• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

BMS drivers in Altec M19?

J

judgeschreber

Well-Known Member
Would this work?

I'm thinking of the BMS 4550. I just acquired a pair of M19's with 802-8G drivers on 811B horns, and I love them, except that on some recordings the HF horn sound bothers my ears. They're certainly not screechy by any means, and they sound good on most recordings, but I'm wondering if they could be improved.

I'm not going to do anything rash, and I'll certainly give these speakers more listening time before I do any mods.

However, I've read over in the Lansing forums how the BMS drivers helped Valencias, and from the descriptions it sounds like it just might be what I'm looking for.

What additional crossover compensation would be required if I chose to go this route?

BTW, I'm using a recently serviced Sherwood S-8000 IV tube receiver for amplification, so I don't think what I'm hearing is a case of the SS/Horn thing. I'll also be trying these out with a MAC1700, which has a warm SS sound, but I haven't gotten around to it yet.
 
Register to hide this ad
Duffinator said:
Zilch should be here soon.
Click to expand...

Funny you should say that, since the monster Valencia thread over in the lansing heritage (which Zilch has contributed extensively to) got me started thinking along these lines.

In fact, with a little further digging in that thread I found a place where the Model 19 X-over was specifically addressed, in part answering some of my question.

http://audioheritage.org/vbulletin/showthread.php?t=14690&page=34

I see the component values for the filter, but what's the layout of an LCR filter?

Also, anyone care to try to talk me out of it? Has anyone tried the BMS drivers and found they just don't sound as good to their ears? Keep in mind I'm trying to tame the tendency towards fatigue and HF "roughness" (what I call that horny screechy sound when it's very very mild).
 
I know of one opinion posted publicly where the poster has enough experience with the vintage altec's for a comparison with the BMS drivers to be meaningful to me................. just my opinion mind you.

See Zilch's post #16 and particularly the link contained therein as Old Guy's quote is not complete in the AK thread.

EDIT:the correct link to the post to which i referred http://www.audiokarma.org/forums/showthread.php?t=124966&page=2

FWIW, i believe changing HF drivers will change the characteristic "sound" of the 19's, maybe a good thing, maybe not. Only you can be the judge. (pun intended):D

EDIT: if you're looking for suggestions, mine would be to first try recapping the XO's. Even in spec there's nothing "great" sounding about the caps Altec used from the factory. Considering their real world performance and entry fee, i don't think you can get hurt trying the BMS drivers. If they didn't fill the bill for you in the 19's they'd surely find new homes in another project of your own or someone elses with ease. Who doesn't love a "win/win" situation?
 
I obtained the best measured performance from BMS 4550 on 811B using the Model 19 crossovers Gary (Gamalot) sent me for evaluation during that study, no additional compensation required. The control settings were outside the factory "Optimum" markings for the stock drivers, is all.

Bear in mind that I did not have actual Model 19s or even Altec woofers here to confirm that there was sufficient adjustability to achieve an optimum overall system balance, but the likelihood is good that a simple driver swap will work. The BMS drivers are more sensitive than the Altecs, but the requisite compensation is similar, so a simple fixed L-Pad on the output may bring the settings into the factory standard range.

My situation is much the same today; I now have my own set of stock Model 19 crossovers for experimentation, but no Altec low frequency system. All of my work was done with relatively lower sensitivity JBL woofers, so someone with Model 19s is just going to have to try the BMS drivers and evaluate the results. If that occurs locally here, I can probably bring a portable measurement system to the site to assist in optimizing the adjustments.

Personally, if I were working with my own Model 19s, I'd spring the extra $30 apiece for 4552nds, as I do believe I can hear a difference in clarity conferred by the neodymium magnets, even though the basic frequency response is the same....

bowtie427ss said:
See Zilch's post #16 and particularly the link contained therein as Old Guy's quote is not complete in the AK thread.
Click to expand...
http://audiokarma.org/forums/showthread.php?p=1333241#post1333241
 
I've been working along a similar line with a pair of Altec Carmels I'm in the process of restoring. I don't have time at this immediate moment to go into gory details but offer the following observations:

1. The Altec 806 driver (and 802 that I used for development work) sounds more natural and "effortless" than the BMS 4550 in a plug-n-play replacement situation.

