I am working on a powered speaker that consists of the following:
Two 15-inch woofers
Two 6.5 inch professional style mid-bass drivers
Again - this is a powered design with a fully active x-over. No passive components to worry about here. The dual 15-inchers will be powered by a TPA3251, the two 6.5's will be powered by a TPA3116 and the tweeter will have it's own TPA3116. Each amp board with individually-adjustable gain/level.
The two 6.5 inch pro mid units have a Fs of around 100hz. I plan to cross the 15 and midbass units over around 220hz. This puts me in the neighborhood of where baffle effects start to kick in if I am not mistaken. And since the dual 15's are on their own amp (with adjustable gain), it should make this adjustment easy.
The two 6.5 inch midbass units will be mounted MTM style (above the dual 15's) with a yet-to-be determined tweeter (although I am 99% certain I know which one I want)
These 6.5 inch midbass units play pretty damn good past 3k.
I currently have THIS tweeter in my Aliexpress basket:
https://www.aliexpress.com/item/1005005687082677.html
For those who do not want to click into the link, here are some basic specs:
These are basically china clones of the Scanspeak D3004/6640 from what I understand
Beryllium dome tweeter
FS = 500Hz
High freq range extends WELL into the 25k region
Faceplate diameter is just over 100mm
Before anyone jumps to conclusion: these are some of the finer things that come out of china. I have personally heard a set of these and they are extremely well behaved tweeters mounted in solid aluminum faceplates - the quality is absolutely top tier. China has their good, bad and ugly. This is their exceptional stuff.
The effective radiating surface of the 6.5 inch midbass units is barely over 5 inches. My plan is to cross the midbass units over to this tweeter at somewhere between 1200 and 1800hz.
Getting up towards 2000 means getting close to the center-to-center limit possible with this combination. Closer to 1200 means the half wavelength is still larger than the center-to-center AND width of the midbass effective radiating surface.
The response of the tweeters I am looking at starts to drop slightly around 1500 - so crossing around 1200 seems like I may be pushing my luck a bit.
My active x-over design is currently a purely analog 12db linkwitz/riley active filters on protoboard. I have not soldered in the components for the mid-tweet x-over yet.
So... am I in the right ballpark here? Would 1500 be reasonable? At 1500, the half-wavelength is around 4.5 inches - about a half inch short of what the C-T-C distange would be from midbass to tweeter (and slightly less than the effective radiating surface of the midbass). 1200 puts the 1/2 wave larger than both, but I am riding the enge of where the FR of the tweeter starts to drop a bit. 1200 is also still over twice the Fs of that driver - so not like I am pushing my luck too much here. Worse case scenario, I need to add a simple shelf filter?
Or am I completely mis-calculating things?
Two 15-inch woofers
Two 6.5 inch professional style mid-bass drivers
Again - this is a powered design with a fully active x-over. No passive components to worry about here. The dual 15-inchers will be powered by a TPA3251, the two 6.5's will be powered by a TPA3116 and the tweeter will have it's own TPA3116. Each amp board with individually-adjustable gain/level.
The two 6.5 inch pro mid units have a Fs of around 100hz. I plan to cross the 15 and midbass units over around 220hz. This puts me in the neighborhood of where baffle effects start to kick in if I am not mistaken. And since the dual 15's are on their own amp (with adjustable gain), it should make this adjustment easy.
The two 6.5 inch midbass units will be mounted MTM style (above the dual 15's) with a yet-to-be determined tweeter (although I am 99% certain I know which one I want)
These 6.5 inch midbass units play pretty damn good past 3k.
I currently have THIS tweeter in my Aliexpress basket:
https://www.aliexpress.com/item/1005005687082677.html
For those who do not want to click into the link, here are some basic specs:
These are basically china clones of the Scanspeak D3004/6640 from what I understand
Beryllium dome tweeter
FS = 500Hz
High freq range extends WELL into the 25k region
Faceplate diameter is just over 100mm
Before anyone jumps to conclusion: these are some of the finer things that come out of china. I have personally heard a set of these and they are extremely well behaved tweeters mounted in solid aluminum faceplates - the quality is absolutely top tier. China has their good, bad and ugly. This is their exceptional stuff.
