3D printed waveguides

[andy19191]

Let my subscription lapse while I'm travelling for work, but I renewed it just now. Site should be back in 48 hours. The SB26 versions are the most popular.

Thanks. I will download it when the site comes back up.

Why are you developing your own software? Does something like ABEC not work well for you?

It is possibly a semi-retirement project otherwise a hobby interest. Depends how it goes. The BEM work is somewhat to the side of the main project but it has been on the to-do list for 20 years over which time I have made a couple of starts. The last long weekend a few months ago produced a very rough but apparently working hack using higher order quad elements (i.e. curved geometry and fewer required elements per wavelength for accuracy). It is this that I intend to tidy up soon, run some basic testing and then demonstrate with a few examples like one of your waveguides. Later developments which are not imminent would be to add a fast mulitipole method to increase efficiency for room simulations and later to adjust the geometry parameters to best fit a target directivity (hence the interest in the parameters you use) much of the software for which will likely arise anyway from some of the work required for the main project.

I have never used ABEC but intend to give it a quick go at some point. I don't use windows, ABEC has crippled IO in the free version and I cannot afford the full version making it of limited interest. I have tried elmer which proved to be broken and judged not worth fixing. I have also tried acousto which worked OK but was clumsy and used zero order elements (as does ABEC) requiring fine grids for accuracy which hurts methods like BEM which use full matrices.

[edlafontaine]

If this is not related to the thread...
For a flush-mount speaker design, what changes to the crossover would you consider making to correspond to placing the tweeter in a waveguide?

[ernperkins]

Three areas of change popped into my head. Of course they all depend on the depth and width of the waveguide.

1. It changes (reduces) the acoustic offset by moving the tweeter back. This will change the phase relationship between the drivers, especially at the crossover point. Many designs use an asymmetrical crossover (e.g. 2^nd order electrical on woofer and 3^rd order electrical on tweeter) to compensate for the acoustic offset. Depending on the waveguide depth the tweeter crossover might need to be changed to 2^nd order electrical since the acoustic offset has been reduced.
2. Depending on the waveguide (mainly width) you’ll see a boost in the tweeter’s lower response. That’s a good effect since the tweeter’s output can be reduced more (lowering distortion) to match the other drivers. But it will require crossover re-work. A potential benefit (again depending on the waveguide) is that you may be able to crossover at a lower frequency.
3. The waveguide will add controlled directivity which effects power response. If you use averaged listening window measurements you’ll see the change and may want to account for it.

In a nutshell: I would take all new tweeter measurements and use those with the current crossover to see the magnitude of the change.

[edlafontaine]

The CSS LD25X
fwiw

The surround has an irregular profile. The surround of this tweeter is half occluded by the waveguide. That portion which is covered by the guide is curved. That portion which is not covered by the guide is "flat" to the dome.
This tweeter was plagued by quality and consistency issues and was not a commercial success. Among the select provided for testing it has what was characterized as "world class" low distortion.

[edlafontaine]

The surface of the faceplate which corresponds to the surround is also curved. The air gap between the 2 increases from the adhered edge toward the dome.

[PWR RYD]

Hi Ed. Are you gonna be able to make it up to InDIYana this year?

[edlafontaine]

Hi Ed. Are you gonna be able to make it up to InDIYana this year?

Hi, I am planning to. It has been awhile.

[Wolf]

I'll add you to my list, Ed. It;s about 7 weeks out...
Wolf

[augerpro]

Goodies starting arrive:




6.5" wide and 1.2" inch deep. I love this SLS nylon. Defect free finishes, if a bit sandpapery. Super tough. Also just bought and Ender 3 Pro, so I'm prototyping improvements to the phase shield so printing errors are reduced. Probably should have had this 4 years ago!

[ani_101]

Hey Brandon, where or how do we get the waveguides? I was following the thread when you started, but then lost touch. Do you have or can point me where I can get the latest information.

[augerpro]

My Soma Sonus website in my sig. Those will only be finalized versions, so new ones like this Bliesma T25 waveguide won't be on there yet.

[ani_101]

Thanks Brandon, got your website. Can i ask where are you getting the Nylon waveguides printed form? and how different are they from the ones you are printing on the ender. the 3D printer has been on my list for a long time - but not sure if it would be of acceptable print quality

[augerpro]

DanP at CSS has been VERY generous in printing these for me. If it wasn't for him there is no way I could do this. Now these Bliesma waveguides were actually paid for by the person who loaned me the tweeter, so he could expedite his project. I used 3dhubs to source a vendor.

[saabracer23]

I’m thinking I might pick up a resin printer for this exact purpose. The resin prints ive seen are very very smooth. Any reason this wouldn’t work?

Dan

[ani_101]

Not Sure, Dan at CSS has got access to some heavy duty industrial 3D printers, which are in a different orbit quality wise.... The hobby grade resin printers are very close to the FDM printers, though each have their own strengths. For an entry level SLA printer you can get a top of the line big volume FDM printer.

But in case you do find a good printer SLA or FDM, please let us know, would be interested in knowing.