Hello,
I have built several speakers from kits and proven designs and I would like to design and build something more specific. I have been using commercial on wall speakers in my living room for a couple years and they can provide a better sounding and balanced lower end when speakers cannot be placed 2-3ft out from the wall. BUT, there are also potential issues at both ends of the frequency response.
The bass response should account for boundary reinforcement, and the treble response can be limited by placement. Wall mounted speakers cannot be toed in and out to balance the treble and the location of the speakers on the wall may be limited by the location of wall studs, fire places, TVs, shelving, etc. Placement near-wall on stands might be a solution for some, but for true on-wall speakers these limitations require some combination of the following “solutions” to get the treble “right” (whatever that means to you).
Goals:
Maybe something that will look like these R speaker concepts:
Room integration concept of the “sweet spot” for mid and high FR:
I will be comparing the final product to my Dali Ikon On-Wall and Revel s16 speakers. My goals are largely the best of both worlds from these speakers. Both are good, but have issues which I hope to solve in this DIY.
Drivers:
Woofer - I would like to use a pair of ScanSpeak Illuminator 18WU/8741T which I got used. WinISD suggests they should be fine in a 12L sealed box which gives a Qtc of 0.79. Have I interpreted this correctly? Are there other considerations at this point?
Tweeter - The SEAS Prestige 27TBCD/GB-DXT (H1499) tweeter appears to fit the prompt. The trade offs are $ and sensitivity. I have plenty of power available and this tweeter appears to have well controlled directivity and reasonable dispersion without the need for custom waveguides. Other suggestions are welcome.
I will measure the drivers in the enclosure for the final crossover design, but as a proof of concept and a learning exercise, I’ve modeled a crossover in VituixCAD for these drivers which I think will provide the in room balance I’m looking for. I’ve also included the modeled effect of the baffle and enclosure. Just hoping to make sure these drivers are compatible before I start laying down money and making saw dust. I have spent about an hour reading about crossover design, so I have a lot to learn and more reading to do.
Any input is appreciated.
Thanks,
Joseph
I have built several speakers from kits and proven designs and I would like to design and build something more specific. I have been using commercial on wall speakers in my living room for a couple years and they can provide a better sounding and balanced lower end when speakers cannot be placed 2-3ft out from the wall. BUT, there are also potential issues at both ends of the frequency response.
The bass response should account for boundary reinforcement, and the treble response can be limited by placement. Wall mounted speakers cannot be toed in and out to balance the treble and the location of the speakers on the wall may be limited by the location of wall studs, fire places, TVs, shelving, etc. Placement near-wall on stands might be a solution for some, but for true on-wall speakers these limitations require some combination of the following “solutions” to get the treble “right” (whatever that means to you).
- An angled front baffle
- Wide, controlled dispersion
- Bright on-axis frequency response such that the off-axis is more ideal
Goals:
- 2-way constant(ish) directivity (without a custom waveguide)
- <12L sealed
- Frequency response to account for boundary reinforcement at low end and slightly bright to balance off axis listening
- I will be using with a sub
- Max SPL> 99 db @ 1m. I listen at around 77 db @ 3m from each speaker equating to 84 db(per speaker) @ 1m, but I would like at least 15 db headroom, hence 99 db.
- Tweeter below the woofer. This is the only way to get the tweeter at ear level in my room.
- Distortion less than -45 dbr above ~100 Hz, 84 db @ 1m. This is the measured room noise floor relative to the preferred volume at listening position. This is more aspirational.
Maybe something that will look like these R speaker concepts:
Room integration concept of the “sweet spot” for mid and high FR:
I will be comparing the final product to my Dali Ikon On-Wall and Revel s16 speakers. My goals are largely the best of both worlds from these speakers. Both are good, but have issues which I hope to solve in this DIY.
Drivers:
Woofer - I would like to use a pair of ScanSpeak Illuminator 18WU/8741T which I got used. WinISD suggests they should be fine in a 12L sealed box which gives a Qtc of 0.79. Have I interpreted this correctly? Are there other considerations at this point?
Tweeter - The SEAS Prestige 27TBCD/GB-DXT (H1499) tweeter appears to fit the prompt. The trade offs are $ and sensitivity. I have plenty of power available and this tweeter appears to have well controlled directivity and reasonable dispersion without the need for custom waveguides. Other suggestions are welcome.
I will measure the drivers in the enclosure for the final crossover design, but as a proof of concept and a learning exercise, I’ve modeled a crossover in VituixCAD for these drivers which I think will provide the in room balance I’m looking for. I’ve also included the modeled effect of the baffle and enclosure. Just hoping to make sure these drivers are compatible before I start laying down money and making saw dust. I have spent about an hour reading about crossover design, so I have a lot to learn and more reading to do.
