Hihi people always want lower hehe
Lowmass where should the crossover be for yoir ribbon ?
By the way i dont use a backplate to eliminate that aspect. Yes i will hurt mt efficiency with 3db or so. Although it looks almost the same with one but if i do i need a tiney bit of shelving . But that might be just my baffle. Anyhow i will not hit those insane high spl with my distortion for sure
but it is a good start. Interesting project at least !! ( yours that is )Hihi people always want lower hehe
Lowmass where should the crossover be for yoir ribbon ?
I use it at 1Khz with 2nd order or higher cross. It can be used anywhere above that. The thrust of development has always been to get to 1khz with below 1% distort at 100db in a ribbon that doesn't have too much vertical dispersion issue and is no wider than about .75 inch.
To my ears once a tweeter is longer than about 3-4 inches or wider than about .75" I start to have issues with a relaxed listening position. And I find that if I dont have to take a woofer past about 1k that its almost too easy to get magic . The ribbon has a smooth roll off below crossover making it very easy to get cross right. No strong resonant hump.
By the way i dont use a backplate to eliminate that aspect. Yes i will hurt mt efficiency with 3db or so. Although it looks almost the same with one but if i do i need a tiney bit of shelving . But that might be just my baffle. Anyhow i will not hit those insane high spl with my distortion for sure
Per requests here is some other info...
1st - is a pulse for comparison from an inexpensive 3/4 inch silk dome. Same setup exactly as the close mic ribbon pulse in post 16 ( but admittedly done the next day). Perhaps not a good comparison but this is what I have at the moment.
Not shown here BUT very interesting to me is that this cheap dome had a very smooth freq response easily competing with the ribbon,low distortion also comparing well with ribbon, AND a CSD that looks even better than the ribbon!
However the Pulse reveals what my ears "see" clearly. The dome is NO match for the ribbon. In fact in listening tests even with a standard, higher distortion ribbon, the dome is not in same league.
In future I will get some better domes for compare,BUT again FR,CSD,and distortion simply do not tell the whole story in this case. At least as far as my ears are concerned, but the pulse is revealing. I suspect the domes nice FR and CSD are simply a well damped diaphragm that hides detail in its lossy construction.
2nd- is FR ( no smoothing) of Dynaplane ribbon ( red trace) and a typical corrugated ribbon construction of same size/mass/corrugation ( blue trace). Both ribbons mounted in same magnet structure and use same transformer
This to show what I find typically in simple corrugated ribbons ( see the sharp wiggle at 8.5 khz ) and some other deviations from smooth lower in freq.
The sharp wiggle at 8.5 Khz is the mass / spring resonance I see in All other "foil only", and to a lesser extent the laminated plastic backed ribbons, that are wider than about 10-12 mm.
That wiggle will have a long ugly tail in a CSD plot and can easily be moved around (lower or higher in freq.) depending on the size/shape of the corrugations.
To my ears this always sounds like added brightness an is fatiguing.
Less wide corrugated ribbons ( below about 10mm) will push this resonance out farther where its less noticed BUT now you have a very small ribbon with its more limited low freq use.
And to be sure we see some manufactures moving to "Flat" foil diaphragms that dont have this trait. However in my experience there are other issues with Flat and getting a small one to goto 1Khz results in diaphragm failure as well as poor control of some other resonances.Much of the work I have done is to eliminate that wiggle and others , in a wider ribbon diaphragm. The red trace tells that story
In my 1st ribbon patent I describe a construction where the corrugations are set within a certain range of angles rather than straight across. This construction increases the ribbons spring rate along its length without increasing mass and results in an improved centering force. This allowing use to lower freq. AND also eliminates that HF wiggle. However continued investigation of those designs has led to the present design we are showing data from in this thred
1st - is a pulse for comparison from an inexpensive 3/4 inch silk dome. Same setup exactly as the close mic ribbon pulse in post 16 ( but admittedly done the next day). Perhaps not a good comparison but this is what I have at the moment.
Not shown here BUT very interesting to me is that this cheap dome had a very smooth freq response easily competing with the ribbon,low distortion also comparing well with ribbon, AND a CSD that looks even better than the ribbon!
However the Pulse reveals what my ears "see" clearly. The dome is NO match for the ribbon. In fact in listening tests even with a standard, higher distortion ribbon, the dome is not in same league.
In future I will get some better domes for compare,BUT again FR,CSD,and distortion simply do not tell the whole story in this case. At least as far as my ears are concerned, but the pulse is revealing. I suspect the domes nice FR and CSD are simply a well damped diaphragm that hides detail in its lossy construction.
2nd- is FR ( no smoothing) of Dynaplane ribbon ( red trace) and a typical corrugated ribbon construction of same size/mass/corrugation ( blue trace). Both ribbons mounted in same magnet structure and use same transformer
This to show what I find typically in simple corrugated ribbons ( see the sharp wiggle at 8.5 khz ) and some other deviations from smooth lower in freq.
