Hi @David McBean
I’m getting to grips with using Hornresp to simulate Ripoles. I used your example of the settings for a Ripole and they make sense, with the front port being made up of 2 x the smaller volume and each rear port being the larger volume. The target being the smallest possible volume to move the resonance peak as far up the frequency range as possible (>300hz if possible)
I have attached your original screenshots from post #13372 which I used as the basis of entering the data for a Ripole.
My main issue is how to convert the settings into actual construction dimensions. Playing with PAR, as the only value that seems to alter the response by altering the front and rear volumes are the 2 PAR values. I’m not sure what point this is measuring from. I assume it’s cm? (Why is it “16” in your example?) e.g, Do I measure this from the port to the centre of the driver or is it the distance from the from the front port to the rear port (phase cancelation distance). The drawing elongates the port but I’m not sure how to interpret this.
For now my process has been the following:
1/ Set the minimum port width (1 front and 2 rear) that allows the drivers (a pair of peerless 12” 840500) to physically fit in the box. In my case this is 130mm for each port, which is too large for the response I’m targeting with regards the resonance peak.
2/ use the area of the baffle (inside) multiplied by the port width to find the port volume.
3/ Alter the PAR value to create the port volume to match the required volume.
I feel that essentially this is backwards to what the design process should be, where the software provides the optimal dimensions rather than the other way around.
Any help and clarification from anyone re PAR and the optimal process is welcomed!
Thanks, Gis
I’m getting to grips with using Hornresp to simulate Ripoles. I used your example of the settings for a Ripole and they make sense, with the front port being made up of 2 x the smaller volume and each rear port being the larger volume. The target being the smallest possible volume to move the resonance peak as far up the frequency range as possible (>300hz if possible)
I have attached your original screenshots from post #13372 which I used as the basis of entering the data for a Ripole.
My main issue is how to convert the settings into actual construction dimensions. Playing with PAR, as the only value that seems to alter the response by altering the front and rear volumes are the 2 PAR values. I’m not sure what point this is measuring from. I assume it’s cm? (Why is it “16” in your example?) e.g, Do I measure this from the port to the centre of the driver or is it the distance from the from the front port to the rear port (phase cancelation distance). The drawing elongates the port but I’m not sure how to interpret this.
For now my process has been the following:
1/ Set the minimum port width (1 front and 2 rear) that allows the drivers (a pair of peerless 12” 840500) to physically fit in the box. In my case this is 130mm for each port, which is too large for the response I’m targeting with regards the resonance peak.
2/ use the area of the baffle (inside) multiplied by the port width to find the port volume.
3/ Alter the PAR value to create the port volume to match the required volume.
I feel that essentially this is backwards to what the design process should be, where the software provides the optimal dimensions rather than the other way around.
Any help and clarification from anyone re PAR and the optimal process is welcomed!
Thanks, Gis
Attachments
For the Ripole folded baffle shown in the attachment,
If:
H = internal height of the enclosure in cm
D = internal depth of the enclosure in cm
W1 = width of the front opening in cm
W2 = width of each rear opening in cm
Then for the example you gave:
S1 = (2 * W2) * H = total cross-sectional area of the two rear openings
S2 = total cross-sectional area of the rear-facing ducts at the middle of the driver
S2S = S2
S3 = total cross-sectional area at the closed ends of the rear-facing ducts
S3S = cross-sectional area at the closed end of the front-facing duct
S4 = cross-sectional area of the front-facing duct at the middle of the driver
S4S = S4
S5 = W1 * H = cross-sectional area of the front opening
L12 = distance from the middle of the driver to the rear of the enclosure
L23 = D - L12 = distance from the middle of the driver to the front of the enclosure
L34 = L12
L45 = L23
In your example, L12, L23, L34 and L45 are the values shown in the input boxes alongside the respective Par labels.
Hornresp will simulate the design you input, it does not provide optimal dimensions. I am not aware of any program that can calculate the optimal dimensions for a Ripole loudspeaker, given the driver.
