Thank you for sharing your excellent work. Would you mind also making available the stl files for the two 5 inch waveguides?
-depends on the resin and its hardner (..assuming it has fully hardned).
You could use almost any resin really that fits your needs. Smooth-on is best for it's specialty products but particularly for it's form material.
The most obvious resin from them would be this (..and it has its technical detail at the right of the page):
EpoxAcast™ 650 + 101 Hardener Product Information | Smooth-On, Inc.
-as far as cost is concerned though, I'd probably experiment with cheap gallon epoxy and hardner and use urn-sand as filler/media. Once you get the balance right it should be very inexpensive overall. Consider a large form with multiple waveguide forms - like a super-sized silicone cookie sheet, and a resin mixture with a lot of urn-sand (as a percentage of the resin's volume) and I'm guess that the total cost would be less than $5 to produce each waveguide (for materials), ..and more likely half that. You could do it even cheaper with cement (as opposed to resin), but cement shrinks and is prone to cracking (and it probably wouldn't bond well to the phase-shield).
Note that the gallon unit of the Smooth-on is still a good price, but remember that it already has its sand filling included - so a lot of the volume is cheap sand, NOT resin:
Buy EpoxAcast(R) 650 and EpoxAcast(R) 650 BLACK Mineral Filled General Purpose Casting Epoxy from Reynolds Advanced Materials
You could use almost any resin really that fits your needs. Smooth-on is best for it's specialty products but particularly for it's form material.
The most obvious resin from them would be this (..and it has its technical detail at the right of the page):
EpoxAcast™ 650 + 101 Hardener Product Information | Smooth-On, Inc.
-as far as cost is concerned though, I'd probably experiment with cheap gallon epoxy and hardner and use urn-sand as filler/media. Once you get the balance right it should be very inexpensive overall. Consider a large form with multiple waveguide forms - like a super-sized silicone cookie sheet, and a resin mixture with a lot of urn-sand (as a percentage of the resin's volume) and I'm guess that the total cost would be less than $5 to produce each waveguide (for materials), ..and more likely half that. You could do it even cheaper with cement (as opposed to resin), but cement shrinks and is prone to cracking (and it probably wouldn't bond well to the phase-shield).
Note that the gallon unit of the Smooth-on is still a good price, but remember that it already has its sand filling included - so a lot of the volume is cheap sand, NOT resin:
Buy EpoxAcast(R) 650 and EpoxAcast(R) 650 BLACK Mineral Filled General Purpose Casting Epoxy from Reynolds Advanced Materials
Last edited:
Waveguide newbie...
I am new to this thread and to waveguides in general, but I do understand how waveguides can do good things for tweeters.
I am currently juggling different ideas related to the Open Source Speaker design over here, and I believe a tweeter + waveguide might be interesting for this project. To get myself into waveguides I got myself a few Visaton waveguides (they were cheap) that I might stuff onto some ScanSpeak 6600 tweeters that I have in my parts box. Those parts don't quite fit, but I will solve that somehow (using the tried and tested MacGyver approach) so that I can have a play with it.
I don't expect this to be the final/best solution, and I actually prefer Be dome tweeters like the ScanSpeak 6640 or the Satori TW29BN. From screening this thread I tend to believe that those tweeters don't play well with (custom?) waveguides. Is that correct? Why? What makes a tweeter a good tweeter for waveguides?
I am new to this thread and to waveguides in general, but I do understand how waveguides can do good things for tweeters.
I am currently juggling different ideas related to the Open Source Speaker design over here, and I believe a tweeter + waveguide might be interesting for this project. To get myself into waveguides I got myself a few Visaton waveguides (they were cheap) that I might stuff onto some ScanSpeak 6600 tweeters that I have in my parts box. Those parts don't quite fit, but I will solve that somehow (using the tried and tested MacGyver approach) so that I can have a play with it.
I don't expect this to be the final/best solution, and I actually prefer Be dome tweeters like the ScanSpeak 6640 or the Satori TW29BN. From screening this thread I tend to believe that those tweeters don't play well with (custom?) waveguides. Is that correct? Why? What makes a tweeter a good tweeter for waveguides?
It's not really anything to do with the dome material. It's the size of the dome that is the issue, and all of these Be tweeters just happen to be on the larger side - 29mm. And with very large surrounds making the throat even wider and pushing the reflections/nulls down in frequency into the 10-20khz area. If you could get the diameter at the outside of the surround to 1.3" or so, you would be in good shape.
