Planning a 2 way with BMR's

Status
Not open for further replies.
Hi all,

I'm in the planning stage of building my first pair of speakers.

I have a small living room 11' x 11' that extends into a dining room with roughly the same dimensions.

My current speakers are Polk RTiA7's RTiA7 - Polk Audio

Polk's are vertically bi-amped using two HK3490 receivers and I use a Tubedac 11 as a DAC/preamp
Harman Kardon HK 3490 XM Ready stereo receiver at Crutchfield.com

Speaker plan:

I would like to use (8) Tectonic Elements TEBM46C20N-4B BMR 3" Full Range Speaker 4 Ohm Wired to 8 ohms

And (4) Aurum Cantus AC-130MKII 5-1/4" Carbon Fiber/Kevlar Woofer I would be happy with bass down to 35HZ but lower if possible.

These will be Vertically Bi-amped with my current equipment sans DAC

I will be using a miniDSP for a Crossover and I figure a crossover somewhere around 300-600HZ.
First questions

1) Would the AC-130MKII's be a good match for the TEBM46C20N-4B BMR's?

2) Is there some kind of golden rule for the number of BMR's (mid/highs) to Woofers?

3) Whats a good crossover range 300, 400, 500 or 600 or something else?

Thanks for your time


- Cory
 
Last edited:
I would like to build a 2-way speaker using multiples of TEBM46C20N-4B BMR

X over @ 600HZ

And 1 or 2 Usher 8945A 7" Carbon Fiber/Paper Woofer's

I want a big sound, was thinking 12 or more of the BMR's

Hmm, with a ~4,592 Hz max 1 WL ctc spacing, the array in theory will need to be concave to focus it at the listening position [LP], which will severely limit useable off axis seating, so will need to be toe'd in; or be at least > 2/3 room height in length to appear as being an 'infinite' array and even then the HF may still sound somewhat 'phasey'.

That said, there is [use to be?] a small four driver array from England that folks liked, but no longer have the info.

The little 8945 has a ~1,945 Hz half space HF corner, whereas the smaller BMR's LF corner is at ~1,266 Hz, so combined with a ~1,636 Hz spacing ctc and a theoretical ~281 Hz Fb in a BMR sealed 0.707 Qtc alignment requiring a ~562 Hz XO, a 600 Hz XO is close enough.

GM
 
Cory,

If this is your first speaker design attempting arrays of tweeters is very ambitious. One of the advantages of speaker projects is the almost always work to some extent i.e. you will get sound out. However optimising such an array of HF drive units will be very difficult. One limited did a design something like this but they used a DSP to manage the phased array they subsequently sold the IP to Yamaha I believe.

The bass drivers look like they have a good fairly well controlled mid response, I would recommend looking for a good standard tweeter to go with them and create a 2.5 way design.

Small drivers are not ideal for very low frequency solutions as you will need high levels of excursion which will mess up the mid range response. 35Hz will push these drivers very hard. I did a simulation of this driver in the optimum box size (as predicted by the simulator you can often optimise manually better). You can see to get down to 35Hz you will need an extended bass shelf design which will make your bass sound somewhat thin. You could EQ this however the excursion limit of the drivers will be exceeded at fairly low power. It also requires a massive box.

The second graph shows the excursion for 20W input this provide around 100dB of output which is loud but not excessive. As you can see the drivers are on their excursion limit at even this fairly modest power.

If you want to play down to 35Hz or lower you really need a large sub to go with the system.
 

Attachments

  • Untitled.jpg
    Untitled.jpg
    508.3 KB · Views: 531
  • excursion.jpg
    excursion.jpg
    507.7 KB · Views: 508
If you want to play down to 35Hz or lower you really need a large sub to go with the system.


due to the vaguely described size and shape of combined listening space, perhaps even multiples of smaller size to mitigate modes -ultimately the room volume/shape take control, and you could be fighting a losing battle with diminishing returns
 
Last edited:
Hi Cory,

I have been developing a range of speaker designs using various commercially available BMR drivers as well as an OEM version using a 4.5 inch (112mm square) chassis and strong Neo motor with a silk surround and treated cotton / linen spider.

BMR's come in all shapes and sizes and just like conventional drivers there are good ones and bad ones!

I have had some great results as well as some disasters so hopefully I can help you avoid the same mistakes!

The attached is a small nearfield studio monitor using our 10 inch bass mid and a single 4.5 inch BMR.
The production versions are really small 312mm by 412mm by 175mm (width height depth) well under a cubic foot external volume and under half a cubic foot (14 litres) internal air volume after internal components.

