A partition halfway up seems like a very promising approach.
Alex
And by that I meant a damping layer.
David S.
And by that I meant a damping layer.
David S.
Yes - of course! I posted last thing at night, and while I had a picture of a bag of felt and wool in my head, what I ended up typing came out quite differently...
Alex
Your port location half way up the box directed out the front baffle isn't helping your cause. Most of what you're measuring is likely standing waves interacting with the port.
The same I was going to say. That's due to the fact that biggest amplitude of standing waves is in the middle.
Your port location half way up the box directed out the front baffle isn't helping your cause. Most of what you're measuring is likely standing waves interacting with the port.
I'm afraid I can't move the port
My strategy now is to try to damp the fundamental mode by adding damping material halfway up the box. Luckily the port isn't halfway up the baffle (more like a third of the way up), so it won't get in the way of the damping (and vice versa).
Alex
Alex,
Your port have a good diameter ? it seems to me a little too small for the 8".
Nice Loudspeaker
Thanks, Jerome - once I have sorted out the crossover and enclosure resonances it should certainly have the potential to be a nice speaker!
As for the port, I'm afraid that again it is too late to change the port without some serious surgery...
Alex
I have finally got round to doing some proper wide frequency range impedance measurements on the box system.
In the above plot, the black solid curve is Audax data for the driver on a baffle. The black dashed line is my original measurement of the driver in my box, showing the sharp resonance at 230 Hz - note that the lower frquency limit for this is only 100 Hz..
The green curve is the same as the latter, but with a couple of extra horizontal shelves, which were my attempt to break up the top-bottom cavity resonance: you can see that they have shifted the frequency downward by a few hertz, as I remarked earlier, but I suspect that the apparent reduction in height of the peak is a sampling artefact. Finally, the red curve is the same enclosure but with 150g of long-haired wool packed into a pillow case fixed halfway between the driver and the port. The peak is now unambiguously reduced and breadened, as you might expect, and its frequency is reduced to 170 Hz or so.
This is the first time I have made any measurements at all of the low-frequency behaviour of the speaker, but I can make certain inferences, which may or may not be valid. The impedance minimum (corresponding to the box tuning frequency, as I understand it) is clearly at around 33 Hz, which is pleasingly close to my design target. I gather that a ported system should ideally have similar amplitudes for the two peaks, but my signal generator doesn't go low enough to catch the lower peak. The damping has hardly affected the upper peak, but seems to have reduced the size of the lower (port) resonance. I guess I might be able to at least partially remedy this by making sure the damping material is kept away from the port.
I think I am nearly there. My original problem was that the impedance peak at 230 Hz casued a big increase in the load presented to my crossover, and hence an unwanted spike in the filter response at that frequency. Now that I have reduced the peak by a factor of two or three, that should improve, and of course I have at the same time ameliorated an unwanted cabinet resonance, which is probably audible even with active drive.
Any comments on my impedance curves from the more experienced designers?
Alex
An externally hosted image should be here but it was not working when we last tested it.
In the above plot, the black solid curve is Audax data for the driver on a baffle. The black dashed line is my original measurement of the driver in my box, showing the sharp resonance at 230 Hz - note that the lower frquency limit for this is only 100 Hz..
The green curve is the same as the latter, but with a couple of extra horizontal shelves, which were my attempt to break up the top-bottom cavity resonance: you can see that they have shifted the frequency downward by a few hertz, as I remarked earlier, but I suspect that the apparent reduction in height of the peak is a sampling artefact. Finally, the red curve is the same enclosure but with 150g of long-haired wool packed into a pillow case fixed halfway between the driver and the port. The peak is now unambiguously reduced and breadened, as you might expect, and its frequency is reduced to 170 Hz or so.
This is the first time I have made any measurements at all of the low-frequency behaviour of the speaker, but I can make certain inferences, which may or may not be valid. The impedance minimum (corresponding to the box tuning frequency, as I understand it) is clearly at around 33 Hz, which is pleasingly close to my design target. I gather that a ported system should ideally have similar amplitudes for the two peaks, but my signal generator doesn't go low enough to catch the lower peak. The damping has hardly affected the upper peak, but seems to have reduced the size of the lower (port) resonance. I guess I might be able to at least partially remedy this by making sure the damping material is kept away from the port.
I think I am nearly there. My original problem was that the impedance peak at 230 Hz casued a big increase in the load presented to my crossover, and hence an unwanted spike in the filter response at that frequency. Now that I have reduced the peak by a factor of two or three, that should improve, and of course I have at the same time ameliorated an unwanted cabinet resonance, which is probably audible even with active drive.
Any comments on my impedance curves from the more experienced designers?
Alex
Last edited:
Alex,
Yes you have all the elements to make a very high end loudspeaker. I like Audax drivers. I browsed your website and the diameter of the port is good, 6.4cm
Perhaps if you put damping wool in the box, the resonance frequency could go lower, perhaps it could be sonically good but perhaps you will have to cut the tube shorter.
I saw the box, you can place the damping in the bottom. I would have filled it till the first brace ? And put a little damping behind the driver, at the parallel face. It is a little tricky because you should not put too much.
Regards.
Yes you have all the elements to make a very high end loudspeaker. I like Audax drivers. I browsed your website and the diameter of the port is good, 6.4cm
Perhaps if you put damping wool in the box, the resonance frequency could go lower, perhaps it could be sonically good but perhaps you will have to cut the tube shorter.
I saw the box, you can place the damping in the bottom. I would have filled it till the first brace ? And put a little damping behind the driver, at the parallel face. It is a little tricky because you should not put too much.
Regards.
More experienced ?Any comments on my impedance curves from the more experienced designers?
It is important to place the damping in the right place.
You should put more damping. I like this picture from Troels :
An externally hosted image should be here but it was not working when we last tested it.
Hi Jerome,
I had noticed the feature at 430 Hz in the datasheet, but I wouldn't describe it as a "peak" like the one I see in my measurements. My feeling is that what I see is a cavity resonance between front and back panels, as the frequency is about right.
I can't believe Troels Gravesen can have a day job! He has finished so many speaker projects. Do you have a url for the particular figure you posted?
Alex
I had noticed the feature at 430 Hz in the datasheet, but I wouldn't describe it as a "peak" like the one I see in my measurements. My feeling is that what I see is a cavity resonance between front and back panels, as the frequency is about right.
I can't believe Troels Gravesen can have a day job! He has finished so many speaker projects. Do you have a url for the particular figure you posted?
Alex
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Loudspeakers
- Multi-Way
- Unexpected resonance