Hello
I have been given a pair of b139 oval drivers a pair of tweeters apparently pulled from b&w, I would like to use these plus a mid range driver something like a peerless 6 1/2 inch to build some speakers, anyone any ideas of size of cabinets or crossover design I would need for such a project, I am new to this. I have built a set of ed frias diy from audio review which I am really pleased with, but that is my total experience in speaker building
Regards Roger
I have been given a pair of b139 oval drivers a pair of tweeters apparently pulled from b&w, I would like to use these plus a mid range driver something like a peerless 6 1/2 inch to build some speakers, anyone any ideas of size of cabinets or crossover design I would need for such a project, I am new to this. I have built a set of ed frias diy from audio review which I am really pleased with, but that is my total experience in speaker building
Regards Roger
In the early 80's I built a pair of transmission line speakers from a kit from Wilmslow Audio based on a design by Chris Rogers and described in Wireless World magazine.
These used the B139 and a Peerless mid. If you squint very hard you can read the scanned article here:
http://www.tech-diy.com/pro9.htm
I don't know how this compares to modern designs but at the time my friends and I loved those speakers. Wish I still had them.
There was an earlier design by Baily in the same magazine which looks simpler to build. There is a scan of that article on the net some where but I can not find it just now.
Having said that I would worry about undertaking such a big project with such old drivers. They may be physically deteriorating now. Check them carefully and maybe measure their parameters for yourself.
These used the B139 and a Peerless mid. If you squint very hard you can read the scanned article here:
http://www.tech-diy.com/pro9.htm
I don't know how this compares to modern designs but at the time my friends and I loved those speakers. Wish I still had them.
There was an earlier design by Baily in the same magazine which looks simpler to build. There is a scan of that article on the net some where but I can not find it just now.
Having said that I would worry about undertaking such a big project with such old drivers. They may be physically deteriorating now. Check them carefully and maybe measure their parameters for yourself.
Design an enclosure in Martin King's worksheets. Forget using ye olde rule of thumb -you won't get anything like the same quality of results as you would from using the software. Anyone got the T/S parameters of the 139? I might be able to come up with an enclosure for you. If you don't want that, I suppose you could try going for one of the 'Alpha TLs', which seem to be the spiritual successor to Bailey's original 'Transmission-Line' misnomer. Again, I don't think these are in the least optimised, but a baboon could build them, they are so simple
You'll either need a mid, or a large tweeter that can go very low: these are bass, not mid-bass units, and they won't thank you for pushing them up too far. I would say definitely worth pursuing though. If memory serves, they'll need a hefty notch-filter somewhere as they have a break-up mode: can't remember exactly where I read that though.
You'll either need a mid, or a large tweeter that can go very low: these are bass, not mid-bass units, and they won't thank you for pushing them up too far. I would say definitely worth pursuing though. If memory serves, they'll need a hefty notch-filter somewhere as they have a break-up mode: can't remember exactly where I read that though.
Hi,
SM is correct in that the B139 has a really bad break up mode
(around 1K I recollect) which makes it difficult to use in a 3-way
design with a simple crossover crossing over to a mid unit.
Consequently its more sensible (if they are in good condition)
to use the B139s for a powered subwoofer or stereo subs.
From what I remember the TS parameters aren't that bass
alignment friendly and a sealed box is probably the best bet.
If you go for this a sub amp with bass boost option might help.
(reflex alignment - ~ 120 litres tuned to ~ 25Hz. Sealed ~ 60L.)
This leaves selection of the midbass unit. Probably best to take
an existing design and shoehorn the B&W tweeters in there
with an L-pad arrangement for impedance and level matching.
The tweeters should also be carefully checked for signs of
distortion - they can make odd noises on their way out.
/sreten.
edit : parameters for one version of the B139
Fs 25 hz
Mmd 43.5 grams
VAS 164 litres
Rscc 6.2 Ohms
Vl 0.56 Mh
Bl 12.3
Qms 5.5
Qes .4
Qts .37
Thse agree with the values given in Win ISD which adds
SPL 84dB
Pe 100W
SM is correct in that the B139 has a really bad break up mode
(around 1K I recollect) which makes it difficult to use in a 3-way
design with a simple crossover crossing over to a mid unit.
Consequently its more sensible (if they are in good condition)
to use the B139s for a powered subwoofer or stereo subs.
From what I remember the TS parameters aren't that bass
alignment friendly and a sealed box is probably the best bet.
If you go for this a sub amp with bass boost option might help.
(reflex alignment - ~ 120 litres tuned to ~ 25Hz. Sealed ~ 60L.)
This leaves selection of the midbass unit. Probably best to take
an existing design and shoehorn the B&W tweeters in there
with an L-pad arrangement for impedance and level matching.
The tweeters should also be carefully checked for signs of
distortion - they can make odd noises on their way out.
/sreten.edit : parameters for one version of the B139
Fs 25 hz
Mmd 43.5 grams
VAS 164 litres
Rscc 6.2 Ohms
Vl 0.56 Mh
Bl 12.3
Qms 5.5
Qes .4
Qts .37
Thse agree with the values given in Win ISD which adds
SPL 84dB
Pe 100W
B139 SP1044 (B version seems to have very sililar parameters)
From KEF spec sheet
Nom Impedance 8 ohms
Freq range 25-500Hz
Sensitiviy 87db
Total flux 1.1mWb
R(E) 7.2
R(e) 4.87
L(E) 1.04mH
L(e) 0.97
Sd 349 sq cm
Xmax 6mm pk-pk (B=7p-p)
X(limit) 12mm pk-pk
Moving Mass 43.5g (B=54g)
R(MS) 1.43 mech ohm
C(MS) 7.43x10e-4m/N
V(AS) 127 litresBl 12.5N?A
F(s) 25 Hz +/- 5Hz
Qm 6
Qe 0.39
Qt ).37
Closed box vol 30-60l
Reflex box vol 60-140 l
Power handling program 100w.
