I've settled on Eminence Beta 12CX and PRV D290py-S compression drivers in a JE Labs style open baffle with a MiniDSP crossover. My design goals are to create a mid-range focused system for low to low-moderate listening (paper thin walls in the new apartment). So far I've only built mono systems so I'm excited to explore soundstage and imaging! I have headphones for bass heavy music, but I'm interested in exploring more jazz, soul, world and opera with this system. Note: I cannot hear past 16000hz.
I'm constructing a rig to measure this speaker 6' off the ground in my back yard. Past this iteration I'm interested in exploring felt baffles with large round over edges, but for the moment I'm content learning the ins and outs of DSP.
I just ordered two 2x15w boards from classdaudio.com to power the 12CXs/tweeters, but I also have an inuke1000 at my disposal if I need more power for the woofers. Question: I like dynamic systems, if I use a more powerful amp to power the woofers and EQ nulls will I loose dynamics? The woofers are 97db with 250watt RMS power handling. The inuke puts out about 160 watts into 8 ohms. Like I said, I'm not trying to get super loud with lean bass. Any tips for designing a DSP based system for quiet listening?
I'm constructing a rig to measure this speaker 6' off the ground in my back yard. Past this iteration I'm interested in exploring felt baffles with large round over edges, but for the moment I'm content learning the ins and outs of DSP.
I just ordered two 2x15w boards from classdaudio.com to power the 12CXs/tweeters, but I also have an inuke1000 at my disposal if I need more power for the woofers. Question: I like dynamic systems, if I use a more powerful amp to power the woofers and EQ nulls will I loose dynamics? The woofers are 97db with 250watt RMS power handling. The inuke puts out about 160 watts into 8 ohms. Like I said, I'm not trying to get super loud with lean bass. Any tips for designing a DSP based system for quiet listening?
Last edited:
It depends how you'll use eq* but i doubt dynamic will be missing even open baffle.
What kind of measurements do you plan?
There is inspirational thread about system like that atm. Pano started one recently and Perrymarshall presented some too:
https://www.diyaudio.com/community/threads/18-coax-on-open-baffle-bm18cx-38.402195/
https://www.diyaudio.com/community/threads/open-baffle-bass-reflex-hybrid.393837/
*Eq nulls: if not SBIR related you can try to boost the depresed area by a bell shape with Q adapted of course. But it's not diferent than using two shelves ( high and low) attenuating outside the area where the dip happen and afterward you can compensate by adjusting overall level driving the amp... try both way and compare them, with dsp it's easy. Choose the one you prefer by comparison.
What kind of measurements do you plan?
There is inspirational thread about system like that atm. Pano started one recently and Perrymarshall presented some too:
https://www.diyaudio.com/community/threads/18-coax-on-open-baffle-bm18cx-38.402195/
https://www.diyaudio.com/community/threads/open-baffle-bass-reflex-hybrid.393837/
*Eq nulls: if not SBIR related you can try to boost the depresed area by a bell shape with Q adapted of course. But it's not diferent than using two shelves ( high and low) attenuating outside the area where the dip happen and afterward you can compensate by adjusting overall level driving the amp... try both way and compare them, with dsp it's easy. Choose the one you prefer by comparison.
Last edited:
I plan on taking driver measurements in free air using an umik, designing the crossover and eq from there and then making adjustments in room once I move in in a few months. I need to do more reading so I’m not entirely sure where/how far to place the mic. Do I need off axis readings?
Yes the more measurements the better, on axis, off axis ( by 10° steps), step response,...
Umik. Usb? I fear it might make time alignement measurements an issue and you might need them.
Umik. Usb? I fear it might make time alignement measurements an issue and you might need them.
Yes USB, I haven’t heard of any issues with time alignment. Is there a latency issue? What sort of budget options are alternatives?
What design software are you using? Two-channel sound card is needed to get equivalent phase info.
I use umik-1 and use only spl measured with REW. First I make individual drivers to give flat response at least 1 octave beyond the intended xo. Then set xo LP and HP, eg. LR4. Then let both drivers play and measure spl. Then adjust tweeter (cd) delay to get flat response. Then reverse polarity of tweeter and check cancellation dip.
You can also use step response to analyze timing/phase match but this requires more experience.
I use umik-1 and use only spl measured with REW. First I make individual drivers to give flat response at least 1 octave beyond the intended xo. Then set xo LP and HP, eg. LR4. Then let both drivers play and measure spl. Then adjust tweeter (cd) delay to get flat response. Then reverse polarity of tweeter and check cancellation dip.
You can also use step response to analyze timing/phase match but this requires more experience.
I use 2 chanels technique and had no experience with usb mic for a while. As i don't know the setup capability i can't help you more about it but with my own attempt on xover for coax i think it's worth to have the info about offset ( in fact i only recently had a coax driver where both drivers are on same Z offset and i played with some of them for some years now).
When using dsp why not go for compensation it allow easily? If it mater to you in outcome you'll notice and keep the treatment if not you switch it off.
There is a way which might be usable to find acoustic offsett but they include additional steps with simulation software though. Have you ever tried Juhazi?
https://app.box.com/s/ouxjjsx0m8bs00cil5iq
http://www.troelsgravesen.dk/Time-Alignment.htm
https://www.speakerdesign.net/sbarticle.html
Another inspirational design from Perrymarshall described there:
Post in thread 'Ultimate Open Baffle Gallery' https://www.diyaudio.com/community/threads/ultimate-open-baffle-gallery.123512/post-6682515
When using dsp why not go for compensation it allow easily? If it mater to you in outcome you'll notice and keep the treatment if not you switch it off.
There is a way which might be usable to find acoustic offsett but they include additional steps with simulation software though. Have you ever tried Juhazi?
https://app.box.com/s/ouxjjsx0m8bs00cil5iq
http://www.troelsgravesen.dk/Time-Alignment.htm
https://www.speakerdesign.net/sbarticle.html
Another inspirational design from Perrymarshall described there:
Post in thread 'Ultimate Open Baffle Gallery' https://www.diyaudio.com/community/threads/ultimate-open-baffle-gallery.123512/post-6682515
Last edited:
My first reference for this project is a little guide from the klipsch forums by Chris Askew. I'll try to tack it on here as an attachment. He uses a USB UMIK like mine and spectrographs to time align.
@krivium
Bonjour! Thank you for bringing the latency issue to my attention. I don't have the budget to invest in more equipment—I'm stuck with what I've got. I have an UMIK and an M1 MacBook Air with REW and MiniDSP installed. On the other hand you passed me some interesting reading—merci beaucoup pour la littérature.
I found this thread with the advice from the creator of REW:
https://www.avnirvana.com/threads/p...rements-with-umik-1-different-usb-ports.6352/
It seems there is a setting to "Adjust clock with acoustic ref" and using shorter sweeps will assuage the problem further. What do you guys think?
@Juhazi
Do you have any links to documentation to the two mic process? I'm just curious.
That Perrymarshall speaker is definitely inspiring. I'd love to widen the sweet spot but I'll stick to getting a flat response and getting the time alignment right for now. Over the long weekend I'm going to cut some wood.
@krivium
Bonjour! Thank you for bringing the latency issue to my attention. I don't have the budget to invest in more equipment—I'm stuck with what I've got. I have an UMIK and an M1 MacBook Air with REW and MiniDSP installed. On the other hand you passed me some interesting reading—merci beaucoup pour la littérature.
I found this thread with the advice from the creator of REW:
https://www.avnirvana.com/threads/p...rements-with-umik-1-different-usb-ports.6352/
It seems there is a setting to "Adjust clock with acoustic ref" and using shorter sweeps will assuage the problem further. What do you guys think?
@Juhazi
Do you have any links to documentation to the two mic process? I'm just curious.
That Perrymarshall speaker is definitely inspiring. I'd love to widen the sweet spot but I'll stick to getting a flat response and getting the time alignment right for now. Over the long weekend I'm going to cut some wood.
The beauty of active DSP is it will allow you to dial in your XO in your location/spot once you settle in and then you can do it old school…..voice the system to what you hear vs measurements alone. You can get a baseline before hand on axis at 1 meter with your mic……this will allow you to see how the compression driver reacts to the woofer as it’s waveguide…..there will be some work to be done there.
Look fwd to your outcome.
Look fwd to your outcome.
Just glancing at that doc I see how less could go wrong. If I get a cheap audio interface with a cheap measurement mic would I be good for time alignment?
For instance:
M-Audio
Behringer
Also, I have cicada insect noise where I live. Is it even worth taking near-field measurements outdoors? I mean if it interferes I could gate the measurements, right?
For instance:
M-Audio
Behringer
Also, I have cicada insect noise where I live. Is it even worth taking near-field measurements outdoors? I mean if it interferes I could gate the measurements, right?
Last edited:
This is what is needed. Behringer or Dayton mic are ok. As long as you have the calibration file that's ok.
2 chanel interface. Done. And you've got a second mike which can be handy.
If outside is noisy it can be an issue. That said there is always windows of lower noise on a whole day. Otherwise indoor gated and you merge with nearfield response ( for low mid/ low) .
2 chanel interface. Done. And you've got a second mike which can be handy.
If outside is noisy it can be an issue. That said there is always windows of lower noise on a whole day. Otherwise indoor gated and you merge with nearfield response ( for low mid/ low) .
Last edited:
And your really just concerned with 500hz and up for time alignment and tweeter response through the XO range
These are the boards I got from classDaudio.com. They are 2x15w bridgeable to 30w mono and I'm using 24v 3a power supplies for LEDs. I'm excited to use them. Like I said earlier my hearing doesn't extend beyond 16000hz so I'm not even sure if the more boutique amps out there would really be effective for me. Maybe I'll plug in my IcePower amps from my last project and see if there's a difference audible to me. Anyone is welcome to chime in on that debate!
Well, I managed to rip some boards this weekend. Not as much work done as I'd like—it just got too hot to work outdoors. I'm using a Kregg jig to make the rip cuts since I don't have a table saw available at the moment.
Here's the last speaker I built. A mono system. It probably doesn't sound as great as it could without some more work done in the DSP. If I could make one change it would have been to use a constant directivity waveguide on the tweeter instead of a horn, but it's been well received. My friend comes over and we spin records until late on it. It was my pandemic project and it does vocals like magic.
My goal as a DIY audiophile is never to have the speaker I want voted out on aesthetic grounds. I really dig the JE Labs Open Baffle plan from the Japanese magazine. I reminds me of Donald Judd's sculptures for its economy. For the last speaker I looked at the old Apple iPod Mini as inspiration and Dieter Rams era Braun products that inspired it in turn.
I was thinking about wrapping the baffle in thick wool felt with a large radius round edge.... but the cost of good felt is prohibitive. In looking forward to a new project in stereo with a more modest budget I'm really thinking a lot about these Judd chairs as I prototype in softwood:
Here's the last speaker I built. A mono system. It probably doesn't sound as great as it could without some more work done in the DSP. If I could make one change it would have been to use a constant directivity waveguide on the tweeter instead of a horn, but it's been well received. My friend comes over and we spin records until late on it. It was my pandemic project and it does vocals like magic.
My goal as a DIY audiophile is never to have the speaker I want voted out on aesthetic grounds. I really dig the JE Labs Open Baffle plan from the Japanese magazine. I reminds me of Donald Judd's sculptures for its economy. For the last speaker I looked at the old Apple iPod Mini as inspiration and Dieter Rams era Braun products that inspired it in turn.
I was thinking about wrapping the baffle in thick wool felt with a large radius round edge.... but the cost of good felt is prohibitive. In looking forward to a new project in stereo with a more modest budget I'm really thinking a lot about these Judd chairs as I prototype in softwood:
Gave it some thought over the week and decided to go with thinner baffles. Saves space, improves imaging and I'll be dialing in the right amount of bass through DSP anyway. The Perry Marshall write-ups gave me much food for thought. I've decided to go for two 12" Bianco OB woofers under the coax for the following reasons:
Only have enough clamps to join one baffle at a time.
Never start without a drawing.
I should have access to a proper shop come December. I'm considering everything I do at this point a 1:1 scale model which I'll refine a second time over. Still looking for inspiration in the furniture of Donald Judd.
I intend to stick with pine all around with flush mounted drivers. Like a long version of the above stool stood on its side. Here is a drawing of how I intend to brace the baffle with 2x2s and chamfer the edges.
I'm thinking a 1" chamfer at least on the sides and perhaps oblique angles on the top. What do you guys think—which chamfer is worth the time and effort?
Thanks for all the input so far, I'm totally amped to start taking measurements!
- I can get bass (I need at least a little).
- Extends the height of the baffle to ear height.
- Most importantly, this should clear up the mids if I regulate the coax woofer to midrange duty and let the woofers do their thing.
Only have enough clamps to join one baffle at a time.
Never start without a drawing.