2. The BMS driver has a pronounced and somewhat broad midrange peak in the 2.5KHz range.

3. The BMS driver clearly has better high-end extension and is more efficient. The efficiency advantage permits some unique opportinities for passive equalization.

4. The BMS driver is not at its best with the stock Altec X-over (symetric version of N-800F in Carmel).

5. A proper parallel notch filter in series with the BMS driver (A-la the Zilchlab design in described in the endless "Look at my new babies" thread) can and does really flatten out the BMS response and results in an audibly extended high end.

6. It seems that a more optimum X-over point for 811 based systems is in the 1-1.2KHz region and not the nominal 800Hz that stock Altec uses if the LF driver(s) are up to it. Actually, in the unsymetric version of the N800-F, an argument could be made that the real X-over pt is closer to 1KHz. The M19 X-over is also around 1.2KHz.

7. The notch filter from #5 above cannot be simply added to an existing X-over network because it introduces a rather nasty impedance peak too close to the turn-over pt for proper operation. The solution I found was the addition of series resonant filter in parallel with the overall HF notch network and BMS driver. This evens out the impedance in the vicinity of the X-over freq for predictable x-over operation. I think this is similar to the "conjugate load" set-up KEF played with back in the 80s?

For my test Carmel, I put together a 2nd order L-R xover @ 1KHz to replace the N-800F and added a zobel to the LF section. It's undergoing a shakedown at the moment and may not be ready for prime-time. However, I can say that with all of these changes so far, the one speaker I've converted over really sounds good. High end is extended compared to the stock unit and it doesn't have a peaky sounding mid-range (honk?) like stock. I think it's an improvement over stock by a decent margin, esp in the neutrality dept.

Although this project isn't complete yet, the one major conclusion at this point is that replacing the stock Altec driver with the BMS required nearly total change of the X-over to get it to really work out to its full potential. I suspect that the same situation exists with the M19 X-over if not moreso because the M19 contains "equalization" specific to the 802 driver. IOW I don't believe the BMS 4550 is a plug-n-play replacement. OTOH, the 4550 is low enough in cost that you can experiment and listen for yourself without too much financial pain if it doesn't work out.

Maybe more to come on the Carmels inc lots of charts and plots in the style of Zilch :thmbsp:as the summer winds down and I find myself cooped up indoors.
 
bowtie427ss said:
I know of one opinion posted publicly where the poster has enough experience with the vintage altec's for a comparison with the BMS drivers to be meaningful to me
Click to expand...

Thanks for that link--that's exactly the kind of impressions I'm looking for.


bowtie427ss said:
Considering their real world performance and entry fee, i don't think you can get hurt trying the BMS drivers. If they didn't fill the bill for you in the 19's they'd surely find new homes in another project of your own or someone elses with ease. Who doesn't love a "win/win" situation?
Click to expand...

Good point :yes:

Zilch said:
I obtained the best measured performance from BMS 4550 on 811B using the Model 19 crossovers Gary (Gamalot) sent me for evaluation during that study, no additional compensation required. The control settings were outside the factory "Optimum" markings for the stock drivers, is all.
Click to expand...

So does that mean you eventually decided against the notch filter at 19Khz? I assume the "8:00" and "max" notations indicate the control settings.

Zilch said:
The BMS drivers are more sensitive than the Altecs, but the requisite compensation is similar, so a simple fixed L-Pad on the output may bring the settings into the factory standard range.
Click to expand...

I couldn't find any specs for the 802-8g, but the 802D has a published sensitivity of 111.7db, and the BMS is published as 113db. Granted, I have no idea how the different horns used affected these specs. But, perhaps only minor compensation is required?

Zilch said:
so someone with Model 19s is just going to have to try the BMS drivers and evaluate the results.
Click to expand...