I see... but: the ScanSpeak 833000 seems to do pretty well in a waveguide. The 833000 is specified with an (effective) dome diameter of 32mm, whereas the ScanSpeak 6640 and 6600 have a smaller dome specified at 30mm. Is there hope?
You have to consider the surround diameter also, since that is what sets the throat diameter.
I don't know why some tweeters do well and others not. Like that 833000 on a round wave guide. I have some ideas though, I think I'm going to revisit using a partial ellipse for the profile of the wall.
I don't know why some tweeters do well and others not. Like that 833000 on a round wave guide. I have some ideas though, I think I'm going to revisit using a partial ellipse for the profile of the wall.
Take a look at the Sd specs, which should take into account half of the surround. The 833000 Sd is 8cm², the 6640 Sd is 7cm². This suggests the diameter of the 6640 is less than that of the 833000. But that's just hand waving... I'll have to put that waveguide on my 6600 and see for myself how this works.
Hi,
I think I used a Mac-based speaker measurement software you developed many years ago... nice to meet here.
A second criteria might be the directivity of the tweeter at high frequencies. A higher directivity tends to minimize interaction with the WG throat which help to get a smooth high frequency range. That might be the reason, why ring radiator tweeter often work well with custom waveguides.
And I'll definitely second augerpro: the diameter of the tweeter (including surrounding).
As you are familiar with German language, you might be interested to take a look at the thread on 'how-to constant directivity'-thread at the German DIY-hifi-forum - quite some information here. Some of the simulations indicate that it might be beneficial, if the tweeter membrane is perpendicular to the WG throat (even if it needs to implement a 'step' at the WG-throat to achieve this).
Bottomline: It's kind of tiral and error to ideally match a tweeter to a custom WG - or you construct the WG to a specific tweeter
I think I used a Mac-based speaker measurement software you developed many years ago... nice to meet here.
Well, not easy to answer. Some issues here (disclaimer - with quite some speculations): Mechanically it's about the adaption to the WG-throat, i.e. is the front plate of the tweeter plane or rounded around the dome? Can you remove the faceplate ('Zwangszentrierung') to fit the dome tho the WG-throat? A good fit of the dome/surround to the throat. With a custom WG that's just by chance.
A second criteria might be the directivity of the tweeter at high frequencies. A higher directivity tends to minimize interaction with the WG throat which help to get a smooth high frequency range. That might be the reason, why ring radiator tweeter often work well with custom waveguides.
And I'll definitely second augerpro: the diameter of the tweeter (including surrounding).
As you are familiar with German language, you might be interested to take a look at the thread on 'how-to constant directivity'-thread at the German DIY-hifi-forum - quite some information here. Some of the simulations indicate that it might be beneficial, if the tweeter membrane is perpendicular to the WG throat (even if it needs to implement a 'step' at the WG-throat to achieve this).
Bottomline: It's kind of tiral and error to ideally match a tweeter to a custom WG - or you construct the WG to a specific tweeter
Thx I forgot that you have to remove the faceplate to mount the tweeter on the waveguide.
This is exactly what I was thinking and why wanted to revisit the ellipse as profile. I can be steep (90 degree) at the dome, but than smoothly transitioning to the baffle.
I wish SB would just release the Be version of the SB26 they are making for Revel!
Oh, wow, that was a looooong time ago!
Well, not easy to answer. Some issues here (disclaimer - with quite some speculations): Mechanically it's about the adaption to the WG-throat, i.e. is the front plate of the tweeter plane or rounded around the dome? Can you remove the faceplate ('Zwangszentrierung') to fit the dome tho the WG-throat? A good fit of the dome/surround to the throat. With a custom WG that's just by chance.
A second criteria might be the directivity of the tweeter at high frequencies. A higher directivity tends to minimize interaction with the WG throat which help to get a smooth high frequency range. That might be the reason, why ring radiator tweeter often work well with custom waveguides.
And I'll definitely second augerpro: the diameter of the tweeter (including surrounding).
As you are familiar with German language, you might be interested to take a look at the thread on 'how-to constant directivity'-thread at the German DIY-hifi-forum - quite some information here. Some of the simulations indicate that it might be beneficial, if the tweeter membrane is perpendicular to the WG throat (even if it needs to implement a 'step' at the WG-throat to achieve this).
Bottomline: It's kind of tiral and error to ideally match a tweeter to a custom WG - or you construct the WG to a specific tweeter
I had a look at that thread, which seems interesting. However, the figures don't show and it's a bit difficult to understand what you did without the figures.