Dynaudio BM15A amplifier modification

I have a pair of dynaudio bm15a that I have never quite been happy with the sound of . theres a slight harshness to the trebble i have the passive version that uses a different tweeter and crossover frequency and i much prefer the sound of those .
I had the idea of using the amplifier and bass driver and altering the inbuilt active crossover to lower the crossover frequency to allow me to crossover to a pair of avi nuetrons about 300htz using their internal passive crossover turning both speakers into one set of three way speakers .
could someone please have a look at the schematic and tell me if this is realistic or not . if not could I somehow patch in my own active crossover before the amplifier part of the pcb and still make use of the amp
many thanks
ps it's the mk2 version I have
 

Attachments

  • bm15a-service-manual.pdf
    351.2 KB · Views: 153
I have no idea how to read the amplifier schematic, except at least everything's through-hole, so the idea might have legs if you can figure out how to set the low-pass frequency, but what of the high pass output? And what would be the plan for the tweeter in a 3-way?

Also, the 24W-100 datasheet says 90db sensitive in box -- I don't think the AVIs' small woofer can keep up. Heard those years and years ago. I remember they had good imaging but didn't have any low end at all and went for some Dynaudio Audience 50s instead.

I think the idea of building a 3-way with the 24W-100 is a good one, but passive would be a lot easier, and you'd need measuring gear, and I don't think the AVI midwoofer is sensitive enough. Maybe sell the BM15As or keep the drivers and design something with a compatible mid.
 
Last edited:
Well if I could lower the crossover frequency to 300htz something like that then the tweeter midrange crossover could be handled by the avi internal crossover . I can't imagine the woofer is as sensitive as that it's one of those heavyweight polypropylene types with a 4inch voice coil . I don't think that mismatched sensitivities would be too much of an issue anyway if I was inserting my own crossover into the plate amp I could adjust levels at the crossover or In dsp .
it's funny I like the bass presentation of the dynaudio bm15a but I much prefer everything else about the nunuetron and they cost me £140 vs £1250
 
Member
Joined 2004
Paid Member
could someone please have a look at the schematic and tell me if this is realistic or not .
With this combo the whole world of modding lays before you: +/-12V and 51V rails to clean up with more caps and new Vregs, opamps to roll, active xo to modify, and 4 channels of mosfet output stage amps, plus complete schematics and parts lists?
But you think the tweeter is harsh...?
 
thank-you . so I've had the opamps socketted when I had the plate amps repaired so that's handy .yeah I think the tweeters harsh someone online said they tracked it down to a reflection off of the woofers frame that isn't flush mounted (which I notice dynaudio never do ) I think it's a bit lazy of them . there's a slight simblance and a harsh edge to the top end that isn't there on the passive version which has a waveguided tweeter and a higher crossover frequency I think
really appreciate any help with this
I have a umik and r.e.w I'm trying to setup a digital crossover I have an 8 channel interface I'm just struggling to get my head around the differnt software options I have a pc and mac it's just a steep learning curve
always wanted take my own speakers.
I'm a carpenter so that's the woodworking sorted just always designing stuff in my head just not quite at the stage of being able to carry the designs through to reality yet
 
Last edited:
I have no idea how to read the amplifier schematic, except at least everything's through-hole, so the idea might have legs if you can figure out how to set the low-pass frequency, but what of the high pass output? And what would be the plan for the tweeter in a 3-way?

Also, the 24W-100 datasheet says 90db sensitive in box -- I don't think the AVIs' small woofer can keep up. Heard those years and years ago. I remember they had good imaging but didn't have any low end at all and went for some Dynaudio Audience 50s instead.

I think the idea of building a 3-way with the 24W-100 is a good one, but passive would be a lot easier, and you'd need measuring gear, and I don't think the AVI midwoofer is sensitive enough. Maybe sell the BM15As or keep the drivers and design something with a compatible mid.
just picked up some dynaudio d76 3inch mid dome and am probly going to incorporate these now instead and keep the avis separate.
I'm thinking a 3 and a half way might be good with 2 24w100 per channel dual opposed sharing 40ltrs . anyone got any suggestions ? seem like it might work ? got plenty of ply and most tools somethingike the picture included
 

Attachments

  • 1690129274641558350826150207602.jpg
    1690129274641558350826150207602.jpg
    203.6 KB · Views: 34
Doesn't this mean cannibalizing your good pair of BM15s?

