the same time more energy would be forced
out through the tapered opening.
The resulting SPL for the horn as well as the PS2000 was about 103-105 dB.
The cut-off for the horn was the assumed 116 Hz, however, the cut-off for the K-coupler
was - as expected - shifted upwards to about 200 Hz.
I clearly found a broad lower K-coupler resonance centered about 400 Hz (100 Hz above measured,
first approximation) and the upper peaks and dips. Also the plots did not show the on-axis response
below the cut-off where the speaker may behave more like a closed box design.
In conclusion: Yes, the PS2000 K-coupler has some significant horn characteristics,
and no, it is not optimized.
We only begin to understand the theory, however, we have some rough models to play around with.
I may have made a lot of errors! However, I have scanned the plots and will send them together
with the contour plot to anybody interested (Fred and Carl have asked).
I think can get them through the (local) fire-wall by know.
Martin Poppe Obituary - Burlington, VT | The Burlington Free Press
"The K-Coupler"
http://home.planet.nl/~ulfman/files/poppe_kcoupler.pdf
some years back, Lars Moseholm did some testing and partial analysis of a little "klam" like the one in my old video
YouTube
Lars Moseholm's 01 report on the Weiss "Rocket"
Posted by freddyi (A) on February 14, 2010 at 14:19:22
Text of Lars report from Job Ulfman's Karlson Speaker Project forum
LAR’S REPORT #5 ON WEISS PS2000 K-COUPLER
"-Dear Group,
The PS2000 taper rate is 4.23 1/m corresponding to a frequency cut-off 116 Hz.
A horn for this rate suitable to the dimensions of the PS2000 (throat area St=500 cm^2
[77.5 in^2] and length L=30 cm [11.8 in]) would have a mouth area Sm=1760 cm^2 [273 in^2].
For comparison the area of the tapered opening of the PS2000 is 256 cm^2 [40 in^2]
and the area of the terminating PS2000 'mouth' is Smk=90 cm^2 [14 in^2].
The horn radiates sound determined by the mouth real impedance.
The PS2000 radiates through the real impedance of the tapered opening, and the end area Smk
(nose area). If the horn is short, or if a lot of energy is left behind in the wave front
before it reaches the nose area in the K-coupler, a lot of reflections take place.
This is well known to all of you. Some more information on the PS2000 speaker.
The mechanical resistance of the driver motor is 5.7 N*sec/m. The mechanical resistance of the driver
suspension losses is 1.6 N*sec/m. The mechanical resistance of the back chamber load (vol. = 5 liters)
on the speaker is 1.65 N*sec/m.
Those resistances determine the SPL together with the real mechanical impedance of the throat.
The optimal (horn) throat area for max. efficiency of the system is 85 cm^2 [13 in^2] which
would give a max. SPL 1W/1m close to 107 dB. The actual PS2000 St is 4 times larger, lowering the SPL.
Based on well known horn equations I estimated the complex input impedance of the above horn.
Then I applied the Poppe equations (which also are the well known horn equations if there
is no loss through the tapered opening), but adjusted for loss - and this is where the transmission
line theory comes in handy - , and from that I also estimated the throat complex impedance of the PS2000.
I had at this point to assume that the K-coupler was a tube
(I have not worked out the clam design yet - so it is a first approximation).
It should also be noted that the PS2000 K-coupler is very short. For all energy to emit through
the tapered opening the length should be about 90 cm [34 in] for that specific taper rate,
so one would expect a lot of reflections.
I found the k_coupler acoustical characteristic capacitance to be Co=9.9*10-^8 m^5/N
and inductance Lo=6.6 N*s^2/m^5.
Due to the tilt of the panel (and square form) Co would decrease and Lo would increase compared
to the tube form assumed. The overall effect would be a relative increase in
throat impedance in the clam type box compared to a tube. At the same time more energy would be forced
out through the tapered opening.
The resulting SPL for the horn as well as the PS2000 was about 103-105 dB.
The cut-off for the horn was the assumed 116 Hz, however, the cut-off for the K-coupler
was - as expected - shifted upwards to about 200 Hz.
I clearly found a broad lower K-coupler resonance centered about 400 Hz (100 Hz above measured,
first approximation) and the upper peaks and dips. Also the plots did not show the on-axis response
below the cut-off where the speaker may behave more like a closed box design.
In conclusion: Yes, the PS2000 K-coupler has some significant horn characteristics,
and no, it is not optimized.
We only begin to understand the theory, however, we have some rough models to play around with.
I may have made a lot of errors! However, I have scanned the plots and will send them together
with the contour plot to anybody interested (Fred and Carl have asked).
I think can get them through the (local) fire-wall by know.
Last edited: