The plan is to do further measurements
1. No damping material at all! As a horror reference!
2. Damping in chamber and IHR and nothing in P2 and P3, regarding P1 I am on the fence. This would represent the classical damping scheme
3. As 2 but with the IHR filled with something solid like a plastic bag filled with sand or rice. This would represent the classical damping scheme wih no IHR
If I focus on the impedance peak at 35 Hz and dip at 45 Hz;
Adding damping in the chamber does nothing
+IHR reduce the peak a 2 Ohm
+P1 reduce the peak 3 ohm
+P2 reduce the peak from a 7 ohm peak to a 1 ohm hump and also reduce the dip at 45 Hz
+P3 and the dip is gone and the peak is just a shoulder of less slope 30-40 Hz.
If I do water fall plots in the bass it will tell me what I know, the room is bad. I wished I had access to a really large room with a good corner and measure the total output at 3 m or so
1. No damping material at all! As a horror reference!
2. Damping in chamber and IHR and nothing in P2 and P3, regarding P1 I am on the fence. This would represent the classical damping scheme
3. As 2 but with the IHR filled with something solid like a plastic bag filled with sand or rice. This would represent the classical damping scheme wih no IHR
If I focus on the impedance peak at 35 Hz and dip at 45 Hz;
Adding damping in the chamber does nothing
+IHR reduce the peak a 2 Ohm
+P1 reduce the peak 3 ohm
+P2 reduce the peak from a 7 ohm peak to a 1 ohm hump and also reduce the dip at 45 Hz
+P3 and the dip is gone and the peak is just a shoulder of less slope 30-40 Hz.
If I do water fall plots in the bass it will tell me what I know, the room is bad. I wished I had access to a really large room with a good corner and measure the total output at 3 m or so
In Kings simulation if I read it correctly the pipe output has - 6dB points at 40 and 200 Hz, I get more like 65-150 Hz and about 12 dB down at 40 and 200 Hz.
Time to investigate what that IHR is doing and the damping material in the pipe
Well, an impressive set of measurements that's for sure. Wish I had the patience to do that to my speaker builds.
When you refer to the -6db points, you need to bear in mind that these are usually calculated relative to the driver's nominal SPL, and not to the horn's maximum level.
As you can see in one of your measurements (horn in corner, mic at 1m), if we consider the driver's nominal SPL to be around the 67.5db mark on the graph (as shown in the 500hz-2khz region where the horn contribution is nill) the horn is a relatively high gain design raising the SPL around 5db above nominal in its bandwidth (40-200hz).
Now take the levels at 65 and 150hz on the graph, go 5db lower, and look at what frequencies get intersected.
Pretty close to 40 and 200hz
edit: try smoothing your graphs, say 1/12 or 1/6 octave smoothing
Last edited:
I have made som more measurements of the pipe output, now I had good use of the small tripod that was included!
Omnimic does some kind of automatic level adjustments in graphs so I am not so sure about output diffrences. My focus is the balance between shelf at 40-60 Hz and the peaks at 95 and 190 Hz.
The top black trace is with damping material in only the chamber and IHR
Blue with P1
Green with P1 and P2 added
Yellow with P1, P2 and P3 added
My interpretation was that the pipe stuffing do reduce the 190 Hz peak a lot but at the expense of bass shelf output.
Here I have no damping in P2 and P3 and I have 42 g in the chamber. I have some of the shelf back and the 190 Hz peak is decent.
I still wanted to get a better shelf and get rid of that pesky 190 Hz. I removed the damping material in P1 and just to try it out I filled the IHR with a bag of rice.
The shelf is good down to 40 Hz, the 190 Hz peak is reduced by 7 db! I still have that 95 Hz 8 dB hump. Perhaps if I retune the IHR to 95 Hz and have some resisitive damping to keep the Q low I could reduce the output 6 db between 70 and 120 Hz?
An externally hosted image should be here but it was not working when we last tested it.
Omnimic does some kind of automatic level adjustments in graphs so I am not so sure about output diffrences. My focus is the balance between shelf at 40-60 Hz and the peaks at 95 and 190 Hz.
The top black trace is with damping material in only the chamber and IHR
Blue with P1
Green with P1 and P2 added
Yellow with P1, P2 and P3 added
My interpretation was that the pipe stuffing do reduce the 190 Hz peak a lot but at the expense of bass shelf output.
An externally hosted image should be here but it was not working when we last tested it.
Here I have no damping in P2 and P3 and I have 42 g in the chamber. I have some of the shelf back and the 190 Hz peak is decent.
I still wanted to get a better shelf and get rid of that pesky 190 Hz. I removed the damping material in P1 and just to try it out I filled the IHR with a bag of rice.
An externally hosted image should be here but it was not working when we last tested it.
The shelf is good down to 40 Hz, the 190 Hz peak is reduced by 7 db! I still have that 95 Hz 8 dB hump. Perhaps if I retune the IHR to 95 Hz and have some resisitive damping to keep the Q low I could reduce the output 6 db between 70 and 120 Hz?
