Overall it looks pretty good. You are making good progress.
It looks to me like the port frequency is tuned a little too low. The woofer's Thiele-Small parameters may differ a bit from the specification. It may loosen up a little with breakin, but probably not much. Using a longer port may help a bit, maybe one tuned for 50 Hz or so.
I don't know whether the peak around 115 Hz is a measurement artifact or something else. I don't see a blip in the impedance response there, so I don't think it is a resonance. Nonetheless, a lot of people like a little extra boost at that frequency.
Looking at the phase and frequency response charts, the drivers are in phase at the crossover frequency. Good job. But, the tweeter's phase starts to diverge quickly just below that. In addition, the SPL is elevated a little around the crossover region. Perhaps you can make adjustments to address those issues?
Also, if you have not done so, make sure that you input the x, y, z cooridinates for your drivers in VituixCAD.
It looks to me like the port frequency is tuned a little too low. The woofer's Thiele-Small parameters may differ a bit from the specification. It may loosen up a little with breakin, but probably not much. Using a longer port may help a bit, maybe one tuned for 50 Hz or so.
I don't know whether the peak around 115 Hz is a measurement artifact or something else. I don't see a blip in the impedance response there, so I don't think it is a resonance. Nonetheless, a lot of people like a little extra boost at that frequency.
Looking at the phase and frequency response charts, the drivers are in phase at the crossover frequency. Good job. But, the tweeter's phase starts to diverge quickly just below that. In addition, the SPL is elevated a little around the crossover region. Perhaps you can make adjustments to address those issues?
Also, if you have not done so, make sure that you input the x, y, z cooridinates for your drivers in VituixCAD.
@TerryForsythe Thank you for checking my work!
Port tuning: I have read lab measurements for the T/S parameters that vary from the factory a little. I’ll make sure I try the lab measurements. What are you looking for to tell the performance of the port? I expect to extend bass a little if all is well but you saw something that worried you. The port length is 5.375” (41Hz). I tried 3.3” (49Hz) and it didn’t seem to be adding much to the boxes low end.
Crossover changes don’t affect the blip at 115Hz (it shows up in REW). I haven’t tried complicated crossovers, though. I do have to go and fix the ESR and DCR values for caps and inductors. I doubt things will change much. Perhaps it’s a baffle issue affecting frequencies just above the blip, accentuating it.
The rate of the tweeter phase changing below the crossover point is noted. I’m not sure I can soften that without breaking something else. I may try moving the crossover point from 2.3KHz nearer to 3KHz. Maybe that’s enough?
The tweeter is free standing with its own 4” baffle. That baffle is in the same plane as the woofer’s so they share the same Z. I have adjusted the Y of the woofer. The twtr and wfr share the same X. I have played with moving Z of the tweeter farther from the listener, but things didn’t improve.
Thanks again
Port tuning: I have read lab measurements for the T/S parameters that vary from the factory a little. I’ll make sure I try the lab measurements. What are you looking for to tell the performance of the port? I expect to extend bass a little if all is well but you saw something that worried you. The port length is 5.375” (41Hz). I tried 3.3” (49Hz) and it didn’t seem to be adding much to the boxes low end.
Crossover changes don’t affect the blip at 115Hz (it shows up in REW). I haven’t tried complicated crossovers, though. I do have to go and fix the ESR and DCR values for caps and inductors. I doubt things will change much. Perhaps it’s a baffle issue affecting frequencies just above the blip, accentuating it.
The rate of the tweeter phase changing below the crossover point is noted. I’m not sure I can soften that without breaking something else. I may try moving the crossover point from 2.3KHz nearer to 3KHz. Maybe that’s enough?
The tweeter is free standing with its own 4” baffle. That baffle is in the same plane as the woofer’s so they share the same Z. I have adjusted the Y of the woofer. The twtr and wfr share the same X. I have played with moving Z of the tweeter farther from the listener, but things didn’t improve.
Thanks again
Last edited:
Have a look at the impedance level at the valley between the peaks. It should usually (unless highly dampened tuning, e.g. aperiodic) be similar to the impedance above the upper peak.
In your case (around 5 ohm) it could be slightly lower, but I guess it's perfectly ok as it is. Eventually check for small air leaks: play a sine wave at tuning frequency, raise the level and listen for high pitched noise.
Edit: the driver may also still need some breaking in.