I remember that @planet10 is an advocate of 2 opposed woofers to address baffle losses, though I don't know how this affects normal box volume calculations, and I think there are some trade-offs in terms of a response dip to anticipate.

Otherwise, the D76 datasheet suggests a nice classic mid that you'd probably want to cross steeply from 600hz or so.
 
Well as long as I don't damage anything I can always put them back together .
it doesn't seem worth selling them as they don't go for loads nowadays.
I wondered about the dispersion I prefer a wide dispersion so I guess I'm better keeping the baffle flat or sloping behind the drivers and trying to minimise diffraction than in any kind of waveguide ?
 
Diffraction is not the same as dispersion. The obvious ways to influence dispersion of the mid and tweeter would be optimal choice of crossover frequency and waveguiding them; yet if using a mid-dome rather than a cone, benefits may be marginal for the complexity of the latter. Then again, the OSMC does it and you say you're a capable woodworker up for baffle prototyping.

https://github.com/mbrennwa/osmcdoc
https://github.com/mbrennwa/osmcdoc/blob/master/osmc_paper.pdf

Some discussion of integrating a waveguide shape into the baffle itself here
Some experiments on baffle shapes here and here
 
thanks I know they're not the same thing I meant that if I'm not going to make a waveguide to narrow the dispersion I could take some measures to try to minimise any diffraction . I'm sure that there is a problem with the active speaker that might be diffraction related . the passive version doesn't seem to have the same issue and uses a diff tweeter (interestingly in a waveguide ) and diff xover points . I think I'll make a baffle acording to best practice as far as i can tell and see what it sounds like .
thanks for the links I'll have a look
 
got some progress today got the boxes made slightly taller than planned to allow for me not liking the mid dome and having to try something else made the sides and top and bottom from 25mm ply and lined them with some leftover cement board . stuck the cement board with bitumen paint and painted over the board with the same . made the boxes in a way that will let me change the baffle out without making a whole new box . they can be lined later on with something nicer looking when I've got the rest of it settled
 

Attachments

  • 20230724_192830.jpg
    20230724_192830.jpg
    327.2 KB · Views: 50
  • 20230724_192815.jpg
    20230724_192815.jpg
    329.1 KB · Views: 42
  • 20230724_192806.jpg
    20230724_192806.jpg
    604.4 KB · Views: 46
the speakers will sit in an alcove next to the fireplace in the bedroom . I'm making some built in wardrobes in a house we're refurbished and they're going to sit in my side of the wardrobe. got a 6 channel amp and motu 828mk2 usb as a multi channel dac fed from an Google TV (unreliable) going to get something better for streaming when funds allow
 

Attachments

  • 16902241083428503912513248915762.jpg
    16902241083428503912513248915762.jpg
    306.2 KB · Views: 45
  • 1690224138260817460895796888792.jpg
    1690224138260817460895796888792.jpg
    353.1 KB · Views: 36
  • 16902241578696811466267198223445.jpg
    16902241578696811466267198223445.jpg
    330.2 KB · Views: 37
  • 16902241782865786027161373225496.jpg
    16902241782865786027161373225496.jpg
    263.7 KB · Views: 41
someone commented earlier that I'd want to cross to the d76 at around 900htz I was expecting perhaps nievely to be able to cover 300 to 3000 odd with that one drover if this isn't the case I might turn it into a 4&1/2 way with a 18sound 6nd420 . I've been thinking about this build for so long now when I drift off to sleep that's its nice to actually be making it at last . took the longest time to get the software sorted for a crossover mostly because its something I didn't really understand . it seems so simple now but I felt like I was going round in circles for ages
 
D76 crossed at 900 Hz would be very wrong.

From "A history of Ruark":

Solstice rounds off Reference Series​


Such was the enthusiasm for Excalibur that many customers asked if a smaller model could be made. The result was a beautiful looking speaker called Solstice. At just over £4,000 UK retail this speaker was still aimed firmly at customers who took their audio seriously, but partnered with electronics of equal quality, the sound Solstice created was truly breathtaking.

tl-2001-solstice.jpg
 
you think it could go a lot lower ? I bought it as a nos replacement part for klein and hummel 0410 which was interesting I like the style of at actives and have always wanted a scm50 or bigger and wondered if I could make a simple waveguide or short horn with big roundover bit on the router .
I've seen those for sale from time to time . and even the next one up with two mid domes per speaker . always assumed they were and atc part In the ruark don't know why I thought that looking at them now