The effective radiating surface of the 6.5 inch midbass units is barely over 5 inches. My plan is to cross the midbass units over to this tweeter at somewhere between 1200 and 1800hz.
Getting up towards 2000 means getting close to the center-to-center limit possible with this combination. Closer to 1200 means the half wavelength is still larger than the center-to-center AND width of the midbass effective radiating surface.
The response of the tweeters I am looking at starts to drop slightly around 1500 - so crossing around 1200 seems like I may be pushing my luck a bit.
My active x-over design is currently a purely analog 12db linkwitz/riley active filters on protoboard. I have not soldered in the components for the mid-tweet x-over yet.
So... am I in the right ballpark here? Would 1500 be reasonable? At 1500, the half-wavelength is around 4.5 inches - about a half inch short of what the C-T-C distange would be from midbass to tweeter (and slightly less than the effective radiating surface of the midbass). 1200 puts the 1/2 wave larger than both, but I am riding the enge of where the FR of the tweeter starts to drop a bit. 1200 is also still over twice the Fs of that driver - so not like I am pushing my luck too much here. Worse case scenario, I need to add a simple shelf filter?
Or am I completely mis-calculating things?
Making some assumptions, my main concern would be whether a single typical dome tweeter is going to keep up with the rest of the system. If you are planning to play to the limits of the dual pro-style midbasses and woofers, I doubt the tweeter will have the volume displacement to deal with where you may want to take it.
If you're just using extra drivers to limit excursion and only have modest SPL expectations, you may be OK though.
If you're just using extra drivers to limit excursion and only have modest SPL expectations, you may be OK though.
Yeah - I considered that.
The dual 15's and dual 6.5's are more to limit distortion/excursion BUT... it WOULD be nice to really crank these outdoors from time to time.
The issue is: the only real solution to higher-SPL tweeters are the bullet mini horn tweeters and... those things sound absolutely horrific. I can not even get REW measurements that show less than 5-10% distortion at even lower levels. The other issue there is: they rarely reach down very well below 2.5k and that puts the x-over frequency far above the ideal range given the C-T-C distances I am working with.
Those MeloDavid tweeters are the beefiest thing that still sounds good I could think of or find. Just 5 watts at 1800hz (92db sensitivity) is pretty darn loud, no?
If someone has a recommendation for something capable of [slightly] higher spl - I am happy to hear. Although, this is NOT a deal breaker.
Any sugestions?
The dual 15's and dual 6.5's are more to limit distortion/excursion BUT... it WOULD be nice to really crank these outdoors from time to time.
The issue is: the only real solution to higher-SPL tweeters are the bullet mini horn tweeters and... those things sound absolutely horrific. I can not even get REW measurements that show less than 5-10% distortion at even lower levels. The other issue there is: they rarely reach down very well below 2.5k and that puts the x-over frequency far above the ideal range given the C-T-C distances I am working with.
Those MeloDavid tweeters are the beefiest thing that still sounds good I could think of or find. Just 5 watts at 1800hz (92db sensitivity) is pretty darn loud, no?
If someone has a recommendation for something capable of [slightly] higher spl - I am happy to hear. Although, this is NOT a deal breaker.
Any sugestions?
I agree that 1500hz may be too low for the dome tweeter.
Do you need MTM? If not, consider just doing MT and crossing at 2500hz or higher. The advantages of MTM are less excursion and a symmetric vertical dispersion pattern. However they require a low crossover point if you follow the 1/4 wavelength center-to-center rule.
Maybe controversial but also worth mentioning: In the past I’ve ignored the 1/4 wavelength rule and gotten good results. I think the interference concerns are worse in theory than in practice.
Do you need MTM? If not, consider just doing MT and crossing at 2500hz or higher. The advantages of MTM are less excursion and a symmetric vertical dispersion pattern. However they require a low crossover point if you follow the 1/4 wavelength center-to-center rule.
Maybe controversial but also worth mentioning: In the past I’ve ignored the 1/4 wavelength rule and gotten good results. I think the interference concerns are worse in theory than in practice.
So then... 1/4 or 1/2? I hear many mention the 1/2 wavelength "rule of thumb". That is what I have been basing my design parameters around.