The sharp wiggle at 8.5 Khz is the mass / spring resonance I see in All other "foil only", and to a lesser extent the laminated plastic backed ribbons, that are wider than about 10-12 mm.
That wiggle will have a long ugly tail in a CSD plot and can easily be moved around (lower or higher in freq.) depending on the size/shape of the corrugations.
To my ears this always sounds like added brightness an is fatiguing.
Less wide corrugated ribbons ( below about 10mm) will push this resonance out farther where its less noticed BUT now you have a very small ribbon with its more limited low freq use.
And to be sure we see some manufactures moving to "Flat" foil diaphragms that dont have this trait. However in my experience there are other issues with Flat and getting a small one to goto 1Khz results in diaphragm failure as well as poor control of some other resonances.Much of the work I have done is to eliminate that wiggle and others , in a wider ribbon diaphragm. The red trace tells that story
In my 1st ribbon patent I describe a construction where the corrugations are set within a certain range of angles rather than straight across. This construction increases the ribbons spring rate along its length without increasing mass and results in an improved centering force. This allowing use to lower freq. AND also eliminates that HF wiggle. However continued investigation of those designs has led to the present design we are showing data from in this thred
Attachments
Last edited:
Thanks for description, and sorry for the slow reply. If I understood your posts correctly the waveguide is measured without a large baffle vs. the no-waveguide on a large baffle. If that is the case, I’d wager some of the lift you are seeing around and below 1khz is from diffraction. You might try comparing the waveguide to a flat baffle of the same outer dimensions to see what lift is coming from diffraction and what is coming from the conical shape. Alternatively, you could add a large baffle around the wave guide. In the end though, if you are getting the response you want when mounted in an enclosure that is what matters.
What is it that is revealed in the impulse that you don’t see in the frequency response? You mention the response was smooth, but the impulse says to me that it has an underdamped HP roll-off about 1500Hz or so. The impulse and response should say the same thing because they are mathematically inseparable. I’m not sure how you did your comparison, but you would need to properly damp this resonance and cross at least an octave above it to make the dome “happy”. I’m not saying the dome would match a ribbon’s sonic capability, just that it may need a bit more attention to achieve best performance.
Thanks Bolserst
The wave guides were tested with and without large baffle. There were 4 different wave guides. You are right about the smaller ones. They have some lift due to diffraction, but the larger ones ( about 16 by 16 inch)didn't show much if any difference with large baffle. A round over at the mouth ( no baffle) was also tried to good success
I have much to learn in this area for sure. Im particulary concerned about the off axis response and some of the medium sized ( about 10 by 10 inches mouth with no baffle) would show the 1 Khz region not dropping much at all as you went off axis until you got to about 55-60 deg, then it drops quick. Not a smooth steady fall at each 20 deg increment. Is this the diffraction at work??
Seems to me with my limited understanding that we want a preety much even drop across the board as we go off axis? But looking at some published stuff using WG ,they seem to ride high at the lower end also.
BTW you say if getting response you want ( with diffraction) then its OK. Is this really tru?? if so, then why use a large WG when a smaller on with diffraction will give desired response? Wont the overall off axis stuff be less perfect using diffraction to get desired response? My desire is not only a sensitivity lift BUT good control of directivity.
You are also right about the dome. I was doing very close mic measures for the pulse response of dome and ribbon. Again much to learn here BUT I was taking advise from Zaph on this measure. Seems when I try to do pulse farther away I don't always get consistent results. Also when I mentioned the domes smooth FR I meant above about 2 khz on out. I seemed to have lost the plot. BUT if I remember right It was smooth rolloff below that too. Its a ferrofluid damped motor and a silk dome about 3/4 inch diameter. Its a very inexpensive tweeter. It biggest issue I saw was a suckout in response between about 4 and 8 k ( I assume the dome is caving in here?) cant remember exactly.
I was then taking freq response of both at about 1/2 meter to avoid issues in my small room and some structure close by. I would always compare FR at 1/2 and 1 meter to make sure I wasn't introducing some big issue. So long as they were close I went ahead and did 1/2 meter.
Distortion measures were at 1 meter
Advice welcomed....
The wave guides were tested with and without large baffle. There were 4 different wave guides. You are right about the smaller ones. They have some lift due to diffraction, but the larger ones ( about 16 by 16 inch)didn't show much if any difference with large baffle. A round over at the mouth ( no baffle) was also tried to good success
I have much to learn in this area for sure. Im particulary concerned about the off axis response and some of the medium sized ( about 10 by 10 inches mouth with no baffle) would show the 1 Khz region not dropping much at all as you went off axis until you got to about 55-60 deg, then it drops quick. Not a smooth steady fall at each 20 deg increment. Is this the diffraction at work??
Seems to me with my limited understanding that we want a preety much even drop across the board as we go off axis? But looking at some published stuff using WG ,they seem to ride high at the lower end also.
