The ring started with the SB26, which has leads glued to the dome plate, and so the waveguide wouldn't seal to the top of the tweeter, which caused an issue in the response. I'd seen this before in a waveguide I made years ago, so just decided to make it a feature everywhere (except Scanspeaks which have a completely different form). That said, I'm changing how it works for the SB26 and removing it from all the other waveguides.
That DI looks pretty fantastic! What is the crossover?
Hi Brandon,
Yes, not bad 🙂. You can see the overall look of my active loudspeaker (post 714)
It's an acoustic LR4 around 2200 Hz as you can guess on the phase response. I'm using a FusionAmp 252.
The polar response was measured with mic at 84 cm from the front plate in tweeter axis with Arta untouched (well still 1/24 default smoothing) with a 5 msec windows then converted in VituixCAD just for presentation.
Nice. Did your try lower? I haven't actually built a speaker with any of these waveguides, so I'm not sure what the "real world" cutoff is for the waveguide, given, say, an LR4 crossover.
Well, you always have to think physics !
We have a small dome with a Fs at 1050 - 1150 Hz, even with a very low distortion unit around Fs like the Bliesma T25B and the help of the waveguide gain, I would not recommend to go much lower than 2 kHz. Maybe with high slope FIR ?
I'm also impressed by the purifi unit which is the easiest driver to correct I had to deal with for the past 10 years. I was reluctant to select it at first due to the low efficiency but it performs very well too (I use 2 passiv drivers on the sides)
We have a small dome with a Fs at 1050 - 1150 Hz, even with a very low distortion unit around Fs like the Bliesma T25B and the help of the waveguide gain, I would not recommend to go much lower than 2 kHz. Maybe with high slope FIR ?
I'm also impressed by the purifi unit which is the easiest driver to correct I had to deal with for the past 10 years. I was reluctant to select it at first due to the low efficiency but it performs very well too (I use 2 passiv drivers on the sides)
You rang?
Pretty well any decent FIR HPF will have essentially no output at an octave below the corner frequency so you don't need really high slopes. In the time domain my personal bias tends away from brickwall filters [1] and toward what I call a "cowboy hat" impulse response: an initial dip below zero, then the obligatory central peak, followed by another dip below zero. These filters are computationally nearly the least demanding FIR possible and they feature about an octave overlap between drivers so the transition will sound more natural.
[1] One gig wanted a very high slope FIR lowpass: at 10 kHz the transition band from full gain to -90 dB was specified as 1 kHz wide. It sounded awful, like a hatpin scratching directly on my eardrum. I asked the chief engineer if I could redesign the thing, and changing the transition bandwidth to 2 kHz made it sound considerably less horrible.
Anyway, we can continue this discussion in another thread. Right now the waveguide saga is massively interesting. Oh, and an advert: 3D print material isn't free so y'all please donate to Augerpro's gofundme so he can continue this excellent work. I did 😀
Thanks for the plug! Everyone has been very generous with donations, I've met my first goal. Thanks everyone who donated!
What's the f3 and f10 of you're design? It looks amazing.
And do I see a infrared slot in the front? If so, that's awesome.
DSP_Geek,
Thanks for your input, and yes brickwall filter are not the solution. My inclination for the moment is to mimic analog filters via IIR and correcting eventually the phase by FIR.
Arnandsway,
I do not want to pollute Augerpro waveguide thread with my latest design so PM me. Just to answer shortly : f3 45,6 Hz / f10 35,6 Hz / 10 liters internal. and Hypex FusionAmp can be IR controlled.
Brandon,
I think you took a wise decision to remove the small throat ring when possible
Thanks for your input, and yes brickwall filter are not the solution. My inclination for the moment is to mimic analog filters via IIR and correcting eventually the phase by FIR.
Arnandsway,
I do not want to pollute Augerpro waveguide thread with my latest design so PM me. Just to answer shortly : f3 45,6 Hz / f10 35,6 Hz / 10 liters internal. and Hypex FusionAmp can be IR controlled.
Brandon,
I think you took a wise decision to remove the small throat ring when possible
Crazy question for augerpro - what are your thoughts on a mini Unity Horn with something similar to these elliptical waveguides using 2 or 4 TC5FC00 mounted such that the cones are as "flush" as possible with the waveguide walls? Any discontinuities or sharp edges might be taken care of using judicious application of felt. Application would be tweeter crossed at around 1.5 kHz or so, with the mini-mid drivers going down to maybe 600-800 Hz to cross to a 6" driver.