Any input is appreciated.
Thanks,
Joseph
You probably already know these:
https://heissmann-acoustics.de/en/wavewall182-wf182bd10-tw030wa09-wandlautsprecher-kinolautsprecher/
https://heissmann-acoustics.de/en/wavewall182-wf182bd10-tw030wa09-wandlautsprecher-kinolautsprecher/
In your scheme I notice quite some baffle step correction (see the electrical transfer function). You don’t need that with on-wall designs. I see you’ll measure the acoustic output before designing the crossover. That‘ll be ok then.
Greets!
In general, best IME to follow the small cinema pioneer's conclusion that these should ideally be dipole and that if no diffuse soundfield sides/at/behind the listening position (Lp), then the entire 'system' should ideally be offset at some golden or acoustic ratio to randomize reflections and of course something to deal with floor/ceiling 'bounce'.
Another more recent addition, though still many decades ago now is to mount at the ceiling/wall juncture combined with toe-in to render the rest ~moot if its polar response is designed for it.
How should I measure an onwall speaker for crossover design?
To measure speakers, I generally do gated measurements (quasi-anechoic) down to 200-300 Hz with the speaker a couple meters from walls, splice in a near field bass measurement, then correct for the baffle using a baffle edge diffraction simulator (Bagby's or in VituixCAD). How might I change this to better capture and onwall speaker?
If this is available online I would like to read it! If not, do you know what issue it was in? Thx
Have the speaker in place on the wall or against a false wall. Measure 0-90 about the usual axes.
Vituixcad may have a problem with this non-freestanding design so you may need to do some of the calculations by hand, and verify the output you get.
The faceted cabinet will have the least effect on the drivers response other than a hemisphere.
So……..here’s the thing…..if your a sit down and listen kinda guy (your image plan suggests AV use), AMT tweeters will have better off axis horizontal response and IMO won’t require any toe-in…..better overall and more even power response given the boundary. Because of the narrower vertical directivity, you won’t have to steer that fwd lobe…..there won’t be one to be deserve concern.
For measurements and prototyping , mount your enclosure to a 4x4 sheet of plywood outside if you can.
SS woofers…….dude…..I know you already have em but the chassis is super deep!…..and pole vented so you’re going to have to allow clearance for that too or you’ll wind up with power compression issues as well as a possible resonance inside the box…..that IS NOT a shallow/on wall driver. Better off to sell em and use the $$$ for a more suitable driver…….Aurum Cantus Amts are the best out there for the $$$…….the Scanspeak sale will fund a pair.
80hz f3?………you’re gonna need more than one sub and if you’re diagram showing a box on the left as the location?……no go……two subs on the front wall within the soundstage phase pattern…..if not, the sun will easily be locatable.
So……..here’s the thing…..if your a sit down and listen kinda guy (your image plan suggests AV use), AMT tweeters will have better off axis horizontal response and IMO won’t require any toe-in…..better overall and more even power response given the boundary. Because of the narrower vertical directivity, you won’t have to steer that fwd lobe…..there won’t be one to be deserve concern.
For measurements and prototyping , mount your enclosure to a 4x4 sheet of plywood outside if you can.
SS woofers…….dude…..I know you already have em but the chassis is super deep!…..and pole vented so you’re going to have to allow clearance for that too or you’ll wind up with power compression issues as well as a possible resonance inside the box…..that IS NOT a shallow/on wall driver. Better off to sell em and use the $$$ for a more suitable driver…….Aurum Cantus Amts are the best out there for the $$$…….the Scanspeak sale will fund a pair.
80hz f3?………you’re gonna need more than one sub and if you’re diagram showing a box on the left as the location?……no go……two subs on the front wall within the soundstage phase pattern…..if not, the sun will easily be locatable.
Another similar design here: https://jblpro.com/en/products/c211
Since you're on the wall, the effective baffle area is large potentially giving good stereo imaging. If you can have it, it's a good setup considering the space occupied, works for cinema as well, with minimal modifications, whether screen or not.
Since you're on the wall, the effective baffle area is large potentially giving good stereo imaging. If you can have it, it's a good setup considering the space occupied, works for cinema as well, with minimal modifications, whether screen or not.
It was in Linear Audio Volume 7. I don't think it is available on line. Here is a small desciption:
https://www.linearaudio.net/index.php/prisma-loudspeaker-shapely-adventure
Thanks for your help. I've got a couple questions and comments.