BTW you say if getting response you want ( with diffraction) then its OK. Is this really tru?? if so, then why use a large WG when a smaller on with diffraction will give desired response? Wont the overall off axis stuff be less perfect using diffraction to get desired response? My desire is not only a sensitivity lift BUT good control of directivity.
You are also right about the dome. I was doing very close mic measures for the pulse response of dome and ribbon. Again much to learn here BUT I was taking advise from Zaph on this measure. Seems when I try to do pulse farther away I don't always get consistent results. Also when I mentioned the domes smooth FR I meant above about 2 khz on out. I seemed to have lost the plot. BUT if I remember right It was smooth rolloff below that too. Its a ferrofluid damped motor and a silk dome about 3/4 inch diameter. Its a very inexpensive tweeter. It biggest issue I saw was a suckout in response between about 4 and 8 k ( I assume the dome is caving in here?) cant remember exactly.
I was then taking freq response of both at about 1/2 meter to avoid issues in my small room and some structure close by. I would always compare FR at 1/2 and 1 meter to make sure I wasn't introducing some big issue. So long as they were close I went ahead and did 1/2 meter.
Distortion measures were at 1 meter
Advice welcomed....
Last edited:
BTW heres a retake that dome, first is FR and 2nd is pulse at 1/2 meter 2 ms gate direct drive no smooth 2dB / div
3rd is the ribbon pulse in same setup within about 5 minuets of each other
I think in past I may have been putting a cap in there on dome to avoid fire and that likely was flattening the hump
3rd is the ribbon pulse in same setup within about 5 minuets of each other
I think in past I may have been putting a cap in there on dome to avoid fire and that likely was flattening the hump
Attachments
Last edited:
Yes, the off-axis “bloom” you are seeing is from diffraction. The frequency range it occurs in is related to the size of the baffle or mouth. The only way around it is to minimize the baffle so that the entire width is driven. Gerrit Boers posted a nice set of polars for the BG Neo10 with several different baffle widths that illustrate this behavior with flat baffles. The "bloom" is more pronounced with dipoles than with monopoles, but this will give you and idea what behavior to expect. The Totem of Tone, an active 3-way dipole and active subwoofer
I meant that since diffraction is defined by the enclosure size you choose, if you have optimized your wave guide to get smooth flattish response with that size, then that is about the best you can do. If making a 24” wave guide is doable for you, then you could push the range of constant directivity down a noticeable amount. Gedlee (Earl Geddes) used a 15” waveguide with careful attention to throat and mouth transitions that has good control down to 1Khz (response overlays attached). I also had posted a polar map of the same speaker if you prefer that format: Monsoon - Surprisingly Good!
Gedlee also posted some general comments/guidance on tradeoffs in size, directivity, and transition frequency you might find useful: Reducing Waveguide Size
Attachments
Thank you much bolserst !!
I will dig into threds you mention.
I have been looking a bit at Geddes work. So far its interesting to me that I too land at about a minimum of about 15 inch mouth width to get what looks acceptable to me for a 1Khz cross.
I seem to get a "decent" response with smaller WG too BUT the 15 or bigger measures better.
I see a few commercial offerings such as the Bema AMT in a WG that claim 1Khz BUT these WGs are only maybe 9-10 inch mouths and from what I see they actually 6 db down at 1K on axis. They don't post the off axis info. Seems to me if we want to go to 1k it would be better to hold level to below that mark for a bit.
Anyway like I said I seem to get a reasonable response with a 9-10 inch WG AND I suppose at some point I have to put the measuring equip aside and simply use the ear to compare the 10 inch and the larger ones to see how much trade off is really an issue.
Also interesting to me is that most of my wave guides so far seem to have their best ( flattest) response at about 20-25 degrees off axis.
And tru straight sides with attention to the mouth/throat transitions always seems to beat any curved sided form.
Thanks again for the clear and helpful response
I will dig into threds you mention.
I have been looking a bit at Geddes work. So far its interesting to me that I too land at about a minimum of about 15 inch mouth width to get what looks acceptable to me for a 1Khz cross.
I seem to get a "decent" response with smaller WG too BUT the 15 or bigger measures better.
I see a few commercial offerings such as the Bema AMT in a WG that claim 1Khz BUT these WGs are only maybe 9-10 inch mouths and from what I see they actually 6 db down at 1K on axis. They don't post the off axis info. Seems to me if we want to go to 1k it would be better to hold level to below that mark for a bit.
Anyway like I said I seem to get a reasonable response with a 9-10 inch WG AND I suppose at some point I have to put the measuring equip aside and simply use the ear to compare the 10 inch and the larger ones to see how much trade off is really an issue.
Also interesting to me is that most of my wave guides so far seem to have their best ( flattest) response at about 20-25 degrees off axis.
And tru straight sides with attention to the mouth/throat transitions always seems to beat any curved sided form.
Thanks again for the clear and helpful response
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.