Your post would perhaps have been better placed in the main Hornresp thread - that way, there would be no risk of me not reading it 🙂.
Attachments
Thank you very much, I will progress from here 🙂
I will post any further questions in the main hornresp forum.
Thanks again for the detailed response.
Gis
I will post any further questions in the main hornresp forum.
Thanks again for the detailed response.
Gis
Hi @David McBean
Further to your reply I’m struggling to include the following variable unless I’m missing something basic;
The reason I was working backwards from the systems port volumes, was that I don’t have “square” volumes with a hight x width rear of the port, but rather follow the circumference of the woofer to create a curved rear of the port. This (should help) stop waves bouncing around on the way out of the port, but more importantly reduces the volume of the port, which in turn moves the resonant peak up in frequency relative to the size of the woofer.
Following your example, I’m not sure how to incorporating this design feature in the figures as I now have a fixed distance to the rear of the port but don’t have a HxW for the rear of the 3 ports.
This principle is the main difference between the original ripole design and the ModalAkustik version which has the rear of the 3 ports curved. By moving the mouth of each port in or out, the volume is able to be tuned as well as the distance between the front and rear which should move the f3 up or down. (Out = peak down and f3 down, In = peak up and f3 up) .
My plan was to use horn real to find the balance where I attain the lowest response for the available max excursion
The only difference in my design is that my 2 drivers are facing the same way, rather than facing each other to further null out any resonances.
Desing showing areas to calculate:
A + woofer = base port area
B = port area extension to fine tune (can be 0)
C = relevant port and port width
Port volume = A + woofer + B x C
D = back of port, now irrelevant?
E = port(s) on other side
F = distance from front to rear port (should influence phase cancellation?)
Btw, should I move this post to the main Hornresp tread (if that’s possible)
Thanks
Gis
Further to your reply I’m struggling to include the following variable unless I’m missing something basic;
The reason I was working backwards from the systems port volumes, was that I don’t have “square” volumes with a hight x width rear of the port, but rather follow the circumference of the woofer to create a curved rear of the port. This (should help) stop waves bouncing around on the way out of the port, but more importantly reduces the volume of the port, which in turn moves the resonant peak up in frequency relative to the size of the woofer.
Following your example, I’m not sure how to incorporating this design feature in the figures as I now have a fixed distance to the rear of the port but don’t have a HxW for the rear of the 3 ports.
This principle is the main difference between the original ripole design and the ModalAkustik version which has the rear of the 3 ports curved. By moving the mouth of each port in or out, the volume is able to be tuned as well as the distance between the front and rear which should move the f3 up or down. (Out = peak down and f3 down, In = peak up and f3 up) .
My plan was to use horn real to find the balance where I attain the lowest response for the available max excursion
The only difference in my design is that my 2 drivers are facing the same way, rather than facing each other to further null out any resonances.
Desing showing areas to calculate:
A + woofer = base port area
B = port area extension to fine tune (can be 0)
C = relevant port and port width
Port volume = A + woofer + B x C
D = back of port, now irrelevant?
E = port(s) on other side
F = distance from front to rear port (should influence phase cancellation?)
Btw, should I move this post to the main Hornresp tread (if that’s possible)
Thanks
Gis
Attachments
Updated version of the above:
@David McBean
Further to your reply I’m struggling to include the following variable unless I’m missing something basic. So I’ll try as much as I can to explain where I’m (still) stuck.
The reason I was working backwards from the systems port volumes, was that I don’t have “square” volumes with a hight x width at the rear of the ports, but rather follow the circumference of the woofer to create a curved rear of the port (see my crude illustration). This (should help) stop waves bouncing around on the way out of the port, but more importantly reduces the volume of the chamber, which in turn moves the resonant peak up in frequency relative to the size of the woofer.
Following your example, I’m not sure how to incorporate this curve at the back of the ports in the figures as I now have a fixed distance to the rear of the port (centre of the woofer to the rear wall of the port is now an arc, which will likely change your calculations for the response as this changes the chamber volume.) but don’t have a height for these 3 rear walls of the ports.