If I ignore the 1/4 and 1/2 wavelength part and stick to a full wavelength - it easily let's me push things up into the 2500 range. Less than 7 inches.
At that point, the SPL is less of a concern, no? I would imagine it does not take much to get a LOT of SPL at 2500+Hz, no?
If I ignore the 1/4 and 1/2 wavelength part and stick to a full wavelength - it easily let's me push things up into the 2500 range. Less than 7 inches.
At that point, the SPL is less of a concern, no? I would imagine it does not take much to get a LOT of SPL at 2500+Hz, no?
Here is another thought....
https://www.aliexpress.com/item/32257141852.html
This is a planar design coupled to an aluminum horn. It digs down into the 500 hertz range - so crossing around 1k or 1.2k should be a cake walk. It also sports a sensitivity of 100db. That extra sensitivity coupled with the horn should help balance the SPL demands when higher than normal, no?
Scratch that last link - I thought I was looking at the price in pesos, not USD. Not sure I want to spend that much on each unit lol..
https://www.aliexpress.com/item/32257141852.html
This is a planar design coupled to an aluminum horn. It digs down into the 500 hertz range - so crossing around 1k or 1.2k should be a cake walk. It also sports a sensitivity of 100db. That extra sensitivity coupled with the horn should help balance the SPL demands when higher than normal, no?
Scratch that last link - I thought I was looking at the price in pesos, not USD. Not sure I want to spend that much on each unit lol..
Last edited:
I am working on the cabinets now.
I can get a FR graph once the cabinets are done in 2-3 weeks.
But this is where the devil is in the details. When I start routing out the flush-mount spots for the tweeter - I need to know what tweeter I am mounting before that part of the cabinets happens!
So I need to actually BUY the tweeters and have them in my hands in order to be able to confirm the fabrication measurements... in order to build the cabinets... in order to get in-cabinet measurements lol.
Chicken before the egg issue here.
Yeah those are $2k each and much too tall for an MTM design.
In the pricey-but-good category, I can recommend the Bliesma T34A-4 as a tweeter that can be crossed low and played loud.
That's the thing with MTM. The spacing issues persist in narrowing radiation, down until you're low enough that it makes no difference.
TM gives symmetric radiation too, it's all about where you point your lobes.
The original MTM was about deliberately making them asymmetric, but mirroring them, with the two sets of lobes overlapping to fill in the nulls.
It’s hard to say without trying it. For example, in this build I use an MTM with a 6khz first-order crossover. In theory that would be a disaster, but in practice they sound really great. Any comb filtering is really subtle, and you need to move around the room and know what to listen for to hear it. YMMV. I would guess you’ll be OK with 2.5khz crossover.
gregulator - thanks for the recommendation.
Placing this in the short list of "very likely candidates". Yeah - a bit pricey, but the specs look really nice and I can cross that around 1.2 without any issue. 3mm linear excursion - jesus....
Anyone have any thoughts on the Dayton AMTPRO-4 Pro? I like that open back design: open baffle for the AMT would be nice. That's a pretty large tranducer and is speced for PA applications. Digs down into the 800hz region - so low xover point is possible.
Can two of these be stacked without risk of comb filtering issues?
Placing this in the short list of "very likely candidates". Yeah - a bit pricey, but the specs look really nice and I can cross that around 1.2 without any issue. 3mm linear excursion - jesus....
Anyone have any thoughts on the Dayton AMTPRO-4 Pro? I like that open back design: open baffle for the AMT would be nice. That's a pretty large tranducer and is speced for PA applications. Digs down into the 800hz region - so low xover point is possible.
Can two of these be stacked without risk of comb filtering issues?
I keep coming back to this monster:
https://www.aliexpress.com/item/1824798165.html
$360 for two of them shipped to me here in Mexico. Horn loaded (105db sensitivity), easily crossed over at 1200 and a very respectable response (if the graph can be believed).
Seriously thinking of taking the chance on this and reporting back for those who may be interested. This looks like a very capable driver for pro-audio applications.
Availability of the Bliesma T34A-4 (especially here in Mexico) is not good. Most sites show it out of stock.