I should have access to a proper shop come December. I'm considering everything I do at this point a 1:1 scale model which I'll refine a second time over. Still looking for inspiration in the furniture of Donald Judd.
I intend to stick with pine all around with flush mounted drivers. Like a long version of the above stool stood on its side. Here is a drawing of how I intend to brace the baffle with 2x2s and chamfer the edges.
I'm thinking a 1" chamfer at least on the sides and perhaps oblique angles on the top. What do you guys think—which chamfer is worth the time and effort?
Thanks for all the input so far, I'm totally amped to start taking measurements!
Hi,
About imaging, my experience with both wide and narrow baffle isn't as clear as what i heard from internet: narrow isn't better than wide, just different rendering.
Narrow is more pinpoint, wider a bit more 'diffuse'. From recordings and mixes i've done i can't say pinpoint is better: with purist stereo recording technique ( 2 mics) rendering is closer to wide baffle ( pinpoint location is rarely experienced ime - even with XY ( coincident capsule, relying on amplitude difference for stereo). Pinpoint rendering fit well multi close miced records though. It's a preference thing and not as black and white as people says over internet. And to say the truth the best i've heard is inwall mounted: no diffraction, better bass ( lower distortion as you experience the inverse of baffle step: you have to lower bass rather than increase it's level as bass radiates in 2pi rather than 4...),... best image rendering i've heard to date.
Maybe it's different for OB ( mine are BR and Closed box and inwall where BR obviously) but i doubt.
Chamfer: it's mainly related to diffraction so if you keep baffle width as close to driver diameter you don't really need it ( except for aesthetical reason). To gives you an idea and put things in perspective: E.Geddes used 2" radius for diffraction management of freq around 1khz for Summa. Shorter radius shift higher in freq the diffraction management...
Maybe it's worth prototyping with a narrow baffle and use removable 'add ons' chamfer to see if it is worth the effort.
One thing i would not do is to have same distance between driver/side and top.
I think you made a good choice to add the 2x12" Bianco. You won't run short on low end and it lowers the displacement needs over a single driver ( the more displacement = the more distortion and given you'll need a lot of eq for low end anyway...). I'm not sure it'll solve issue with bass+neighbours but you'll have a very nice OB loudspeakers set! And you can always lower listening level anyway. 😉
About imaging, my experience with both wide and narrow baffle isn't as clear as what i heard from internet: narrow isn't better than wide, just different rendering.
Narrow is more pinpoint, wider a bit more 'diffuse'. From recordings and mixes i've done i can't say pinpoint is better: with purist stereo recording technique ( 2 mics) rendering is closer to wide baffle ( pinpoint location is rarely experienced ime - even with XY ( coincident capsule, relying on amplitude difference for stereo). Pinpoint rendering fit well multi close miced records though. It's a preference thing and not as black and white as people says over internet. And to say the truth the best i've heard is inwall mounted: no diffraction, better bass ( lower distortion as you experience the inverse of baffle step: you have to lower bass rather than increase it's level as bass radiates in 2pi rather than 4...),... best image rendering i've heard to date.
Maybe it's different for OB ( mine are BR and Closed box and inwall where BR obviously) but i doubt.
Chamfer: it's mainly related to diffraction so if you keep baffle width as close to driver diameter you don't really need it ( except for aesthetical reason). To gives you an idea and put things in perspective: E.Geddes used 2" radius for diffraction management of freq around 1khz for Summa. Shorter radius shift higher in freq the diffraction management...
Maybe it's worth prototyping with a narrow baffle and use removable 'add ons' chamfer to see if it is worth the effort.
One thing i would not do is to have same distance between driver/side and top.
I think you made a good choice to add the 2x12" Bianco. You won't run short on low end and it lowers the displacement needs over a single driver ( the more displacement = the more distortion and given you'll need a lot of eq for low end anyway...). I'm not sure it'll solve issue with bass+neighbours but you'll have a very nice OB loudspeakers set! And you can always lower listening level anyway. 😉
Last edited:
I was also thinking that with a narrower baffle there would be less bass to start out with…
Could you explain further please? I assume you mean off setting the tweeter from center, but do you mean to vary the distance between woofers?
Could I get away with an inch between the driver and the edge of the baffle and still get away from the diffraction?
Could you explain further please? I assume you mean off setting the tweeter from center, but do you mean to vary the distance between woofers?
Could I get away with an inch between the driver and the edge of the baffle and still get away from the diffraction?