This principle of the curved rear chamber is the main difference between the original ripole design and the ModalAkustik version which has the rear of the 3 ports curved. By moving the mouth of each port in or out, the volume is able to be tuned as well as the distance between the front and rear which should move the f3 up or down. (Out = peak down and f3 down, In = peak up and f3 up) .
My plan was to use horn real to find the balance where I attain the lowest response for the available max excursion and the highest resonance peak (more important imo)
The only difference in my design is that my 2 drivers are facing the same way, rather than facing each other to further null out any resonances.
Design showing areas to use to calculate response.
A + woofer = base port area (for port chamber volume calculation)
B = port area extension to fine tune (can be 0)
C = relevant port and port width
Port chamber volume = A + woofer + B x C
D = back of port, now irrelevant?
E = port(s) on other side
F = distance from front to rear port (should influence response through phase cancellation?)
Woofer = SD 466cm2, Xmax +/-12.5mm but target <+/-10mm
David, please note that I’m fully aware that I’m not telling you anything new with respect to the variable’s impacts on the response calculated by hornresp. I’m just trying to be as clear as possible in showing my interpretation so that this is of value to anyone that is trying to come to grips with the variables of a ripole implementation. So if I’m making any kind of clear blunders here, please don’t hesitate to point these out, I’m here to learn 😬🥴
Btw, should I move this post to the main Hornresp tread (if that’s possible)
Thanks
Gis
@David McBean
Further to your reply I’m struggling to include the following variable unless I’m missing something basic. So I’ll try as much as I can to explain where I’m (still) stuck.
The reason I was working backwards from the systems port volumes, was that I don’t have “square” volumes with a hight x width at the rear of the ports, but rather follow the circumference of the woofer to create a curved rear of the port (see my crude illustration). This (should help) stop waves bouncing around on the way out of the port, but more importantly reduces the volume of the chamber, which in turn moves the resonant peak up in frequency relative to the size of the woofer.
Following your example, I’m not sure how to incorporate this curve at the back of the ports in the figures as I now have a fixed distance to the rear of the port (centre of the woofer to the rear wall of the port is now an arc, which will likely change your calculations for the response as this changes the chamber volume.) but don’t have a height for these 3 rear walls of the ports.
This principle of the curved rear chamber is the main difference between the original ripole design and the ModalAkustik version which has the rear of the 3 ports curved. By moving the mouth of each port in or out, the volume is able to be tuned as well as the distance between the front and rear which should move the f3 up or down. (Out = peak down and f3 down, In = peak up and f3 up) .
My plan was to use horn real to find the balance where I attain the lowest response for the available max excursion and the highest resonance peak (more important imo)
The only difference in my design is that my 2 drivers are facing the same way, rather than facing each other to further null out any resonances.
Design showing areas to use to calculate response.
A + woofer = base port area (for port chamber volume calculation)
B = port area extension to fine tune (can be 0)
C = relevant port and port width
Port chamber volume = A + woofer + B x C
D = back of port, now irrelevant?
E = port(s) on other side
F = distance from front to rear port (should influence response through phase cancellation?)
Woofer = SD 466cm2, Xmax +/-12.5mm but target <+/-10mm
David, please note that I’m fully aware that I’m not telling you anything new with respect to the variable’s impacts on the response calculated by hornresp. I’m just trying to be as clear as possible in showing my interpretation so that this is of value to anyone that is trying to come to grips with the variables of a ripole implementation. So if I’m making any kind of clear blunders here, please don’t hesitate to point these out, I’m here to learn 😬🥴
Btw, should I move this post to the main Hornresp tread (if that’s possible)
Thanks
Gis
Attachments
Last edited:
What would the accuracy be if I set:
S1 = woofer Sd (or circular baffle area A?)
S3 = 1/2 woofer Sd (or circular baffle area A?)
