Hi Nuuk and thanks for the welcome!
I've been not writting in this forum for too much time, I've been spending a lot of time studying electronics and acustics. It's good to see you again, and checkin that your doing fine without box
Until now, I've been checking bass range, and if I conect a reflex speaker just to check,
, it sounds like a ...... box.
I still have to check the chip forum, but I built a discrete amp that has better highs than my previous GC, and now I'm planning the bass amps, those XLS are current eaters
I've been not writting in this forum for too much time, I've been spending a lot of time studying electronics and acustics. It's good to see you again, and checkin that your doing fine without box
Until now, I've been checking bass range, and if I conect a reflex speaker just to check,
, it sounds like a ...... box.I still have to check the chip forum, but I built a discrete amp that has better highs than my previous GC, and now I'm planning the bass amps, those XLS are current eaters
Hi Raka,
I didn't notice the fold back of the baffle and I also thought intuitively that the H and W frames would have cavity resonance issues, I'd have an urge to line the inside walls with thick damping material if I ever did one of those. Linkwitz has commented for many years about box and room cavity resonances so I'd expect that he's got it under control, they're probably above the pass band of the woofers. Still I understand your thinking.
You might want to measure the distortion of the XT18s or even just listen for it with a sinewave input just to see if they have any unusual characteristics at high displacements. I'm sure you know to be careful of voice coil heating with continuous tones.
Looks like a very nice design, and that woodworking is really something. Maybe someday I'll do a system with a fine finish!
I didn't notice the fold back of the baffle and I also thought intuitively that the H and W frames would have cavity resonance issues, I'd have an urge to line the inside walls with thick damping material if I ever did one of those. Linkwitz has commented for many years about box and room cavity resonances so I'd expect that he's got it under control, they're probably above the pass band of the woofers. Still I understand your thinking.
You might want to measure the distortion of the XT18s or even just listen for it with a sinewave input just to see if they have any unusual characteristics at high displacements. I'm sure you know to be careful of voice coil heating with continuous tones.
Looks like a very nice design, and that woodworking is really something. Maybe someday I'll do a system with a fine finish!
Those look fabulous, Raka!
Have you experimented with the wings on the sides of your dipoles? They look the same dimension to me and I do recall reading a few times that making different size/shape wings helps.
Also, this may be a dumb question, but are you a female? Obviously it's a female picture you have up, but I'm not sure if that's you or someone else. Just curious because it's always nice to find a female that gets interested in DiY speakers like us guys do!
Now...if I could only get my fiancee into speakers....oh....how my life would be made easier!
Have you experimented with the wings on the sides of your dipoles? They look the same dimension to me and I do recall reading a few times that making different size/shape wings helps.
Also, this may be a dumb question, but are you a female? Obviously it's a female picture you have up, but I'm not sure if that's you or someone else. Just curious because it's always nice to find a female that gets interested in DiY speakers like us guys do!
Now...if I could only get my fiancee into speakers....oh....how my life would be made easier!
PB2 said:
I wasn't going to comment on this as Raka was happy with low output, but now it's come up again ...
There is a good example of dealing with resonances in dipole woofers (or similar variants) in John K's Nao design. What this illustrates is that the resonances can be dealt with effectively, and that it's worthwhile doing for the extra output and extension.
Raka, do you know what your fequal is with your current baffle?
mazeroth said:
<<< also wants to know
My guess is the pic is probably Raka's partner, I can only ever remember coming across one female on this forum. It seems that women like a good sound system, but don't get the same satisfaction usually out of talking about it, or building their own.
mazeroth said:
At prototype stage I tried different wing sizes, and my target was not to have big dips in the frequency response (on and off axis), and reasonable bass capacity, since I intend to equalize in active. Next project, I will try to make more studies in order to get a very well controlled radiation pattern. Not that I won't get one this project, just that will try different patterns and their relation with sound quality.
About the resonances, that is a good point, but at least this "box" is a big step compared to a box standard one. I'm studing about measurement, and resonance locations, so this is unexplored yet. Once the thing is finished, and playing, I'll make some tests.