It's one of the few drivers I can think of that are almost as small as a typical tweeter, but still with excellent performance into the sub-1 kHz regime (depending on output level, hence the proposed use of more than one).
It's one of the few drivers I can think of that are almost as small as a typical tweeter, but still with excellent performance into the sub-1 kHz regime (depending on output level, hence the proposed use of more than one).
Last edited:
I'd love to do a Unity/Synergy horn at some point. 3D printing certainly lends itself to that. But I know zero about them.
Bonjour JC,
Have the DI plots in post 719 been generated by the full "Kimmoste Vcad" procedure with 19 (5 degree) measurements horizontal and vertical?
Have the DI plots in post 719 been generated by the full "Kimmoste Vcad" procedure with 19 (5 degree) measurements horizontal and vertical?
No, preliminary results, horizontal only at 10° step I'm still trying to refine the filters. Full set will be made later. Don't expect vertical polar plot that good, we have non coincidence of drivers (145 mm)
What kind of printing settings/materials are recommended?
SLS or FDM? Nylon or Petg? How much % infill (20,40,60,80)? Using the platform 3Dhubs.com there are so many possibilities, it's kinda overwhelming for a layman.
SLS or FDM? Nylon or Petg? How much % infill (20,40,60,80)? Using the platform 3Dhubs.com there are so many possibilities, it's kinda overwhelming for a layman.
I frankly don't know much about 3D printing. Maybe an expert lurking about can chime in. Here are some beliefs I have though...SLS nylon process which uses a laser to sinter powdered nylon is tough as hell. I love it. I think it is 100% infill. Only available from commercial outfits though. But I've heard nylon likes to absorb water, possibly causing warpage over time? So it would be great to find a sealant I could apply.
PLA is super easy to print with, but isn't terribly chemically stable, so breaks down over time. Probably causing some warping here too? Not very tough. I would suspect high infills would be desirable to lessen sound transmission but have no real data to back that up.
Petg is tough and AFAIK chemically stable and would be my pic for a hobbyist for a final speaker build. But it is tougher to print with, you need to have the printer settings tuned for your machine and filament. Infill would be similar to PLA I would think.
PLA is super easy to print with, but isn't terribly chemically stable, so breaks down over time. Probably causing some warping here too? Not very tough. I would suspect high infills would be desirable to lessen sound transmission but have no real data to back that up.
Petg is tough and AFAIK chemically stable and would be my pic for a hobbyist for a final speaker build. But it is tougher to print with, you need to have the printer settings tuned for your machine and filament. Infill would be similar to PLA I would think.
AIUI nylon is a polyamide, and i.materialise offers waterproofing process for this material at a slight premium. An upload of 6x1ASB26v1.zip generated a single piece quote of $85.34 for straight polyamide, with the waterproofed version at $124.79, and a pair goes for $168.26 and $217.56 respectively. My 3D printing experience is negligible so I'm far from an expert, however a friend is quite happy selling her sculptures through i.mat., including this piece in polyamide.
Haven't the faintest. She's an artist, not a 3D specialist 😀
One presumes it must be reasonably effective, given i.materialise's decent reputation.
Last edited:
Hi,
First, thanks Brandon for all the work done.
Regarding waveguide, I am more in the understanding phase than a build phase. So I have basic questions if you can help me (sorry the information have already been provided).
Firstly, do you have some interesting technical documents about waveguide design to recommend ?
Secondly, I obviously see the different responses when the size of the waveguide increases (especially with the newly added 8 inch design), but what do you take into account for the choose of your waveguide size on a project (aside from geometric constrains) ? Point of crossover ? Directivity match between driver ? Sensitivity ?
Thanks !
First, thanks Brandon for all the work done.
Regarding waveguide, I am more in the understanding phase than a build phase. So I have basic questions if you can help me (sorry the information have already been provided).
Firstly, do you have some interesting technical documents about waveguide design to recommend ?
Secondly, I obviously see the different responses when the size of the waveguide increases (especially with the newly added 8 inch design), but what do you take into account for the choose of your waveguide size on a project (aside from geometric constrains) ? Point of crossover ? Directivity match between driver ? Sensitivity ?
Thanks !