PAR 1 = Distance from centre of woofer to C - port mouth
PAR 2 = Distance from centre of woofer to D - port rear
Then alter variable B (which influences PAR 1 and F?) and working backwards from the Hornresp calculation of the chamber volumes and the minimum practical port mouth width, until I get all the variables balanced as much as possible?
Thanks
Gis
S1 = woofer Sd (or circular baffle area A?)
S3 = 1/2 woofer Sd (or circular baffle area A?)
PAR 1 = Distance from centre of woofer to C - port mouth
PAR 2 = Distance from centre of woofer to D - port rear
Then alter variable B (which influences PAR 1 and F?) and working backwards from the Hornresp calculation of the chamber volumes and the minimum practical port mouth width, until I get all the variables balanced as much as possible?
Thanks
Gis
Using your sketch and dimensions:
C = width of the single front-facing port
D = enclosure height = 2 * (Sd / Pi) ^ 0.5 + the gap between the driver and the top and bottom of the enclosure
E = total width of the two rear-facing ports
F = depth of enclosure
Then:
S1 = height * rear port width = D * E
S2 = S1
S2S = S1
S3 =S1
S3S = height * front port width = D * C
S4 =S3S
S4S = S3S
S5 = S3S
L12 = height / 2 = D / 2
L23 = F - L12
L34 = L12
L45 = L23
It would appear that maybe you are considering the wrong cross-sectional areas.
There is probably little to be gained by moving any of the posts, now that this thread is well established in its own right.
Hi @David McBean
My apologies for the ongoing questions, but I hope I'm almost there 🙂
I've put your suggested calculations together based on a design that should be very close to reality. Could you confirm this is now all correct?
The attached drawing shows how I have interpreted the dimensions in these calculations.
Your values didn't use the PAR values, so do I set up a generic horn file rather than a Ripole so that it shows the options of the L12, 23, 34 and 45 in the Hornresp interface or do I simply enter these in the PAR fields?
I've assumed (since its a radius) that the gap value is single measurement (not top + bottom?)
The distance measured for this value is the SD value of 466cm2 converted to the diameter of the driver (this = 244mm and = centre of the roll surround)
Thanks again!!
Gis
My apologies for the ongoing questions, but I hope I'm almost there 🙂
I've put your suggested calculations together based on a design that should be very close to reality. Could you confirm this is now all correct?
The attached drawing shows how I have interpreted the dimensions in these calculations.
Your values didn't use the PAR values, so do I set up a generic horn file rather than a Ripole so that it shows the options of the L12, 23, 34 and 45 in the Hornresp interface or do I simply enter these in the PAR fields?
I've assumed (since its a radius) that the gap value is single measurement (not top + bottom?)
The distance measured for this value is the SD value of 466cm2 converted to the diameter of the driver (this = 244mm and = centre of the roll surround)
Thanks again!!
Gis
| Answer | ||||
| C = width of the single front-facing port | 100.00 | mm | ||
| D = enclosure height = 2 * (Sd / Pi) ^ 0.5 + the gap between the driver and the top and bottom of the enclosure. | 295.00 | Gap = 52mm around when measured from a diameter of 244mm (Sd) | SD = 466cm2 or diameter = 244mm | 2 * (466 / Pi) ^ 0.5 + 5.2 |
| E = total width of the two rear-facing ports | 200.00 | mm | ||
| F = depth of the enclosure | 390.00 | mm | ||
| Then: | ||||
| S1 = height * rear port width = D * E | 590.00 | |||
| S2 = S1 | 590.00 | |||
| S2S = S1 | 590.00 | |||
| S3 =S1 | 590.00 | |||
| S3S = height * front port width = D * C | 295.00 | |||
| S4 =S3S | 295.00 | |||
| S4S = S3S | 295.00 | |||
| S5 = S3S | 295.00 | |||
| L12 = height / 2 = D / 2 | 14.75 | |||
| L23 = F - L12 | 24.25 | |||
| L34 = L12 | 14.75 | |||
| L45 = L23 | 24.25 |