I have read this (and have no personal experience), but looking at the measurements of AMT and ribbon tweeters in the $100-$200 USD range on Test Bench and HiFiCompass, they don't really seem to have better dispersion or directivity than the Seas DXT. Am I missing something?
If I give the SS Illuminator 1 in (25mm) behind it, it should fit well in an enclosure with the front baffle 6 in (150mm) from the wall. This is within the range that I see in on-wall designs. My understanding is that I shouldn't hear any difference in timing as long as the front baffle is within 10 in (250mm) of the wall. I have considered the SB Acoustics SB17CAC35-8 and Wavecor WF182BD which are slimmer and should give a similar bass response in a 12L sealed box, but with less SPL than the Illuminator and marginally higher distortion. So, if the SS Illuminator's only tradeoff is that it needs a deeper box, 6 in (150mm) vs a 4.5 in (115mm), that seems like a good trade. Are there other considerations I'm not accounting for?
The transfer function magnitude suggests closer to 70 Hz f3. In my current setup the sub is crossed at 90 Hz, and you are correct, it is locatable. BUT, I wouldn't say easily. I spend a lot of time just listening in the dark, and the location of the sub very rarely pops out at me. Maybe because there is a chair between my favorite spot and the sub, or maybe because I'm a pin head and the distance between my ears is so small. In any case, the sub can't really move, so I'll be happy if I can drop its crossover by 10 Hz.
Thanks!
You might even consider installing a big "false wall" for flush mounted 2pi 2-way speakers including the sub?
Perhaps. The reinforcement and cancellation due to the reflection off the front wall will requires handling in a high quality on-wall design. With a 3 way there are several ways the issue can be designed out but a 2 way is more limited in what can be done. If the complexity, size and inefficiency of a passive cardioid cabinet is inappropriate then making the cabinet shallow and wide will raise the frequency of the cancellation dip and weaken it. Making a cabinet deeper for a deep driver is going to make the cancellation worse. Note that cancellation dips cannot generally be equalized out in the crossover they need to be designed out. The reinforcement can generally be equalized out which I think you are taking on board?
Here is the designmaster Albert Von Schweikerts surroundspeakers from around 1995, i have all of this modell "Virtual Reality series"
A fantastic wall surroundspeaker and i have real high demands.......and they go quite deep
The modell TS150 and the big TS200 surround speakers (Virtual Reality series)
They are switchable between monopole and dipole operation via a switch, which allows tailoring the expansiveness / openness of the sound depending on the program materials.
Also have 2 pairs of their "big brother" the TS310 floor surround with 3 x 5 inch mid and 3 x tweeter with a 10 inch woofer, and i mesaured it once without preperations.
Regards John
A fantastic wall surroundspeaker and i have real high demands.......and they go quite deep
The modell TS150 and the big TS200 surround speakers (Virtual Reality series)
They are switchable between monopole and dipole operation via a switch, which allows tailoring the expansiveness / openness of the sound depending on the program materials.
Also have 2 pairs of their "big brother" the TS310 floor surround with 3 x 5 inch mid and 3 x tweeter with a 10 inch woofer, and i mesaured it once without preperations.
Regards John
Attachments
-
20201209_210214.jpg -
1836650-a-pair-of-von-schweikert-ts200-monopole-dipole-surround-speakers.jpg -
1836648-a-pair-of-von-schweikert-ts200-monopole-dipole-surround-speakers.jpg -
1292667-von-schweikert-audio-ts150-dipolemonopole-surround-speakers-in-maple-finish.jpg -
1292666-von-schweikert-audio-ts150-dipolemonopole-surround-speakers-in-maple-finish.jpg -
Von S TS-310.jpg
The thought is that it will balance the boundary reinforcement at low frequencies and balance the off axis listening at high frequencies. With a slightly bright frequency response the 40 degree horizontal listening window might be +2 to -2 db at 10 kHz. With a flat frequency response and the same dispersion, the 40 degree horizontal window would be 0 to -4 db. In my crude test using integrated amp tone controls and my current speakers, +2 db to treble was barely noticeable, but -4 db was obvious.
That is the idea, but I'd like to hear other opinions or experiences. Maybe I will get a mini-DSP to test FR options in room before designing a final passive crossover?
If you’re serious on this, polar measurements (0-90 degrees, in both axis) are almost mandatory. But that requires some measurement setup: you’ll have to rotate the microphone in a quarter circle around the loudspeaker.
Alternatively optimize flat response at 15 degrees. Lots of designers go this easy way and something has to be said for it.
Alternatively optimize flat response at 15 degrees. Lots of designers go this easy way and something has to be said for it.