I studied industrial and furniture design and also took a plastic class. Unfortunately it was not my favourite especially working with epoxy but with your help i will master the task.
The last time i put a tweeter in a smaller sphere i worked with an electrical wood saw to cut the hole. That was not too difficult. For experimenting i have a sticky sealant until i come up with a method to fix the driver permanently.
The last time i put a tweeter in a smaller sphere i worked with an electrical wood saw to cut the hole. That was not too difficult. For experimenting i have a sticky sealant until i come up with a method to fix the driver permanently.
Alex, look at post 307. You see a very tiny tweeter on top of the nylon sphere and the aproximate position of the midrange is shown as a back ring.
Epoxy now can be applied by using a mixing head, to you can apply it just like using hot melt glue. However, I do not know how it works on plastic, but there is also glue expecially made for joning plastic. I have used it and it seems to work quite well.
Glues just do not work well on untreated polyethylene or polypropylene. You need surface treatment, heat, or both.
There's a chemical way to prepare the surface (chromic acid) so it will take adhesive, but that's really toxic and dangerous. Effective, though.
There's a chemical way to prepare the surface (chromic acid) so it will take adhesive, but that's really toxic and dangerous. Effective, though.
Joachim,
Gluing PE is near hopeless. It is normally heat-bonded or as Sy pointed out, special glues exist that require flame treating the surface (it does not look good afterwards)...
For a prototype I would make a foam (EVA) gasket, to seal things well and use the so-called "Decken Anker" bolts (or maybe just ordinary wide expansion wall plugs) with a suitable arrangement so that the bolts hold the midrange purely mechanical, by tension.
Head over to your local Obi and have a look what you can find.
Ciao T
Gluing PE is near hopeless. It is normally heat-bonded or as Sy pointed out, special glues exist that require flame treating the surface (it does not look good afterwards)...
For a prototype I would make a foam (EVA) gasket, to seal things well and use the so-called "Decken Anker" bolts (or maybe just ordinary wide expansion wall plugs) with a suitable arrangement so that the bolts hold the midrange purely mechanical, by tension.
Head over to your local Obi and have a look what you can find.
Ciao T
Hi Joachim
I had to find a way to bond ferrite to glass filled nylon once, which has very low surface energy. Epoxies, acrylics, silicones were all useless. The only thing I found that worked well was flexible modified polyurethanes that are commonly used in glazing applications: Sikaflex-256 or similar.
Good luck
I had to find a way to bond ferrite to glass filled nylon once, which has very low surface energy. Epoxies, acrylics, silicones were all useless. The only thing I found that worked well was flexible modified polyurethanes that are commonly used in glazing applications: Sikaflex-256 or similar.
Good luck
Last edited:
Hello,
I would like to source a pair of 3 liter spheres in the US, does anyone know where?
DT
All just for fun!
I would like to source a pair of 3 liter spheres in the US, does anyone know where?
DT
All just for fun!
Hello Joachim,
I sent a PM.
Friday i received some OEM drivers made by
Tymphany for Sonus faber.
Tymphany TPY04W16S0086 5" Woofer 16 Ohm | Parts-Express.com
The modeled sealed enclosure is 2.7 l. Looks like a mid-range that may perform OK at least to play with.
DT
All just for fun!
I sent a PM.
Friday i received some OEM drivers made by
Tymphany for Sonus faber.
Tymphany TPY04W16S0086 5" Woofer 16 Ohm | Parts-Express.com
The modeled sealed enclosure is 2.7 l. Looks like a mid-range that may perform OK at least to play with.
DT
All just for fun!
It seems that it's possible to improve CSD as well.
Below, before equalized by Ultimate Equalizer
An externally hosted image should be here but it was not working when we last tested it.
and being equalized by Ultimate Equalizer
An externally hosted image should be here but it was not working when we last tested it.
Both frequency and phase are flat.
I probably should redo the measurement with the same sample rate. Don't know what I was thinking when I did this.
With this kind of performance, other forms of distortion are going to be even more important.
Last edited:
DT, i got your PM. Sorry that i did not answer so far. I was very busy and had to acomodate a lot of visitors. That driver looks fine to me. Unfortunately i can see no frequency response and phase response data. Are you able to measure it ! 3 Liters may be a little small for 5" driver.
Soongcs, yes, when you equalize for flat frequency respone and phase you get much less stored energy. Your result looks very encouraging.
Hi All ! I got the woofers today so i will build the bass cabinets sonn.
Soongcs, yes, when you equalize for flat frequency respone and phase you get much less stored energy. Your result looks very encouraging.
Hi All ! I got the woofers today so i will build the bass cabinets sonn.
