Zozo:
Congratualtions!! Even if you cannot get it down any farther, you have already gone a great deal toward suppressing the "spike".
It actually looks like a usable woofer curve the way it is. If we combine it with Andrew's advice and apread some stuffing around the middle, the "spike" will be reduced still further.
Of course, our aim is to attempt to see if we can suppress the "spike" compeletely through the Helmhotz resonator. However, even if we don't get any further than where we are, you have already shown a usable curve.
Now, one thing. Are you measuring just the woofer of your system, or the output of the whole system-woofer and port together?
Congratualtions!! Even if you cannot get it down any farther, you have already gone a great deal toward suppressing the "spike".
It actually looks like a usable woofer curve the way it is. If we combine it with Andrew's advice and apread some stuffing around the middle, the "spike" will be reduced still further.
Of course, our aim is to attempt to see if we can suppress the "spike" compeletely through the Helmhotz resonator. However, even if we don't get any further than where we are, you have already shown a usable curve.
Now, one thing. Are you measuring just the woofer of your system, or the output of the whole system-woofer and port together?
Zozo said:
Congratulations Zozo. I had a strong feeling that this concept was going to be a winner. My own adjustable test prototype won't be completed until next week so I'm studying your curves with great interest.
Re. your question ... try this link Helmholtz Resonator calculations. It will give you frequencies, volumes, port sizes & Qs. I believe that adding some filling to the chamber would lower the frequency and lower the Q, giving a wider frequency range.
Again congratulations Zozo and also kelticwizard who started this particular wagon rolling. Some day they'll thank us
Steve
I can't help but note that Zozo's first curve, which was unsatisfactory, the two ports were strongly interacting. The tuning frequency of the overall box was affected. That is because Fb of the box was 40 and the Helmholtz Resonator was around 70 Hz-less than an octave apart.
Here is Aoao's first curve:
http://www.diyaudio.com/forums/attachment.php?attachmentid=8059
In Zozo's second curve, he tunes to around 200, and the Fb of the box-40 Hz, is almost unaffected. The closer Fb is to the Helmholtz Frequency, the more they affect each other. As the two frequencies move apart, the less they affect each other.
This has been noted before in the "6th order bandpass box", where the two chambers are tuned to different frequencies. The two frequencies must be at least an octave apart, preferably more, or bad things happen to the response.
Here is an illustration of the 6th Order Bandpass Box. Despite the drawing, the two vents do pass through the wall of the enclosure and go all the way through to the outside of the box. The two chambers do not have to be equal size. As stated previously, tuning both chambers within an octave of each other usually yields uneven response:
Here is Aoao's first curve:
http://www.diyaudio.com/forums/attachment.php?attachmentid=8059
In Zozo's second curve, he tunes to around 200, and the Fb of the box-40 Hz, is almost unaffected. The closer Fb is to the Helmholtz Frequency, the more they affect each other. As the two frequencies move apart, the less they affect each other.
This has been noted before in the "6th order bandpass box", where the two chambers are tuned to different frequencies. The two frequencies must be at least an octave apart, preferably more, or bad things happen to the response.
Here is an illustration of the 6th Order Bandpass Box. Despite the drawing, the two vents do pass through the wall of the enclosure and go all the way through to the outside of the box. The two chambers do not have to be equal size. As stated previously, tuning both chambers within an octave of each other usually yields uneven response:
Attachments
With the possible interaction of the ports-decidedly lessened as the space between the tuning frequencies increases-we might wish to consider the idea that the Helmholtz Resonator might have a different tuning in a Vented Box from a Closed Box of equal volume. Just a guess. The farther apart the tuning of the Helmholtz Resonator is from Fb, the closer the tuning of the Vented Box will be to the Closed Box.
This is speculation on my part.
This is speculation on my part.
kelticwizard said:My guess is that doubling the volume raises the Q and therefore takes the "spike" down more. That is just a guess.
It's a little more complicated than that.