The Dayton AMTPRO-4 seems known to be less-than-stellar FR-wise and falls short around 18k.
https://www.aliexpress.com/item/1824798165.html
$360 for two of them shipped to me here in Mexico. Horn loaded (105db sensitivity), easily crossed over at 1200 and a very respectable response (if the graph can be believed).
Seriously thinking of taking the chance on this and reporting back for those who may be interested. This looks like a very capable driver for pro-audio applications.
Availability of the Bliesma T34A-4 (especially here in Mexico) is not good. Most sites show it out of stock.
The Dayton AMTPRO-4 seems known to be less-than-stellar FR-wise and falls short around 18k.
Looks like it's down about 20 dB from 2 kHz to 1 kHz though.
Without seeing distortion measurements, it's hard to judge tweeter suitability for low cross points.
Using this driver in MMT setup, crossed around 2k... I can get that driver mounted close enough that the only distance which matters is center of the top mid and basically... the edge of this ribbon driver, no?
The midbass drivers should not start beaming until around 2800, so pushing the xover up as far as 2200 or 2400 would be fine I imagine.
I am assuming the vertical center of a ribbon driver does not matter since it is a line source - correct?
Or am I sweating too much over the issues of potential beaming and driver spacing? lol
The midbass drivers should not start beaming until around 2800, so pushing the xover up as far as 2200 or 2400 would be fine I imagine.
I am assuming the vertical center of a ribbon driver does not matter since it is a line source - correct?
Or am I sweating too much over the issues of potential beaming and driver spacing? lol
Yes. What I'm trying to say is that beaming isn't a good or bad thing. When you want it, you want it and when you don't, you don't. Eg waveguides beam deliberately, and they are popular.
You have chosen a ribbon. They will produce an asymmetric radiation pattern until low enough in frequency that they lose their line status and look more like a point. They will then be more axisymmetric and they will be wide.
You have chosen a ribbon. They will produce an asymmetric radiation pattern until low enough in frequency that they lose their line status and look more like a point. They will then be more axisymmetric and they will be wide.
Right approach assuming piston size for directionality.
Problem is it is not one 6.5" it is x2
So vertical it is closer to 13"
plus a tall tweeter in the middle
center to center becomes closer to a 15" speaker
off axis response.
MTM gurantee is waste of time.
Just do woofer / woofer / mid mid tweet.
Even then your closer to a 12" off axis
if you mount the mids dead tight rim to rim.
Basically one mid could have high end rolled off
to fill baffle step. then no upper frequency cancellation.
actually easier to do passive
thats just me though.
Depending on SPL goals
200 Hz crossover for mids could get little close
to what it can do thermally/ and distortion wise.
600 Hz for the 15" is pretty normal.
15" has a little more magic for mid bass.
With center to center of 2x 15"
would have to calculate
might be closer to 300 to 400 Hz
for ok crossover point before directionality.
If you sim the driver accurately on the baffle
size and positions you plan on using.
with even sim off axis info
You will quickly see how much a waste of time MTM is
Not being negative.
Its just not needed or have any advantage at all.
unless the worst possible vertical and worst possible center to center is the design goal.
go for it.
Problem is it is not one 6.5" it is x2
So vertical it is closer to 13"
plus a tall tweeter in the middle
center to center becomes closer to a 15" speaker
off axis response.
MTM gurantee is waste of time.
Just do woofer / woofer / mid mid tweet.
Even then your closer to a 12" off axis
if you mount the mids dead tight rim to rim.
Basically one mid could have high end rolled off
to fill baffle step. then no upper frequency cancellation.
actually easier to do passive
thats just me though.
Depending on SPL goals
200 Hz crossover for mids could get little close
to what it can do thermally/ and distortion wise.
600 Hz for the 15" is pretty normal.
15" has a little more magic for mid bass.
With center to center of 2x 15"
would have to calculate
might be closer to 300 to 400 Hz
for ok crossover point before directionality.
If you sim the driver accurately on the baffle
size and positions you plan on using.
with even sim off axis info
You will quickly see how much a waste of time MTM is
Not being negative.
Its just not needed or have any advantage at all.
unless the worst possible vertical and worst possible center to center is the design goal.
go for it.