The performance is staggering in my opinion (biased!) and gob smacking in the opinion of others....I will be doing some demo's of this prototype pair over the next few weeks and some of the DIY guys will no doubt be posting their opinions....!! 😉

My general advice is if you are using the small PE BMR's use lots of them....at least 4 per side, even better with 8 or 16 and build tall shallow cabinets you can hang on the wall.

For the 3 inch (Sd of only 19cm 2?) PE driver in a small room and average listening SPL's here are some suggestions for crossover points to a small bass / mid driver.

Single BMR - Min 400Hz best with 800Hz.
Twin BMR - 300Hz to 500Hz.
Four BMR - 250Hz to 400Hz
Eight BMR - 150Hz to 300Hz.
Sixteen BMR - 80Hz to 200Hz.

Forget all passive crossovers!
One must use DSP and Eq to crossover and tune the BMR's. The bass will also benefit massively from ditching the passive crossovers.
An decent (Yamaha is my choice) AV home cinema amp with its DSP / crossover / room Eq and a seperate power amp to drive the bass / mids ( use the L&R sub outs) will demolish the performance of any passive crossover.....
Even the any snake oil $1,000 per inductor / capacitor / resistor and golden wire, unobtainium solder all hand built into special "tone" wood crossover boxes sitting on meteorite component supports.....Sorry, I hate passive crossovers!!

You can forget about all the problems that afflict conventional speakers
using cones / domes and all the usual stuff you see on DIY sites.....Comb filtering / beaming / lobing horizontal / vertical dispersion / rear wall / side wall / floor & ceiling bounce, reflections etc etc.....

A well designed and installed (Eq'd) on wall BMR array (taller than 75% of the floor to ceiling height) smashes all those issues into tiny irrelevant pieces and simply fills a room (small or large) with the most even, well balanced and natural sounds you can imagine.....Build it and they will come!!

Good luck and hope this helps.
Cheers
Derek.
 

Attachments

  • 25 OCT SM15 AND 10 138.jpg
    25 OCT SM15 AND 10 138.jpg
    894.3 KB · Views: 546
and not having heard a BMR of any of the types Derek or Cory have mentioned, I can only posit that they might not all be created equal

Derek - pardon if the question has been answered elsewhere - what are the BMR's seen in your photos & pretty renderings?
 
Hi Cory,

I have been developing a range of speaker designs using various commercially available BMR drivers as well as an OEM version using a 4.5 inch (112mm square) chassis and strong Neo motor with a silk surround and treated cotton / linen spider.

BMR's come in all shapes and sizes and just like conventional drivers there are good ones and bad ones!

I have had some great results as well as some disasters so hopefully I can help you avoid the same mistakes!

The attached is a small nearfield studio monitor using our 10 inch bass mid and a single 4.5 inch BMR.
The production versions are really small 312mm by 412mm by 175mm (width height depth) well under a cubic foot external volume and under half a cubic foot (14 litres) internal air volume after internal components.

The performance is staggering in my opinion (biased!) and gob smacking in the opinion of others....I will be doing some demo's of this prototype pair over the next few weeks and some of the DIY guys will no doubt be posting their opinions....!! 😉

My general advice is if you are using the small PE BMR's use lots of them....at least 4 per side, even better with 8 or 16 and build tall shallow cabinets you can hang on the wall.

For the 3 inch (Sd of only 19cm 2?) PE driver in a small room and average listening SPL's here are some suggestions for crossover points to a small bass / mid driver.

Single BMR - Min 400Hz best with 800Hz.
Twin BMR - 300Hz to 500Hz.
Four BMR - 250Hz to 400Hz
Eight BMR - 150Hz to 300Hz.
Sixteen BMR - 80Hz to 200Hz.

Forget all passive crossovers!
One must use DSP and Eq to crossover and tune the BMR's. The bass will also benefit massively from ditching the passive crossovers.
An decent (Yamaha is my choice) AV home cinema amp with its DSP / crossover / room Eq and a seperate power amp to drive the bass / mids ( use the L&R sub outs) will demolish the performance of any passive crossover.....
Even the any snake oil $1,000 per inductor / capacitor / resistor and golden wire, unobtainium solder all hand built into special "tone" wood crossover boxes sitting on meteorite component supports.....Sorry, I hate passive crossovers!!

You can forget about all the problems that afflict conventional speakers
using cones / domes and all the usual stuff you see on DIY sites.....Comb filtering / beaming / lobing horizontal / vertical dispersion / rear wall / side wall / floor & ceiling bounce, reflections etc etc.....

A well designed and installed (Eq'd) on wall BMR array (taller than 75% of the floor to ceiling height) smashes all those issues into tiny irrelevant pieces and simply fills a room (small or large) with the most even, well balanced and natural sounds you can imagine.....Build it and they will come!!

