Help with Silk Dome Tweeter Waveguide

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
We debated the term in Earl's thread not too long ago. I tend to agree with your definition and call it a waveguide when it is primarily designed for directivity reasons. In this case there is no compression driver so horn seems inapropriate. Still a semantic issue.

As to the shape, the most constant directivity comes from flares that start as straight conical and then flare to a wider angle at a distance about 2/3 out. Look at Keele's papers or copy something as used by Genelec. I would start with a 120 degree initial flare angle. Narrow angles will give worse frequency response.

AES Papers -- Official website of D.B.Keele

David S.
p.s. I've had better luck with non-circular guides such as the Snell one on my Avatar. Better frequency response.
 
Last edited:
By the way, you can prototype conical flares easily with manilla card stock: the thin cardboard that file folders are made from. I have made many samples with 2 part conical flares that way. Mock something up in cardboard and after a few trials, if you have something you like, then make it in wood.

10 minutes of geometry will let you figure out how to cut ring sections that can be put together to form conical flares that join up. The inner radius matches the tweeter diameter just across the surround. I always taped the sections together with small strips of masking tape. For a tweeter, the cardboard is more than rigid enough to give the same frequency response as a final molding. It let me try all maner of initial flare, final flare and linking radius.

Here is an example of a flare that I did at PSB. It is essentially 2 conical sections with a soft radius linking section. Worked very well and was well modeled initially in cardboard.

David S.
 

Attachments

  • psb cw180.jpg
    psb cw180.jpg
    64.1 KB · Views: 225
Getting the curve out

Hi Folks,

I think i've more or less finalised the curvature of my horn/waveguide so now i need to get the contour accurately out onto printed paper at the correct size.

This is so i can cut a hardboard template for the lathe. Can someone suggest the best way to do this. I've tried exporting the data and pulling it into excel and creating a graph from the datapoints but it doesnt want to align properly. I'm also dubious about the accuracy of excel and the size of the graph.

Thanks,

Steve.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.