The fequal for the midbass is around 200Hz and for the woofer is around 100Hz, but I'm talking by memory, so don't believe me.
The other question will remain unanswered as irrelevant, sorry.
Raka,
Very nice looking speakers. I have been working on a set of dipoles for about 9 months with the following drivers -
Peerless 850146
Audax HM130c0
ScanSpeak 9500
I'm using 2 woofers and 2 mids per side, with 24db LR active crossovers(see my post today in this forum on the crossover). If you get the compensation right you will love the sound of your dipoles.
I'm anxious to see what you think of the XLS woofers. I like their extra Xmax, but was somewhat concerned with their low Qts in an open baffle.
BobK
Very nice looking speakers. I have been working on a set of dipoles for about 9 months with the following drivers -
Peerless 850146
Audax HM130c0
ScanSpeak 9500
I'm using 2 woofers and 2 mids per side, with 24db LR active crossovers(see my post today in this forum on the crossover). If you get the compensation right you will love the sound of your dipoles.
I'm anxious to see what you think of the XLS woofers. I like their extra Xmax, but was somewhat concerned with their low Qts in an open baffle.
BobK
Some interesting issues have come up regarding driver choice for dipole woofers. First there is the issue of Qts. I have some crude midbass drivers on an open baffle for my computer speakers - one is a small and ultra cheap speaker from a pair of bookshelves which had no bass at all - they have a tiny magnet and probably a fairly high Qts. The other driver is a car audio 3 way. Normally I'd expect more bass from the car speaker, but on the open baffle, the cheapie has a lot more bass. Why does high Qts mean more bass in an open baffle? I have yet to figure this out.
Some interesting points:
* Adire's only driver designed specifically for dipole use has a Qts of 0.4 - not particularly high
* Linkwitz seems to pay no attention to Qts in the selection of a dipole woofer - he is more concerned with distortion performance and driver excursion permissible without audible turbulence
I suspect that lower Qts merely means more eq is needed - this is not a problem. The output will in the end be determined by how much excursion you can actually use. You can get around low Qts, but you can't get around things like distortion or turbulence with high excursion.
I have heard someone say that the Tumult is the ideal dipole woofer, but it seems to me it would be very difficult to use that much excursion without getting excessive turbulence. Can anyone with experience with the Tumult comment?
I believe one of the advantages of the XLS is that the distortion performance towards the upper end of the bass is very good. As the ability of the ear to perceive distortion appears to decrease as you go down lower, then it seems more critical at the upper end. The XLS may be a little better in that case than it immediately appears on paper.
Some interesting points:
* Adire's only driver designed specifically for dipole use has a Qts of 0.4 - not particularly high
* Linkwitz seems to pay no attention to Qts in the selection of a dipole woofer - he is more concerned with distortion performance and driver excursion permissible without audible turbulence
I suspect that lower Qts merely means more eq is needed - this is not a problem. The output will in the end be determined by how much excursion you can actually use. You can get around low Qts, but you can't get around things like distortion or turbulence with high excursion.
I have heard someone say that the Tumult is the ideal dipole woofer, but it seems to me it would be very difficult to use that much excursion without getting excessive turbulence. Can anyone with experience with the Tumult comment?
I believe one of the advantages of the XLS is that the distortion performance towards the upper end of the bass is very good. As the ability of the ear to perceive distortion appears to decrease as you go down lower, then it seems more critical at the upper end. The XLS may be a little better in that case than it immediately appears on paper.
You can raise the Qts of a driver by simply making the motor weaker. That will require less EQ in a dipole but have you really gained anything? I like to think of dipole EQ as cutting the highs rather than boosting the lows. From that perspective, all that really matters is the sensitivity at the lowest frequency you want to play. Whether you cut more or less as you go up in frequency is sort of irrelevant. This, of course assumes that you are using active EQ. If you want to try to do a dipole without EQ, you will need a high-Q driver to get something near a flat response.
