Geddes on Waveguides

Thanks everyone!

I have a simple task of taking 1/2" wood and routing out holes equal to the measurements on those spreadsheets.

I would love to get the spread sheet broken down to 1/2" chunks so I can match the angles with router bits.

Then glue all the pieces together to make a simples waveguide out of wood.\

Im just trying to get a workable molding. The DDS OS waveguide is very hard to source, Geddes doesnt sell his waveguides alone any more. It would be nice ot source a GOOD OS waveguide.
 
Last edited:
Geddes doesnt sell his waveguides alone any more. It would be nice ot source a GOOD OS waveguide.

Just to be clear, I did, but I found that there was no value in it for me. I believe that this is why there aren't any on the market right now. It seems like a simple part, and maybe it is if you have a lot of money for tooling, etc. but in small lots its very expensive and people just don't want to pay a price that is interesting to a manufacturer.

Some time ago there was a long thread about making waveguides. The overwhelming postion at that time was that they would be easy to make and that I was just gouging the market. OK, where are they then? That thread was the reason that I stopped selling them. Simply put the waveguide, foam plug and mounting plate are the key to what I make and sell and represent the greatest value-added by far. I am not willing to give this away without due compensation for this value added.

I don't make stand alone waveguides, they are molded into a custom baffle. It is extra work for me to cut them out. I have, and will, sell the baffles with foam plug and mounting plate. Just contact me on the side because I do not advertise this. But be fore warned, they are not cheap.
 
After going through months of effort and lots of resin, cloth, foam, wood, plastic, sandpaper, and filler putties, I can confirm- one-offs are exceedingly obnoxious to do in synthetic materials. My production method wasn't ideal either, however. If I had it to do over, I'd do it in wood rings. The results I'm getting are exceedingly nice, but I'm nearing $150 in materials and a minimum of 40 "pure" work hours (no downtime, that's all spraying, spreading, sanding, cutting time). One can go quick and dirty, I got usable results from a foam mold and a plaster cloth layup, but

After all these months, I'm being a little anal about the result- I want nothing less than complete perfection, as I'm so vested in this.

Next ones I make with wood and a router. In retrospect, that would have been much MUCH easier and cheaper. It'd also be a good excuse to upgrade the router head to a 3hp unit. It's currently a 1.75 porter cable but I'd like a little more oomph for when I'm doing larger-scale router ops.
 
Hi Earl,

I'm not sure if this is sarcasm or what.

Not sarcasm, just curiosity. I do not understand how it could be possible to establish the Thiele-Small parameter values for a given driver without physically handling the device. Paranormal activity is the only explanation that I can come up with.

It would even work for compression drivers if measured on a plane wave tube.

I think we just touched the device.... :).

Kind regards,

David
 
Use JohnK's, catapult's or my spreadsheet.

Hi JoshK,

Could you please post a link to your latest spreadsheet. I currently have a copy of OS WG 2.xls, and when I increase the value of the initial exit angle, the radius of the waveguide throat decreases - it does not stay at the specified value. (I have changed the driver exit angle radians formula from =6*PI()/180 to =C10*PI()/180).

Thanks in anticipation.

Kind regards,

David
 
Hi JoshK,

Could you please post a link to your latest spreadsheet. I currently have a copy of OS WG 2.xls, and when I increase the value of the initial exit angle, the radius of the waveguide throat decreases - it does not stay at the specified value. (I have changed the driver exit angle radians formula from =6*PI()/180 to =C10*PI()/180).

Thanks in anticipation.

Kind regards,

David

Here is the link to mine: http://www.musicanddesign.com/codes/OS_wave_guide.zip
 
After going through months of effort and lots of resin, cloth, foam, wood, plastic, sandpaper, and filler putties, I can confirm- one-offs are exceedingly obnoxious to do in synthetic materials. My production method wasn't ideal either, however. If I had it to do over, I'd do it in wood rings. The results I'm getting are exceedingly nice, but I'm nearing $150 in materials and a minimum of 40 "pure" work hours (no downtime, that's all spraying, spreading, sanding, cutting time). One can go quick and dirty, I got usable results from a foam mold and a plaster cloth layup, but

After all these months, I'm being a little anal about the result- I want nothing less than complete perfection, as I'm so vested in this.

Next ones I make with wood and a router. In retrospect, that would have been much MUCH easier and cheaper. It'd also be a good excuse to upgrade the router head to a 3hp unit. It's currently a 1.75 porter cable but I'd like a little more oomph for when I'm doing larger-scale router ops.

Do you have any molding that are decent that you would sell?
 
I thought it would be fun to look at some nice polars of an OSWG.
 

Attachments

  • nice oswg polar.gif
    nice oswg polar.gif
    41.4 KB · Views: 386
Hi soongsc,

Would it be possible to post the dimensions of the waveguide?

* Throat radius (or diameter or area)
* Throat entry half-angle (if not 0 degrees)
* Coverage half-angle
* Axial length from throat to mouth

Thanks.

Kind regards,

David
* Throat radius (13mm)
* Throat entry half-angle ( 0 degrees)
* Coverage half-angle (~31.5)
* Axial length from throat to mouth (300mm)

I just posted the best looking parts.:)
 
Hi Soongsc,
would you post a sim for 120 degree oswg with 1.4" throat?
Thanks...
Work is up to my nose right now trying to solve issues in many fronts. Plus results are driver dependent. AxiDriver is free if you don't need to save the sim data, and quite easy to learn. Why not give it a try?

For my application, I could not get good results much beyond 60deg. But others may consider different tradeoffs. The one that I posted was done just to investigate what kind of limits I could expect.
 
Hello Michael and Soongsc,

A lot of good things abour shape optimization can be seem in Morgans 's thesis.

You can download the different parts at:

Adelaide Research and Scholarship: Optimisation techniques for horn loaded loudspeakers.

I recommand specially the reading of chapter 5 to 8.

http://digital.library.adelaide.edu.au/dspace/bitstream/2440/41350/2/03chapters5-8.pdf


Few simulated shape are looking similar to yours...

Best regards from Paris, France

Jean-Michel Le Cléac'h

... The differences between my Gaussian MinPhase contour the LeCleach contour and oblade spheroide contour are quite distinct with respect to tolerating a "conical duct" in front of the horn contour....
 
For my application, I could not get good results much beyond 60deg.

It can be shown that the HOM content of the solution to the OS waveguide increases dramatically after about 45 degree 1/2 angle. Basically, what this says is that diffraction begins toi dominate the solution for these wider angles and the device is working less like a waveguide/horn and more like a contoured baffle. This is the reason why I do not really consider a waveguide beyond 45 dgrees, or consider any device of these larger angles to actually be a waveguide - although technically it is. At these larger angles one can no longer assume a single mode of wave propagation at any frequency.