Let's take a volume of 10L and a tuning frequency of around 150Hz. We can tune the Q by changing the port length and adjusting the port diameter to match (and vice versa).
So ...
Volume: 10L - Frequency: 150Hz
Port length: 1.0cm - Port diameter: 8.7cm - Q: 0.04
Port length: 2.5cm - Port diameter: 10.0cm - Q: 0.11
Port length: 5.0cm - Port diameter: 11.7cm - Q: 0.20
Increasing the volume whilst keeping the frequency constant gives:
Volume: 20L - Frequency: 150Hz
Port length: 1.0cm - Port diameter: 16.4cm - Q: 0.01
Port length: 2.5cm - Port diameter: 17.9cm - Q: 0.03
Port length: 5.0cm - Port diameter: 20.0cm - Q: 0.06
Port length: 20.0cm-Port Diameter: 28.5cm - Q: 0.16
So, it seems that Q is not only a function of volume.
My guess is that increasing volume while holding the Q and frequency constant will increase the absorbency of the resonator.
Is that a possibility?
Steve
I am not sure about increasing the absorbancy of the Helmholtz Resonator.
But it looks like my guess is the exact opposite of the facts. Increasing the volume decreases the Q of the Helmholtz Resonator. But making the port longer and wider increases it.
Ahh, where would we be without conterbalancing variables to make our lives complicated?
One thing to remember in Helmholtz Resonators: The volume of the port is subtracted from the volume of the Resonator itself. In small Helmholtz Resonators with large ports, this can affect tuning.
But it looks like my guess is the exact opposite of the facts. Increasing the volume decreases the Q of the Helmholtz Resonator. But making the port longer and wider increases it.
Ahh, where would we be without conterbalancing variables to make our lives complicated?
One thing to remember in Helmholtz Resonators: The volume of the port is subtracted from the volume of the Resonator itself. In small Helmholtz Resonators with large ports, this can affect tuning.
Steve:
Don't know. All my experience with Helmholtz Resonators are with Vented Boxes with a speaker on the front and the port either front or back. The Vas of the speaker in combination with the desired F3 of the speaker system determined the volume of the Helmholtz Resonator, and that was that.
How a Helmholtz Resonator would act with no speaker attached to it and the sound coming from a source not physically connected to it I do not know, other than to predict which frequency it is tuned to.
I might get a chance to check that LspCAD program and report what I see either late tonight or tomorrow.
Don't know. All my experience with Helmholtz Resonators are with Vented Boxes with a speaker on the front and the port either front or back. The Vas of the speaker in combination with the desired F3 of the speaker system determined the volume of the Helmholtz Resonator, and that was that.
How a Helmholtz Resonator would act with no speaker attached to it and the sound coming from a source not physically connected to it I do not know, other than to predict which frequency it is tuned to.
I might get a chance to check that LspCAD program and report what I see either late tonight or tomorrow.
Kelticwizard & 7V,
Thanks for your positive comments and the help giving me. Here are my latest meas with some added stuffing. I think in this setup this is the best I can achive. As you can see the response is within +/- 1dB which is not bad at all.
I have added some more damping to the chamber, but it didn't helped to further reduce the ripple :
Thanks for your positive comments and the help giving me. Here are my latest meas with some added stuffing. I think in this setup this is the best I can achive. As you can see the response is within +/- 1dB which is not bad at all.
I have added some more damping to the chamber, but it didn't helped to further reduce the ripple :
Attachments
I have a D'Appolito configuration tall narrow enclosre project going on at the moment also. At current state I have a ready cabinet and one also has drivers in it. I have also made some preliminary measurements which also show a much bigger resonanceproblem than I had anticipated.....
My box is 20l and 125cm high inside. Other dimensions are about 13x12cm. So hard one to get working but I will try and later today I will be doing more measurements and posting them here also.