The rice was used to fill the IHR to accoustically close it. Reducing the volume of the driver chamber is not something I have tried.
I have used a online calculator mh-audio.nl - Acoustic
The original IHR has very wide band with 105-405 Hz, if I try to lower the resonance from 255 Hz to around 95 Hz by using a smaller and longer port the band with shrink to much less one 25mm hole in 33 mm thick board result in 89 Hz and 0.43 Hz bandwith two holes 25mm diameter gives 116 Hz and 1.5 Hz...
Regarding T/S paremeters look at page 2 post 18, I am using the driver with SN 404
I have used a online calculator mh-audio.nl - Acoustic
The original IHR has very wide band with 105-405 Hz, if I try to lower the resonance from 255 Hz to around 95 Hz by using a smaller and longer port the band with shrink to much less one 25mm hole in 33 mm thick board result in 89 Hz and 0.43 Hz bandwith two holes 25mm diameter gives 116 Hz and 1.5 Hz...
Regarding T/S paremeters look at page 2 post 18, I am using the driver with SN 404
Now I have blocked the IHR and drilled two 25 mm diameter holes in two layers of 11 mm OSB so the tuning frequencys are 100 Hz and 128 Hz with one and two holes open respectively. Here I have the speaker chamber even more densly stuffed 63 grams!
The black curve is closed vents and it is quite similar to the rice bag in previous measurements with less damping, the main difference is less of a peak at 190 Hz.
The brown and red curve are two measurements with one hole open (tuning 100 Hz) taken before and after the two hole measurement. It does nothing for the 90 Hz peak but lowers the 100-160 Hz range by 5 dB or so.
The blue is with both vents open and a calculated tuning of 128 Hz and it takes out 5-10 dB from 130 to 210 Hz. I think it looks like the best combination so far, but I would rather have reduced that 90 Hz hump than the one around 160 Hz.
I will search my house for tubes of 25mm OD with such an item I could tune all the way down to 60 Hz.
The black curve is closed vents and it is quite similar to the rice bag in previous measurements with less damping, the main difference is less of a peak at 190 Hz.
The brown and red curve are two measurements with one hole open (tuning 100 Hz) taken before and after the two hole measurement. It does nothing for the 90 Hz peak but lowers the 100-160 Hz range by 5 dB or so.
The blue is with both vents open and a calculated tuning of 128 Hz and it takes out 5-10 dB from 130 to 210 Hz. I think it looks like the best combination so far, but I would rather have reduced that 90 Hz hump than the one around 160 Hz.
I will search my house for tubes of 25mm OD with such an item I could tune all the way down to 60 Hz.
And only at 4 degrees celsius does it get very near to 1g/cc (maximum water density at 1 bar of pressure).
Well at 15C the density has decreased to 0.999 and at 45C it is still 0.99 so for a kitchen scale that is good enogh. I would need a calibrated Voll pipette to get the error of added volume down into the 0.1% range, I make food in my kitchen not analytical chemistry.
I found a tube containing vitamin C that was close to 25mm I am gluing it to one of the vents I hope the latex sealant have cured by tomorrow.
If I would make a new prototype I would make "manholes" or rather "handholes" to access the compartments and have most of the speaker closed. There is a lot of vibrations going on when the whole side is structurally disconnected to the rest of the enclosure.
I found a tube containing vitamin C that was close to 25mm I am gluing it to one of the vents I hope the latex sealant have cured by tomorrow.
If I would make a new prototype I would make "manholes" or rather "handholes" to access the compartments and have most of the speaker closed. There is a lot of vibrations going on when the whole side is structurally disconnected to the rest of the enclosure.
Drilled holes for the IHR
Extended with a tube containing vitamin C previously.
Horn mouth measurements
The one long vent with the lowest tuning just lowers the 40-60 Hz, so what ever I throw at that 90 Hz peak it is really stubborn.
Corresponding impedance measurements
I think I will go with the two short vents, fiddle with the damping of the two chambers and then glue on the side and take it to some serious listening and some final set of measurements.
Extended with a tube containing vitamin C previously.
Horn mouth measurements
The one long vent with the lowest tuning just lowers the 40-60 Hz, so what ever I throw at that 90 Hz peak it is really stubborn.
Corresponding impedance measurements
I think I will go with the two short vents, fiddle with the damping of the two chambers and then glue on the side and take it to some serious listening and some final set of measurements.
Dr Boar,
The 90 Hz peak corresponds to a quarter wave cavity length of about 1.05 meters, looking at the MJK plans for this horn, there appears to be wall forming the bottom of the Helmholz resonator and the wall forming the top wall of final passage before it turns 90 deg and exits at the mouth, that has an approximate length matching this distance. Try putting some highly acoustic absorbent material like thick felt on these walls to destroy the resonator cavity. It might be worth noting that avoiding parallel faced walls in the internal passages may reduce these peaks in general. Just 2 or 3 degrees offset from perfect 90 deg normal will do wonders. I run across this all the time in the optical design world where we trace back inadvertent Fabry-Perot etalon peaks by backing out the length and look for corresponding offending parallel walls or surfaces.