Raka,
The rest of us can only judge you by what you post. In your avatar you are a female, so as far as I am are concerned, Raka is a female.
On the internet you can be anything you want people to think you are. If you do not wish to have a gender you can become something else.
As for me? Lately I am a living room with...ah..large speakers. Did anyone notice I'm not wearing a grille cloth?
The rest of us can only judge you by what you post. In your avatar you are a female, so as far as I am are concerned, Raka is a female.
On the internet you can be anything you want people to think you are. If you do not wish to have a gender you can become something else.
As for me? Lately I am a living room with...ah..large speakers. Did anyone notice I'm not wearing a grille cloth?
Paul,
High Qts means the driver has a natural low end boost that provides some of the dipole EQ that you would normally need to get accurate bass, which is why you heard more bass from the ostensibly inferior driver.
Linkwitz feels that this is a crude way to get flat response and he prefers to use active EQ to properly shape the response. I'm on his side in this, although I do feel that using the ultra low Qts of the Peerless XLS is going too far.
I use 12 inch drivers in my speakers and feel that the Adire DPL12, TC Sounds or Stryke AV12 would all be better choices than the XLS.
One other thing to note is that having a higher Q speaker doesn't mean you can play your speakers louder, just that you will need less EQ to get them there.
Cheers
Steve
High Qts means the driver has a natural low end boost that provides some of the dipole EQ that you would normally need to get accurate bass, which is why you heard more bass from the ostensibly inferior driver.
Linkwitz feels that this is a crude way to get flat response and he prefers to use active EQ to properly shape the response. I'm on his side in this, although I do feel that using the ultra low Qts of the Peerless XLS is going too far.
I use 12 inch drivers in my speakers and feel that the Adire DPL12, TC Sounds or Stryke AV12 would all be better choices than the XLS.
One other thing to note is that having a higher Q speaker doesn't mean you can play your speakers louder, just that you will need less EQ to get them there.
Cheers
Steve
As far as I understand it, Linkwitz prefers a rolloff of the driver equlal to Qt=0.5, for sonic reasons and group delay. If you have a Qts lower than that, you can correct it, making use of the excursion. With a Qts>0.5 you can't.
Ideally, the lower the excursion the better, so using the XLS is kind of compromise, hoping that the price to pay for the excursion is cheap than the price of the quality you get.
BTW, a very low Qes dictates the control of the amp, so I think it's good to have it as low as the XLS10, but I haven't performed serious tests, it's what the theory tells me.
Anyway, yesterday I plugged my Audigy and Kx drivers to my Spicatto, and with raw equalization, this thing sounds very good. The tests were done in a small room, but I have to say that the SS9700 is a fantastic driver, and that the dipole bass and midbass are just great. Switching back to reflex seems imposible now. BTW, those are my first full range speaker system (down to 20Hz), and it's great to know that there was a bass player in the band
Ideally, the lower the excursion the better, so using the XLS is kind of compromise, hoping that the price to pay for the excursion is cheap than the price of the quality you get.
BTW, a very low Qes dictates the control of the amp, so I think it's good to have it as low as the XLS10, but I haven't performed serious tests, it's what the theory tells me.
Anyway, yesterday I plugged my Audigy and Kx drivers to my Spicatto, and with raw equalization, this thing sounds very good. The tests were done in a small room, but I have to say that the SS9700 is a fantastic driver, and that the dipole bass and midbass are just great. Switching back to reflex seems imposible now. BTW, those are my first full range speaker system (down to 20Hz), and it's great to know that there was a bass player in the band
Raka said:..........speaker system (down to 20Hz), and it's great to know that there was a bass player in the band
Yes I am with you there, since I implemented my dipole subs into the system I thought to myself what on earth myself and others have been listening to previously and making comments about how nice and good sounding our system produces, especially to the jazz records. One of my friends came over and commented I have too much bass and demanded to turn down the volumn of the low output from my active crossover
He did not even aware he has been missing all those low frequency detail recorded on the CDs all the years.