***
One answer I know already, that was current couple of pages back. Namely the peak hight in similar kind of enclosure... The question was how can the peak be so much smaller in zozo's box than in pioneers. I'm quite sure it's a matter of where you put the midbass element hightwise. In my D'Appolito gonfig the driver closer to the top shows a much higher peak than the lower one. I quess it's the matter of how much the driver exites the standing wave and in this respect the upper one in worse.
I hope I'll get more finding soon.
Ergo
My box is 20l and 125cm high inside. Other dimensions are about 13x12cm. So hard one to get working but I will try and later today I will be doing more measurements and posting them here also.
***
One answer I know already, that was current couple of pages back. Namely the peak hight in similar kind of enclosure... The question was how can the peak be so much smaller in zozo's box than in pioneers. I'm quite sure it's a matter of where you put the midbass element hightwise. In my D'Appolito gonfig the driver closer to the top shows a much higher peak than the lower one. I quess it's the matter of how much the driver exites the standing wave and in this respect the upper one in worse.
I hope I'll get more finding soon.
Ergo
Well, at the moment I have cabinets already made. As I did not expect so big problems with standing waves I did both, top and bottom just parallel regular surfaces. I have an option to open the bottom if really needed but I plan to make measurements and try different stuffing ideas etc. and try to avoid total rebuilding of the enclosure.....
current state
Here you can find more pix..... The other one is a parallel project of my "best know how - high end - no compromise" type of speaker. The tower are "low cost - good quality" type of project.
http://f5.infonet.ee/ergo/newspeaker/
Ergo
current state
An externally hosted image should be here but it was not working when we last tested it.
Here you can find more pix..... The other one is a parallel project of my "best know how - high end - no compromise" type of speaker. The tower are "low cost - good quality" type of project.
http://f5.infonet.ee/ergo/newspeaker/
Ergo
7V said:
I am using a cheap Voltcraft (Radioshack) SPL meter, built-in soundcard and SATLive measurement SW. Not professional, but reliable enough for experimenting.
Ergo,
It would be interesting to see your measurements on this tall enclousure. I think the peak height is also depends on the height, width, depth ratio. The taller and narrower the enclosure the bigger the peak, I guess.
Another affecting fact is the D'Appolito configuration (different woofer positions).
Zozo
ergo said:
If this standing wave is a real problem and you want to try the Helmholtz Resonator (HR) approach, there is a way that would avoid opening up the cabinet.
Make a small box as an HR, tuned to the standing wave frequency, and mount it on the back of the speaker at the bottom. Then drill a hole at the bottom rear of the cabinet, opening the speaker into the HR. In the (hopefully unlikely) event that it all goes horribly wrong, you could just throw away the HR box and block over the hole.
Steve
ergo, I've just taken a look at your pics.
The method of construction where you use the oval segments with a flatted front is exactly what I'm doing on my latest speaker, with a couple of differences. Have you finished it yet and if so, how do you find this design for cabinet vibration?
I'm waiting to get my "segments" delivered from the CNC machining shop but I reckon that it should be pretty good, partly because the design eliminates the vibration modes that are present in flat panels and partly because the wall has a varying thickness.
On the subject of this thread - this type of construction makes it easy to put in an HR. You just design one of the segments near the base to be solid with a hole of the right size. That's why I'm so keen to explore the "HR compartment" concept.
Let us know how it goes.
Steve
The method of construction where you use the oval segments with a flatted front is exactly what I'm doing on my latest speaker, with a couple of differences. Have you finished it yet and if so, how do you find this design for cabinet vibration?
I'm waiting to get my "segments" delivered from the CNC machining shop but I reckon that it should be pretty good, partly because the design eliminates the vibration modes that are present in flat panels and partly because the wall has a varying thickness.
On the subject of this thread - this type of construction makes it easy to put in an HR. You just design one of the segments near the base to be solid with a hole of the right size. That's why I'm so keen to explore the "HR compartment" concept.
Let us know how it goes.
Steve
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.