The 90 Hz peak corresponds to a quarter wave cavity length of about 1.05 meters, looking at the MJK plans for this horn, there appears to be wall forming the bottom of the Helmholz resonator and the wall forming the top wall of final passage before it turns 90 deg and exits at the mouth, that has an approximate length matching this distance. Try putting some highly acoustic absorbent material like thick felt on these walls to destroy the resonator cavity. It might be worth noting that avoiding parallel faced walls in the internal passages may reduce these peaks in general. Just 2 or 3 degrees offset from perfect 90 deg normal will do wonders. I run across this all the time in the optical design world where we trace back inadvertent Fabry-Perot etalon peaks by backing out the length and look for corresponding offending parallel walls or surfaces.
Velocity of sound is about 344 m/s a quarter of that is 86 and 90 Hz then would be 0.95 m. Standing waves are at 1/2 length if it is between two surfaces. The bounce at an opening due to the impedance change is at 1/4 wave.
But I think you are right about that it is the pipe that is causing that peak. I will try putting rolls of damping material in the middle of each section (velocity maxima) to see what I can find. I also have some scrap pieces that I can use to make corner reflectors, so I will do that as well. This I can do today but actual measurements have to wait until the wife is out of the house, those test signals fray the nerves...
But I think you are right about that it is the pipe that is causing that peak. I will try putting rolls of damping material in the middle of each section (velocity maxima) to see what I can find. I also have some scrap pieces that I can use to make corner reflectors, so I will do that as well. This I can do today but actual measurements have to wait until the wife is out of the house, those test signals fray the nerves...
Here you have the corner reflectors 50x50x175mm apart from the corner thing in the IHR that is only about 80 mm long. Then there are some bitumen loaded fibre board added as well
Measurment setup, minus my body gracefully draped on top of the box...
The topmost jagged curve is the empty pipe, then follows with damping in chamber, chamber &IHR and then complete damping the last one remove the dip at 40 Hz and also the one at 150 Hz.
Driver response 5 cm on axis.
Red undamped, big dips at 40 85 and 150 Hz
Brown with chamber and IHR damped 40Hz dip lower in frequency and less deep, 85 Hz dip unaffected 150 Hz less deep.
Black full damping, 40 and 150 Hz much reduced, 90 Hz slightly lower in frequency and slightly less deep
Horn opening:
Black undamped humps at 40 90 and 150 Hz, corresponding to the dips in driver response
Red damping in chamber and IHR, as the one above but 150 Hz reduced 6 dB or so.
Brown full damping. 40 Hz reduced by 6 dB, the 90 Hz lowered some few Hz and dB and the 150 Hz is gone by 10 dB also this in good agreement with driver response and impedance plots.
The foam seal was hard to get rid of very sticky but now the side is glued on with 23 g BAF in the IHR and 25 g in speaker chamber. I used PU glue as it expands and fills any minor cracks.
Next up is some listening tests.
Measurment setup, minus my body gracefully draped on top of the box...
The topmost jagged curve is the empty pipe, then follows with damping in chamber, chamber &IHR and then complete damping the last one remove the dip at 40 Hz and also the one at 150 Hz.
Driver response 5 cm on axis.
Red undamped, big dips at 40 85 and 150 Hz
Brown with chamber and IHR damped 40Hz dip lower in frequency and less deep, 85 Hz dip unaffected 150 Hz less deep.
Black full damping, 40 and 150 Hz much reduced, 90 Hz slightly lower in frequency and slightly less deep
Horn opening:
Black undamped humps at 40 90 and 150 Hz, corresponding to the dips in driver response
Red damping in chamber and IHR, as the one above but 150 Hz reduced 6 dB or so.
Brown full damping. 40 Hz reduced by 6 dB, the 90 Hz lowered some few Hz and dB and the 150 Hz is gone by 10 dB also this in good agreement with driver response and impedance plots.
The foam seal was hard to get rid of very sticky but now the side is glued on with 23 g BAF in the IHR and 25 g in speaker chamber. I used PU glue as it expands and fills any minor cracks.
Next up is some listening tests.
Yes I have left out the damping of the pipe. With a 50 cm2 cone area I try to keep all the radiation resistance I can get. There must be a way to reallocate colors in Omnimic so I get consistent coloring schemes across different graphs as well as better contrast in than red and brown.
I will try out different levels of damping in the speaker chamber, and perhaps try sticking a piece of felt in the first section of the horn.
I also have discovered that the side walls of the horn mouth vibrates, tapping it gives a sustained high pitch tone putting a crossbrace of only 5mm diameter hardwood reduces it to a dead "thunk". I could record it and put in on youtube, but then I have to get an account on youtube and so on so that may or may not happen.
I will try out different levels of damping in the speaker chamber, and perhaps try sticking a piece of felt in the first section of the horn.
I also have discovered that the side walls of the horn mouth vibrates, tapping it gives a sustained high pitch tone putting a crossbrace of only 5mm diameter hardwood reduces it to a dead "thunk". I could record it and put in on youtube, but then I have to get an account on youtube and so on so that may or may not happen.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.