:scratch2:

Zilch said:
If that occurs locally here, I can probably bring a portable measurement system to the site to assist in optimizing the adjustments.
Click to expand...

I'm not local to you, but I do have a basic measurement setup here. It's not really use-able below about 150Hz or so (I imagine due to room resonances or some such thing), but it should at least be able to get an idea of how the drivers blend.

If I get those drivers, I'll be sure to post results.

Zilch said:
Personally, if I were working with my own Model 19s, I'd spring the extra $30 apiece for 4552nds, as I do believe I can hear a difference in clarity conferred by the neodymium magnets, even though the basic frequency response is the same....
Click to expand...

Thanks for the recommendation.
 
Steve O said:
I've been working along a similar line with a pair of Altec Carmels I'm in the process of restoring. I don't have time at this immediate moment to go into gory details but offer the following observations:

1. The Altec 806 driver (and 802 that I used for development work) sounds more natural and "effortless" than the BMS 4550 in a plug-n-play replacement situation.

2. The BMS driver has a pronounced and somewhat broad midrange peak in the 2.5KHz range.

3. The BMS driver clearly has better high-end extension and is more efficient. The efficiency advantage permits some unique opportinities for passive equalization.

4. The BMS driver is not at its best with the stock Altec X-over (symetric version of N-800F in Carmel).

5. A proper parallel notch filter in series with the BMS driver (A-la the Zilchlab design in described in the endless "Look at my new babies" thread) can and does really flatten out the BMS response and results in an audibly extended high end.

6. It seems that a more optimum X-over point for 811 based systems is in the 1-1.2KHz region and not the nominal 800Hz that stock Altec uses if the LF driver(s) are up to it. Actually, in the unsymetric version of the N800-F, an argument could be made that the real X-over pt is closer to 1KHz. The M19 X-over is also around 1.2KHz.

7. The notch filter from #5 above cannot be simply added to an existing X-over network because it introduces a rather nasty impedance peak too close to the turn-over pt for proper operation. The solution I found was the addition of series resonant filter in parallel with the overall HF notch network and BMS driver. This evens out the impedance in the vicinity of the X-over freq for predictable x-over operation. I think this is similar to the "conjugate load" set-up KEF played with back in the 80s?

For my test Carmel, I put together a 2nd order L-R xover @ 1KHz to replace the N-800F and added a zobel to the LF section. It's undergoing a shakedown at the moment and may not be ready for prime-time. However, I can say that with all of these changes so far, the one speaker I've converted over really sounds good. High end is extended compared to the stock unit and it doesn't have a peaky sounding mid-range (honk?) like stock. I think it's an improvement over stock by a decent margin, esp in the neutrality dept.

Although this project isn't complete yet, the one major conclusion at this point is that replacing the stock Altec driver with the BMS required nearly total change of the X-over to get it to really work out to its full potential. I suspect that the same situation exists with the M19 X-over if not moreso because the M19 contains "equalization" specific to the 802 driver. IOW I don't believe the BMS 4550 is a plug-n-play replacement. OTOH, the 4550 is low enough in cost that you can experiment and listen for yourself without too much financial pain if it doesn't work out.

Maybe more to come on the Carmels inc lots of charts and plots in the style of Zilch :thmbsp:as the summer winds down and I find myself cooped up indoors.
Click to expand...

Thanks for the detailed response and your impressions! I'd love to see those charts and plots if/when you get around to it :yes:

I didn't understand your entire post on first reading, but I'll revisit it later to try and get a handle on it all. Although I'm fascinated with speaker design, I am still very much a novice regarding the more technical aspects. AK is a great place to learn, though :yes:
 
judgeschreber said:
So does that mean you eventually decided against the notch filter at 19Khz? I assume the "8:00" and "max" notations indicate the control settings.
Click to expand...
If you do this, I will address notching the 19 kHz "spike" again. The particular filter shown with the Model 19 crossover was an early attempt, and that was considerably refined with more precision in subsequent iterations. I think it's important to have that in there, as the spike seems to be a resonance artifact with peaked distortion, but it's a simple add-on and not essential to initially evaluating the swap. I can certainly see it in the measurements and it's shown in the factory specs, but I don't hear it.

