Was thinking about this... I think i'll stay on with the balls for now, but I've got access to a Tevo Little Monster 3D printer (very large) which will allow me to print horns/cones/reflectors up to 13 inch diameter.
I could do a hybrid perhaps, using a ball for the woofer and print something more elaborate for the compression driver.
Yes, I was looking at those garden ornaments! Also, I'm wondering if a simple straight sided cone would be better than a sphere... Have you seen this site? Omnidirectional speakers
Im not sure about the straight sided cone, but i guess it wouldnt be hard to construct one and test.
Yea I've had a look at that site. Some of the reflectors are quirky.
just thinking as well (a bit off topic of omni's) - with the whole FAST/WAW speaker thing, I wonder how effectively you could get a woofer assisted compression (WAC?) to work. I've seen some of the dayton 10" subwoofers have ok response up to 1-1.5KHz, and presumably you could get a nicer compression driver that would cover 1.5Khz up to 18 or 20Khz. It would be asking a lot of both drivers though.
Yea I've had a look at that site. Some of the reflectors are quirky.
just thinking as well (a bit off topic of omni's) - with the whole FAST/WAW speaker thing, I wonder how effectively you could get a woofer assisted compression (WAC?) to work. I've seen some of the dayton 10" subwoofers have ok response up to 1-1.5KHz, and presumably you could get a nicer compression driver that would cover 1.5Khz up to 18 or 20Khz. It would be asking a lot of both drivers though.
UPDATE! - Sorry for the long delay
Hi All,
Sorry the stuff actually arrived a few weeks ago but totally snowed under with work so wasnt able to get anything.
Everything arrived more or less as expected. The compression driver had a tiny bit of damage but minor.
The spheres were of surprisingly good quality for the price.
away on a business trip for a week soon but will get back into this once ive got the time.
Cheers!
Hi All,
Sorry the stuff actually arrived a few weeks ago but totally snowed under with work so wasnt able to get anything.
Everything arrived more or less as expected. The compression driver had a tiny bit of damage but minor.
The spheres were of surprisingly good quality for the price.
away on a business trip for a week soon but will get back into this once ive got the time.
Cheers!
Attachments
New year, back at work after the break and finally put some time in on the CNC.
Just made a stand basically to hold all the cones (and woofer) at equal height during testing. The top ring pops out and can be swapped for the other horns etc as seen in the photo.
Will set up the software on the raspberry pi and DSP in the next few days. Hopefully will be able to start measuring response this week or next.
Just made a stand basically to hold all the cones (and woofer) at equal height during testing. The top ring pops out and can be swapped for the other horns etc as seen in the photo.
Will set up the software on the raspberry pi and DSP in the next few days. Hopefully will be able to start measuring response this week or next.
Attachments
Did some messing around last week,
Got the amp going. The Beocreate software is a bit clunky and it took a few tries to get the image on the raspberry working, but seems to be ok for now.
Installed SigmaStudio and had a play with that - all working and the real time updates over ethernet are quite handy.
Got REW running - still getting the hang of it but can take basic horn measurements now.
I also went out and bought a Behringer UCA222 interface (the cheapest i could find) in order to have an optical audio output from my laptop, as the Beocreate has no analog in, and the only other options were bluetooth (buggy) or airplay. The optical interface is fine though.
To do:
- properly measure the compression driver and metal ball/reflector combinations
- Build a small enclosure for the woofer for tests. Will do something out of ply on the CNC
- Measure thiele and small of woofer
Also on another note, the lack of volume control on the optical audio output is a bit of an issue if this is to be used for TV/home theater applications. Still trying to figure out a work around but it appears as there is a way the raspberry can control the DSP volume control in real time although not currently implemented
Got the amp going. The Beocreate software is a bit clunky and it took a few tries to get the image on the raspberry working, but seems to be ok for now.
Installed SigmaStudio and had a play with that - all working and the real time updates over ethernet are quite handy.
Got REW running - still getting the hang of it but can take basic horn measurements now.
I also went out and bought a Behringer UCA222 interface (the cheapest i could find) in order to have an optical audio output from my laptop, as the Beocreate has no analog in, and the only other options were bluetooth (buggy) or airplay. The optical interface is fine though.