Good luck and hope this helps.
Cheers
Derek.

Hi Derek,

It helps immensely, I was hoping you would post🙂

I read your post from awhile back which got me interested in BMR's and active xover's.

Being new I have a lot of research to do and all the posts I receive are so helpful in pointing me in the right direction.

Off to do more reading and of course I will be back for more training!

Tonight i'm researching using cement enclosures for the BMR's🙂

Thanks everybody!
 
Hmm, with a ~4,592 Hz max 1 WL ctc spacing, the array in theory will need to be concave to focus it at the listening position [LP], which will severely limit useable off axis seating, so will need to be toe'd in; or be at least > 2/3 room height in length to appear as being an 'infinite' array and even then the HF may still sound somewhat 'phasey'.

That said, there is [use to be?] a small four driver array from England that folks liked, but no longer have the info.

The little 8945 has a ~1,945 Hz half space HF corner, whereas the smaller BMR's LF corner is at ~1,266 Hz, so combined with a ~1,636 Hz spacing ctc and a theoretical ~281 Hz Fb in a BMR sealed 0.707 Qtc alignment requiring a ~562 Hz XO, a 600 Hz XO is close enough.

GM

Thanks GM, I looked up C2C and now know the formula and what its for but am not sure how you came up with the combined 1,636hz?
 
and not having heard a BMR of any of the types Derek or Cory have mentioned, I can only posit that they might not all be created equal

The BMR's I will be using are less equal than Dereks 4.5's which are OEM with cloth surrounds.

I believe there are other 4.5's available but the response curve has a deep canyon around 2K if I remember correctly.

The 3 inch have a pretty flat response curve from 200-20k.
 
Thanks GM, I looked up C2C and now know the formula and what its for but am not sure how you came up with the combined 1,636hz?

You're welcome!

The acoustic center of two vertically oriented drivers is the center, so the minimum ctc distance assuming they are all butted up against each other is the physical length of one driver of the pair plus half the length of the [HF] driver it's mated to.

GM
 
The BMR's I will be using are less equal than Dereks 4.5's which are OEM with cloth surrounds.

I believe there are other 4.5's available but the response curve has a deep canyon around 2K if I remember correctly.

The 3 inch have a pretty flat response curve from 200-20k.

Hi Cory,

Major peaks and troughs in a drivers frequency response are never a good thing, but not always a deal breaker.
I believe accurate time domain (CSD plots show time and frequency performance) performance is far more important than an accurate frequency (amplitude) domain performance.

The reason....? Because, in certain circumstances one can fix frequency response errors with Eq, but one can never fix time domain errors....Not even with DSP Servo control systems.
Please see the attached doc which explains my reasoning in a bit more detail.


Re the frequency response of drivers in general and BMR's in particular I have found that Eq can be a great solution or actually cause more harm than good....It all depends on the drivers "power response".

The power response is the term describing how much a driver suffers from "beaming". ie directly on axis the driver might give you a nice even response covering 200 Hz up to say 10KHz, this is good, but off axis, say 30 degrees it might be 10dB down, and by 60 degrees it might be 20dB or 30dB down.

This looks bad on a graph, but is fatal in real rooms due to the dominance of reflected sound in most listening positions.
One exception is in a treated studio environment sitting very close (around 1 meter ear to speaker distance) to near-field monitors, but in most domestic or commercial spaces reflected sound dominates both measurements and subjective sound.

I believe all beaming is bad, even controlled beaming or "controlled directivity" as the marketing dept of horn speakers call it!...Broad even coverage over as wide an area as possible is good!

The Holy Grail being a totally even 180 degree coverage (plus / minus 90 degrees) in both vertical and horizontal planes. This is deemed almost impossible to achieve (until the pulsating Plasma Sphere drivers are invented!) and many loudspeaker designers believe "constant directivity" is the only alternative .....I disagree!

One very important and point which is misunderstood by many designers is that you can not successfully Eq any driver (or combination of drivers) which do not have a very flat power response.
How flat....? Accuracy = to or better than plus / minus 2.5 dB over both vertical and horizontal axis and covering at least a 140 degree (plus minus 70 degree) is required to achieve a balanced sound.
If you Eq a driver(s) on axis that suffers from beaming the already bad off axis performance will be significantly degraded and all the subsequent room reflections will amplify these errors.

I hope I have not rambled on too much some of this makes sense?
The bottom line is that Eq'ing conventional drivers can be problematic in domestic rooms, but is easy with BMR's and produces great results!

All the best
Derek.
 

Attachments

Status
Not open for further replies.