Looks quite good, I would attenuate a couple of dB the area from 500Hz to 5kHz.
Sorry for the belated answer but I di not see your mention. I am just following the indications given by Kimmosto (I hope correctly): it's not about room interaction but the diffraction responses at 10m/1m distance (with no full space/directivity checked) along with the the full space help to simulate correctly the levels between LF anf HF parts.
See these merging examples:
Shadowplay62 many thanks for filling in a big gap in my knowledge re getting the basic LF part set correctly.
Usually I adjust the LF using measurements and listening to come to a happy medium, but having a representative model will help me with future builds.
Usually I adjust the LF using measurements and listening to come to a happy medium, but having a representative model will help me with future builds.
Last edited:
Oops. I meant shortening the port. 🙂 Anyway, at 41Hz the you are down 10dB, so it already is in the range where the bass output is falling fast. With a little bit higher tuning frequency, it should extend the bass tuning a little bit lower before the drop. What is your box air volume for the woofer? Assuming the published Thiele-Small parameters are correct, I get the following QL=7 alignments:
SBB4:
Box volume: 15L
Tuning frequency: 49Hz
SQB3
Box volume: 17.1L
Tuning frequency: 45Hz
SC4
Box volume: 18.8L
Tuning frequency: 43Hz
I wouldn't do that if you have a typical room without a lot of room treatment. It will make the directivity index a little worse. That will degrade the tonal balance if you have typical room reflections. If you have sound treatment on the side walls and a carpeted floor, it may not matter too much, though.
I'm not sure what is causing the phase deviation of the tweeter below 2.1kHz. You could try removing the resistors on the output side of the tweeter's filter, and adding a series resistor to the input side. Of course, that would require a change to the capacitor and inductor values for the tweeter.
If that does not improve the phase, you could experiment with filling in the gap between the tweeter enclosure and the sides of the cabinet and see if that has any effect on the phase. Maybe cut a couple of small pieces of wood and put them there and see how that impacts the phase. You also could see if adding sound absorber there has any impact.
If none of the above result in an improvement, I would stick with what you have. As it is, the phase alignment is quite a bit better than many commercially available speakers I have seen.
I would position the tweeter in the manner that has the least amount of diffraction, which probably means keeping the tweeter baffle aligned with the woofer baffle. Try to address the phase as discussed above.
The box is 20 liters. About 1/3 of that volume is loose packed fiberglass insulation - this is about twice what it had before. There was a bump in the vent pipe due to a pipe coupling. I worked on smoothing that out while making it flare (not a lot). I think it will help with chuffing and it seems like it might effectively shorten the pipe, but not by much.
I'll play around with the crossover per your suggestions trying to keep the crossover frequency around 2.5KHz and try and improve the tweeter's phase at the low end. I have some sound absorber material that I can try around the tweeter.
I think I'm confused. The T/S parameters are accepted in the Enclosure interface of Vituixcad2 but I cannot see what mechanism causes any of that information to be available to crossover design window. Why?
@shadowplay62
I took your advice about lowering the sound level in the 500Hz to 5KHz range. That's for the reference to the merger examples. I don't understand them yet and will have to learn some more.
I'll try that. Thanks.
I'm away from home for a few days, so I cannot physically change the speakers until home. Thanks to all for your help and patience!
That is a good size for the woofer based on the published Thiele-Small parameters. Maybe with some break in your f3 frequency will get a little lower.
I think that when you enter measured impedances and frequency responses for the drivers, VituixCAD uses those instead of the Thiele-Small parameters. Assuming the measurements are accurate, the model will be more accurate using the measured data.
One more note:
It is important to keep the area around the port opening clear. If there is insulation near the port, it will interfere with the port's airflow, throwing off the port tuning frequency.
Also, I try to densly pack the insulation at the top and bottom of the cabinet, then loose fill in the rest of the cabinet, except keeping the area around the port clear.
Not a problem in my box. I've kept the pipe's internal end clear.
The way I've mounted my loose fiberglass insulation is to glue it to a 3mm aluminum/bitumen sheet that, itself, is adhered to the speaker wall. I insulate one wall of each opposing pair - the theory being it should be enough to discourage standing waves and mitigate box wall vibrations from acting in concert between wall pairs. I expect there to be less or no effect on the lower LF range. When I rap on the box, it is more punky now.