To do:
- properly measure the compression driver and metal ball/reflector combinations
- Build a small enclosure for the woofer for tests. Will do something out of ply on the CNC
- Measure thiele and small of woofer
Also on another note, the lack of volume control on the optical audio output is a bit of an issue if this is to be used for TV/home theater applications. Still trying to figure out a work around but it appears as there is a way the raspberry can control the DSP volume control in real time although not currently implemented
Attachments
Yep looked into the ALSA volume control, should work. Leaving it for now as the optical full volume is fine for testing.
Did more horn tests.
Firstly did a basic 1m on axis sweep, results below. was done at slightly less than 1w into the CD for reference. The Selenium/JBL horn seemed best, and the 8 inch Dayton horn seemed ok as well.
Didnt include the plots, but the JBL horn was a lot more directional and had much more falloff on high freqs when off axis. the 8 inch Dayton was ok.
Did some very rough ball tests - photo of the setup. Didn't really know how to do it, but the setup shown roughly allowed to simulate a reflective back wall and ceiling, and gave me some flexibility with mic position.
I was just playing around, and tried most sphere size and position tests, you can see the results vary a lot but there is an obvious trend.
Worth noting in most of the cases the effect of the sphere was minimal, and a similar response was measured when the sphere was removed, and the mic position remained 90 degrees from the direction of the horn.
The only worthwhile difference was when i had a 300mm sphere offset from the center of the horn, so that it was much more directional - that is the highlighted measurement in the images.
Still no woofer tests, need to do the enclosure.
Perhaps an angled or curved plate of steel/aluminum would be a better reflector. Not quite omni, but still.
Welcome to any thoughts or suggestions
Did more horn tests.
Firstly did a basic 1m on axis sweep, results below. was done at slightly less than 1w into the CD for reference. The Selenium/JBL horn seemed best, and the 8 inch Dayton horn seemed ok as well.
Didnt include the plots, but the JBL horn was a lot more directional and had much more falloff on high freqs when off axis. the 8 inch Dayton was ok.
Did some very rough ball tests - photo of the setup. Didn't really know how to do it, but the setup shown roughly allowed to simulate a reflective back wall and ceiling, and gave me some flexibility with mic position.
I was just playing around, and tried most sphere size and position tests, you can see the results vary a lot but there is an obvious trend.
Worth noting in most of the cases the effect of the sphere was minimal, and a similar response was measured when the sphere was removed, and the mic position remained 90 degrees from the direction of the horn.
The only worthwhile difference was when i had a 300mm sphere offset from the center of the horn, so that it was much more directional - that is the highlighted measurement in the images.
Still no woofer tests, need to do the enclosure.
Perhaps an angled or curved plate of steel/aluminum would be a better reflector. Not quite omni, but still.
Welcome to any thoughts or suggestions
Attachments
Hi. Interesting. I tend to agree, even a straight sided cone would be better than a sphere. The more directional high frequencies, will tend to be reflected back down, others will curve around a sphere and be reflected off the ceiling. You have to ask yourself, how many omnidirectional speakers out there use a sphere as the reflector?
Yea i did a few measurements where the mic was about 45 degrees to the horn and i agree, it seems as though the high frequencies appear to curve around the sphere.
Will be interesting to see how the woofer/sphere combo behaves, but definitely seems less than ideal for the horn/CD.
The Duevel Enterprise speakers look like they have a more suitable reflector for the HF
Will be interesting to see how the woofer/sphere combo behaves, but definitely seems less than ideal for the horn/CD.
The Duevel Enterprise speakers look like they have a more suitable reflector for the HF
1950's GA Briggs design for wideband 8"
https://www.audioxpress.com/assets/upload/files/weems_drainpipe8.pdf
https://www.audioxpress.com/assets/upload/files/weems_drainpipe8.pdf
Maybe a different planet is required
. I was thinking about how you could try a few ideas with those spheres. All of them ruined the aesthetics of course. The horn is 25cm ID, the sphere is 14cm OD, the disc is 23cm, and the cone is 5cm ODx3cm. Compare them to just using sphere in post #21. The field at the top of the sphere is a little jagged from simulation resolution used.Attachments
Last edited:
It does. The problem is that it's hard to change direction unless you can provide both an inner and outer surface to support that direction change.
The sound will bounce off the walls as you can light up the room and it can even out by diffusion. You also need enough room for spacing and delays.
The sound will bounce off the walls as you can light up the room and it can even out by diffusion. You also need enough room for spacing and delays.