I guess it takes time for people to adjust their ears especially those been conditioned by so many years of limited full range speakers.
Regards,
chris
Raka said:
The SPL generated by a driver is purely a function of displacement and the space it's operating into. Two drivers with the same Sd will have the same excursion at a given SPL. With the same nominal sensitivity the lower Q driver will just take more voltage to get there. This causes the amplifier to clip at lower SPLs with decreasing frequency and can be regarded as a feature: Linkwitz's 60W amplifier cannot cause over excursion in the Orion bass drivers above 22Hz.
A raw driver+baffle/enclosure with Qtd = .2 equalized for the same response as one with Qtd = .5 with the same resonance clips 8dB sooner there. With such a driver, you'd need to start with less than 10W to have the same fail-safe behavior.
The excursion trade-off comes from where and how the system rolls-off . With flat response dipole excursion increases 8X/octave. Lower order roll-offs lead to higher excursions. Beneath a fixed F3 point a system with a second order roll-off Q=.5 will have more output and excursion than one with Q=.7.
paulspencer wrote:
"There is a good example of dealing with resonances in dipole woofers (or similar variants) in John K's Nao design. What this illustrates is that the resonances can be dealt with effectively, and that it's worthwhile doing for the extra output and extension. "
John K. is basically doing a short TL, I've been lining TLs for over 30 years, this is probably the source of the urge to line an H frame. I don't follow John K's work much, but I find his discussion of lossy vented systems to be highly flawed, I use them, have simulated and built many with more damping than is normaly used, and believe me they work very well. All vented systems are not alike, I never thought of calling what I do a variant on vented but perhaps I should.
John K. wrote this about vented systems:
"Here, in part is where the apperant delima enters the picture
for vented boxes are based on the consept the the vent
makes the box act like an osscialtor. For the vented system to
perform well, this oscillator should have little damping. "
A vented system is an oscillator? I don't think so, this is an absurd claim. A vented system approximates a 4th order high pass filter, high pass filters are obviously not oscillators. Statements such as this make me question his work. He took a particular angle on the problem that misses the advantages. There are other advantages that he misses completely.
I also believe that there are errors in his dipole discussion since his bass section is a short transmission line, which I've studied in great detail.
For those who think dipole is the only way, I estimate that I need about 8 times the VD for a dipole to keep up with vented. I am curious about looking into the room interface and room mode excitation but that's for another day.
"There is a good example of dealing with resonances in dipole woofers (or similar variants) in John K's Nao design. What this illustrates is that the resonances can be dealt with effectively, and that it's worthwhile doing for the extra output and extension. "
John K. is basically doing a short TL, I've been lining TLs for over 30 years, this is probably the source of the urge to line an H frame. I don't follow John K's work much, but I find his discussion of lossy vented systems to be highly flawed, I use them, have simulated and built many with more damping than is normaly used, and believe me they work very well. All vented systems are not alike, I never thought of calling what I do a variant on vented but perhaps I should.
John K. wrote this about vented systems:
"Here, in part is where the apperant delima enters the picture
for vented boxes are based on the consept the the vent
makes the box act like an osscialtor. For the vented system to
perform well, this oscillator should have little damping. "
A vented system is an oscillator? I don't think so, this is an absurd claim. A vented system approximates a 4th order high pass filter, high pass filters are obviously not oscillators. Statements such as this make me question his work. He took a particular angle on the problem that misses the advantages. There are other advantages that he misses completely.
I also believe that there are errors in his dipole discussion since his bass section is a short transmission line, which I've studied in great detail.
For those who think dipole is the only way, I estimate that I need about 8 times the VD for a dipole to keep up with vented. I am curious about looking into the room interface and room mode excitation but that's for another day.