Its always nice to look at people "fighting" diffraction (like good ol' Don Quijote) - it seems that only few - Thorsten and Elias around here for example - are willing to look any deeper into the topic
Aleksander your nice tweeter actually *is* at a horn (at certain frequencies at least) - no question about - and I'm pretty sure it performs as you said - just adding the turnaround effect mentioned by Elias as well.
For Joachim to apply your findings would require to scale up your "horn" for the mid and the bass and the sub as well (ok possibly this is a little bit too pronounced 🙂 ).
Before mentioned Cepstrum or wavelet analysis would certainly show or some simple "mind experimenting" would do equally
--
As a side note those "turnaround" diffraction will add to ripples as will the diffraction effects from the rim of the speaker tweeters and mid (hence different "FR" for different close mic distance)
All in all I'm very much reminded by this thread on good ol' Klinger - which certainly is in the mind of the German speaking around here...
(well a little bit pimped up by some nowadays name dropping 😉 )
Michael
Aleksander your nice tweeter actually *is* at a horn (at certain frequencies at least) - no question about - and I'm pretty sure it performs as you said - just adding the turnaround effect mentioned by Elias as well.
For Joachim to apply your findings would require to scale up your "horn" for the mid and the bass and the sub as well (ok possibly this is a little bit too pronounced 🙂 ).
Before mentioned Cepstrum or wavelet analysis would certainly show or some simple "mind experimenting" would do equally
--
As a side note those "turnaround" diffraction will add to ripples as will the diffraction effects from the rim of the speaker tweeters and mid (hence different "FR" for different close mic distance)
All in all I'm very much reminded by this thread on good ol' Klinger - which certainly is in the mind of the German speaking around here...
(well a little bit pimped up by some nowadays name dropping 😉 )
Michael
Yes Michael. This is nothing new as the Elipson speakers from the 50th show. Where we have made progress though is in linearity of drivers and crossover design.
In the end it matters what comes out of it and as i said before, the sound of the Stoll speakers is unique in terms of 3 dimensional imaging so there must be something going on that is very right and different to anything i came across. Certainly a waveguide design put in the farfield does not cut the mustard here. At least not for me although you can see in my MPL thread that i experiment with more narrow DI too. In my setup the big MPL does work pretty well too. The soundstage i get is very high and wide but not extremely deep. That fits life music very well and gives the impression to sit quite close to the stage with musicians playing from quite upstairs. The Stoll speakers though listened from close up give the impression that the speakers totally disappear. It is a kind of surround sound experience with very good imaging and depth. I think what is really important is to do the details right and i am also prepared to work with external damping when i can not control diffraction well enough with only the outer shape.
In the end it matters what comes out of it and as i said before, the sound of the Stoll speakers is unique in terms of 3 dimensional imaging so there must be something going on that is very right and different to anything i came across. Certainly a waveguide design put in the farfield does not cut the mustard here. At least not for me although you can see in my MPL thread that i experiment with more narrow DI too. In my setup the big MPL does work pretty well too. The soundstage i get is very high and wide but not extremely deep. That fits life music very well and gives the impression to sit quite close to the stage with musicians playing from quite upstairs. The Stoll speakers though listened from close up give the impression that the speakers totally disappear. It is a kind of surround sound experience with very good imaging and depth. I think what is really important is to do the details right and i am also prepared to work with external damping when i can not control diffraction well enough with only the outer shape.
IMO what you (and Dominic Stoll) are after is the same effect those love who are operating nude speakers.
With both attempts the diffraction alignment issue is cut down not to try to go for rounding the edges - which is the technically pretty different WG / horn attempt - but to cut down time to an absolute minimum as where those diffraction caused "echoes" occur.
Cutting down a baffle finds its end when you reach the basket so to say (and there are possibly other factors involved as well).
Anyway - you should stop in "fighting" diffraction (I mean we already had a looooot of junk discussion with someone who claimed to do "least diffraction WGs" already).
Diffraction is a necessity !
Its a *necessity* of the transport media between you (the listener) and the speaker.
Of course - only if you do not prefer to listen with a tube attached between your ears and the speakers (or prefer point source omni).
So - what you can focus on is to get the *effects* of that unavoidable and "always equally big" diffraction (!) into some alignment that suits your taste.
By the way - to what extend did you succeed on the dampening of the turnaround effect by absorptive material?
Michael
With both attempts the diffraction alignment issue is cut down not to try to go for rounding the edges - which is the technically pretty different WG / horn attempt - but to cut down time to an absolute minimum as where those diffraction caused "echoes" occur.
Cutting down a baffle finds its end when you reach the basket so to say (and there are possibly other factors involved as well).
Anyway - you should stop in "fighting" diffraction (I mean we already had a looooot of junk discussion with someone who claimed to do "least diffraction WGs" already).
Diffraction is a necessity !