Yes, those are the settings, as I now recall. Getting the response absolutely flat (+/- 1.5 dB or less) like that uses much of the available compensaton adjustment range. BMS 4550 has about 9 dB more sensitivity than 806, so I'd expect 6 dB more than 802. One or another of the 802 curves I published may show the actual measured differential.

I'm pleased to know you have measurement capabilities. It's really hard to do this sort of development work without that. Several other members are working with different Altec systems using BMS, but progress seems to have been slow over the summer; perhaps we'll be seeing some informative updates in the coming months.... :yes:

Edit: Why, there's an update from Steve O NOW, in fact! :thmbsp:

Steve O said:
Maybe more to come on the Carmels inc lots of charts and plots....
Click to expand...
:guitar:​
 
Hey Zilch,

In these charts you posted in the L-H forum, it looks like the sensitivity is pretty close between the BMS and the 802, with the 811B horn and M19 crossover, optimized for the respective drivers.

http://www.audioheritage.org/vbulletin/showpost.php?p=160763&postcount=493

The bottom chart has a comparison between the two drivers. Am I reading this correctly? Are they really that close in sensitivity?

If so, then I could just drop in the BMS drivers, twiddle the knobs a little, and be good to go, right?
 
judgeschreber said:
I could just drop in the BMS drivers, twiddle the knobs a little, and be good to go, right?
Click to expand...
I'd like to say so, of course, but I may have adjusted the gain between readings to show them superimposed like that.

The time stamps indicate that I tried the BMS first, then the 802, then readjusted the crossover for BMS and measured it again, as a black curve is the most recent sample.

While I do keep a lab notebook, it's not that detailed as to what the levels may have been.... :dunno:
 
Zilch said:
I'd like to say so, of course, but I may have adjusted the gain between readings to show them superimposed like that.
Click to expand...

Yeah, that was my suspicion as well. . .

Well, I'll go ahead and put those drivers on order (I'm leaning towards the 4552nd's), and after a recap of the x-overs, I'll take a few measurements to determine whether it needs an L-pad or not.

Unfortunately I probably won't get to it until November, but when I do, I'll be sure to post about it.
 
Zilch said:
Several other members are working with different Altec systems using BMS, but progress seems to have been slow over the summer; perhaps we'll be seeing some informative updates in the coming months.... :yes:
Click to expand...
Hey, you talkin' ta me?! :D
 
BMS, why not?

http://www.eminence.com/pdf/definimax-12ho.pdf

I'm putting the Eminence Definimax in a sealed box for a two way sealed design with the low end supported by sealed 15" Titanic subs for my early winter project. 1200 Hz and above will be the BMS 4550 with hopefully Altec 811B horns. I will use electronic crossover and equalization.

I'll be reporting to AK when complete. :music:
 
Redboy said:
Hey, you talkin' ta me?! :D
Click to expand...
Heh, heh.

YOU just gotta do some quality listening, is what.

[Then, we'll change stuff up.... :p: ]

Gibsonian said:
1200 Hz and above will be the BMS 4550 with hopefully Altec 811B horns. I will use electronic crossover and equalization.
Click to expand...
For ~$30 with throat adapters, try a pair of the JBL PT waveguides, too, or first, maybe.... :yes:
 
For ~$30 with throat adapters, try a pair of the JBL PT waveguides, too, or first, maybe....
Click to expand...

Well alright, I'm game. Now you've got to tell me where I find these waveguides.............search yielded nothing but JBL eng. doc. and JBL speakers with these waveguides as part of a speaker system - couldn't find any individual JBL waveguides. I'm not limited to 1 project, and I'm thinking compression drivers are in my long term future.
 
You must log in or register to reply here.
Back
Top Bottom