PB2 said:
I would call John K's Nao woofer system a U frame open back woofer. You could also call it an open back woofer. TL doesn't really adequately describe it, even though it has a resonance in a similar way to a TL. I think it is appropriate to draw a distinction here on the basis of design goals - a TL delibrately uses the tuning to extend LF performance. By contrast a U frame has this resonance effect as a side effect of increasing the effective path difference of front and rear sound waves from the driver. The resonance resulting from the length of the U frame is a problem to be alleviated, not the means to achieve the desired performance.
If you look at the material John has presented on dealing with this resonance, then you will see that it certainly appears sound, well explained and documented with measurements to demonstrate the points that he makes.
If you wish to question his work, then do him the courtesy of doing it appropriately, with clear and specific points on which you differ. Dont' be vague. It appears as if you question his "dipole discussion" on the basis of your issues with his discussion on vented alignments. I consider this irrelevant. If he has made mistakes in one area, it does not mean any of his discussion of U frames is flawed.
If you wish to question his work, it might also be appropriate to contact John.
I was hoping to have this thread go back to a discussion of Raka's speakers.
I do not want to get into a back and forth with you, I simply offered a comment to what I see as casual and flawed discussions by John K. His vented discussion is just one example, and his U frame analysis is another. He frequents this forum and can comment if he likes, but to be honest I don't want to get into a back and forth with him or you.
You state that it is not a transmission line and are wrong, he himself calls it a duct. Transmission line equations apply to the U frame structure. His analysis on that page, as I see it has many flaws based on what I know. My own analysis of transmission lines was done at the college level under the supervision of an expert professor in acoustics, audio, and electronics. I derived a full Thiel and Small model. John K. derives a model but the equations are not shown, and his derivation has errors based on the little information that he offers and what I know having done it in the past.
I have offered people the courtesy and long explanations in the past on the net, however the outcome is ususally not positive. Instead I offer my perspective and people can take it for what it's worth.
Everyone should be cautious of non peer reviewed work and I am simply offering a caution. Why don't we get back to Raka's work?
I do not want to get into a back and forth with you, I simply offered a comment to what I see as casual and flawed discussions by John K. His vented discussion is just one example, and his U frame analysis is another. He frequents this forum and can comment if he likes, but to be honest I don't want to get into a back and forth with him or you.
You state that it is not a transmission line and are wrong, he himself calls it a duct. Transmission line equations apply to the U frame structure. His analysis on that page, as I see it has many flaws based on what I know. My own analysis of transmission lines was done at the college level under the supervision of an expert professor in acoustics, audio, and electronics. I derived a full Thiel and Small model. John K. derives a model but the equations are not shown, and his derivation has errors based on the little information that he offers and what I know having done it in the past.
I have offered people the courtesy and long explanations in the past on the net, however the outcome is ususally not positive. Instead I offer my perspective and people can take it for what it's worth.
Everyone should be cautious of non peer reviewed work and I am simply offering a caution. Why don't we get back to Raka's work?
Nice disussion you have here, very interesting indeed in the theorical way, and as far as my point of view is correct, I see that the subject Bass reflex vs ML TL vs duct vs dipoles is giving things to think about the next step in the speakers construction.
About my speakers, I'm listening in stereo right now with them, and sound is very very good. About the imaging, the violinist is almost there, not in front of you, but a couple of steps beyond the speakers, with lots of air around. The amps I'm using now are a HT Denon and a diy design of a good designer here called PCP, a mosfet circuit that sounds very good. Not as good as an Aleph3 but clearly better than a Nad320.
Now I'm starting with the final filter and pcb design, as I suspect that the PC and the kx drivers are not "highend". I will show the details as soon as I have a final version.
About my speakers, I'm listening in stereo right now with them, and sound is very very good. About the imaging, the violinist is almost there, not in front of you, but a couple of steps beyond the speakers, with lots of air around. The amps I'm using now are a HT Denon and a diy design of a good designer here called PCP, a mosfet circuit that sounds very good. Not as good as an Aleph3 but clearly better than a Nad320.
Now I'm starting with the final filter and pcb design, as I suspect that the PC and the kx drivers are not "highend". I will show the details as soon as I have a final version.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.