Its a *necessity* of the transport media between you (the listener) and the speaker.
Of course - only if you do not prefer to listen with a tube attached between your ears and the speakers (or prefer point source omni).
So - what you can focus on is to get the *effects* of that unavoidable and "always equally big" diffraction (!) into some alignment that suits your taste.
By the way - to what extend did you succeed on the dampening of the turnaround effect by absorptive material?
Michael
Last edited:
Great to hear about new progress!
I'm thinking probably my measurements need to be in an anechoic chamber to get things better.
From this point on many Off Topic posts have been moved to this thread:http://www.diyaudio.com/forums/everything-else/173283-cmp-framing-what-you-mean.html_________________________________________________________________
To be honest i do not understand some things that have being brought up here too. That may be ralated to language ( yes, i hate too when we can not universally agree on terms )
like "phase linearity", "linear phase", "constant groop delay", "time alignement", "time coherence" etc. and partly simply to missunderstanding or stupidity.
This "turnaround effect" was certainly new to me and i studied it with Cepstral Analysis.
I reported what i found by measuring a very small and round tweeter "open air" and compared it with a conventional tweeter with a frontplate. In case of the tiny tweeter i could not find a market difference, with and without damping the back. The conventional tweeter like measured by Elias had a lot more problems with this test so i can conclude that when the object is small enough for the frequency range in question the turnaround effect is a third order effect that should not be overemphasized.
Last edited by a moderator:
Today i got some interesting drivers. The new Vifa 19mm Aluminium-Ceramic tweeters,
the Scan Discovery 15cm midranges and the 25cm Peerless woofers for the ZDL.
I will measure the Vifa tweeters in open air like i did with the even smaller OX tweeter. The Scan midrange uses a fiber cone that is coated and cut at the edge (NRS- non resonant sourround). The cone apears to have a NAVI shape and is very stiff but well damped. The surround is a low loss foam. Sensitivity is quite high at 92dB but it is a 4 Ohm design. The magnet system has an aluminum Faraday ring to cut back inductance and lower modulation distortion. This driver coud be used in the deeper midrange in a 4 Way ZDL. I already have made good experiences with the 30cm version of the woofer so i do not expect a bad surprise here.
the Scan Discovery 15cm midranges and the 25cm Peerless woofers for the ZDL.
I will measure the Vifa tweeters in open air like i did with the even smaller OX tweeter. The Scan midrange uses a fiber cone that is coated and cut at the edge (NRS- non resonant sourround). The cone apears to have a NAVI shape and is very stiff but well damped. The surround is a low loss foam. Sensitivity is quite high at 92dB but it is a 4 Ohm design. The magnet system has an aluminum Faraday ring to cut back inductance and lower modulation distortion. This driver coud be used in the deeper midrange in a 4 Way ZDL. I already have made good experiences with the 30cm version of the woofer so i do not expect a bad surprise here.
Attachments
Brian, i will not let that happen here. I heard several of Stolls speakers. That gives me the advantage of ear witness
I know how they sound and i know what i want. Making the sound sources mechanical very small and round gives a flat on and off axis response over a very wide frequency range provided the drivers are of good quality. The next step is to mount them as close together as posible while avoiding interference and then aligne the acoustic centers. Then a crossover has to be designed that does add as little group delay as posible. I am personally not very interested in "name dropping" to make the thread more exiting or give the impression that something amasing is going on here but the child needs a name. I am dedicated to design a DIY speaker within resonable cost and efford that gives a glimps of high end heaven. It should measure well, actually better then 90% of the current crop of high end speakers over 10.000,-€, and it should sound well. As prove of pudding i invite everybody that is contributing something good here to my home to listen to the finished speaker. That is one of the problems Michael, as much as we can theorytise how a good speaker should be designed and measured, ultimately soundquality can not be measured nore described in words. I do not remember who said that but "It is like dancing to architecture".
I know how they sound and i know what i want. Making the sound sources mechanical very small and round gives a flat on and off axis response over a very wide frequency range provided the drivers are of good quality. The next step is to mount them as close together as posible while avoiding interference and then aligne the acoustic centers. Then a crossover has to be designed that does add as little group delay as posible. I am personally not very interested in "name dropping" to make the thread more exiting or give the impression that something amasing is going on here but the child needs a name. I am dedicated to design a DIY speaker within resonable cost and efford that gives a glimps of high end heaven. It should measure well, actually better then 90% of the current crop of high end speakers over 10.000,-€, and it should sound well. As prove of pudding i invite everybody that is contributing something good here to my home to listen to the finished speaker. That is one of the problems Michael, as much as we can theorytise how a good speaker should be designed and measured, ultimately soundquality can not be measured nore described in words. I do not remember who said that but "It is like dancing to architecture".
- Status
- Not open for further replies.