These units very robust and age does not seem to be a problem.
Roll surround lasts well.
From KEF spec sheet
Nom Impedance 8 ohms
Freq range 25-500Hz
Sensitiviy 87db
Total flux 1.1mWb
R(E) 7.2
R(e) 4.87
L(E) 1.04mH
L(e) 0.97
Sd 349 sq cm
Xmax 6mm pk-pk (B=7p-p)
X(limit) 12mm pk-pk
Moving Mass 43.5g (B=54g)
R(MS) 1.43 mech ohm
C(MS) 7.43x10e-4m/N
V(AS) 127 litresBl 12.5N?A
F(s) 25 Hz +/- 5Hz
Qm 6
Qe 0.39
Qt ).37
Closed box vol 30-60l
Reflex box vol 60-140 l
Power handling program 100w.
These units very robust and age does not seem to be a problem.
Roll surround lasts well.
Cheers guys. Assuming it's the SP1044, this thing is a doddle to use in a Quarter-wave resonator (TL)
Two options here: a Mass Loaded Tapered Quarter Wave Tube, and the next post will have a straight, Mass-Loaded Transmission Line. For more on these geometries Freo, if you haven't seen them before, look at the Projects section on Martin King's site: www.quarter-wave.com
The Project 2 ML TQWT shows the basic cabinet shape. Looks good in-room. If it's too tall for you, you can simply fold it in half, the top to the rear, and it will perform exactly the same.
ML TQWT first.
Line-length: 60"
Width: 16" internal at the base, 4" internal at the top.
Depth: 13.75" internal
Driver centered in the cabinet at 30" from the internal base and 30" from the internal top.
Use 0.25lbs ft^3 of stuffing from the top to just below the driver.
Circular, round port 3" wide by 2" long 3" up from the base.
Here's the projected frequency response:
You can cure the slight rise with either a touch of series resistance or a 1 long port instead. However, I deliberately aimed for a slightly falling response or this thing will become very bass-heavy in-room.
Two options here: a Mass Loaded Tapered Quarter Wave Tube, and the next post will have a straight, Mass-Loaded Transmission Line. For more on these geometries Freo, if you haven't seen them before, look at the Projects section on Martin King's site: www.quarter-wave.com
The Project 2 ML TQWT shows the basic cabinet shape. Looks good in-room. If it's too tall for you, you can simply fold it in half, the top to the rear, and it will perform exactly the same.
ML TQWT first.
Line-length: 60"
Width: 16" internal at the base, 4" internal at the top.
Depth: 13.75" internal
Driver centered in the cabinet at 30" from the internal base and 30" from the internal top.
Use 0.25lbs ft^3 of stuffing from the top to just below the driver.
Circular, round port 3" wide by 2" long 3" up from the base.
Here's the projected frequency response:
You can cure the slight rise with either a touch of series resistance or a 1 long port instead. However, I deliberately aimed for a slightly falling response or this thing will become very bass-heavy in-room.
Attachments
Scott, I have a pair as well for a project that got put on the back burner. I was thinking more for stand alone subs, and I got good results with an eight foot straight line with a 4-1 taper, but never got round to building anything. The full spec sheet is on P10's website, here, and my measured T/S parameters matched well with the published ones.
To be honest, I wouldn't run these at more than about 200Hz, but that does give a wide range of options for decent mids.
To be honest, I wouldn't run these at more than about 200Hz, but that does give a wide range of options for decent mids.
Thank you Sir, you are a Scholar and a Gentleman!
And here's the response of a straight, Mass Loaded Transmission Line. For more on these, Look at Martin's FE208Sigma and Lowther MLTL projects. They look like bass-reflex cabinets, but they don't behave like them!
45" internal height.
Internal width 11.5"
Internal depth: 14.35" (So & Sm=3Sd)
Driver at 33.75" up from internal base. (0.25 of line length from the top)
3" wide by 2" long port 3" up from the internal base.
0.25lbs ft^3 of stuffing from the top to just below the driver.
Note: with both of these designs, you can put the assumed midrange and tweeter wherever you like, within reason. They shouldn't affect the cabinet etc. In the case of the MLTL, I'd place them above the B139. Actually, if you did that, and were using a midrange unit, you could lower the B139 to 22.5" from the internal base of the cabinet, which would remove the touch of ripple you can see. Again, I was aiming deliberately for a slightly drooping response to compensate for room-gain.
In the ML TQWT, unfolded, I'd probably place the mid and tweeter below the B139. Folded -your call. These MathCad sheets of Martin's are staggeringly accurate by the way. This is exactly how these designs will perform in anechoic conditions, assuming they are built correctly and the drivers are working as they should. However, as our friend has above mentioned, the bit you want to be paying attention to in both cabinets and graphs is below around 200Hz: above that, you'll need a decent midrange unit. That's fine: place it in a location like those I suggested above where there is some stuffing and that should flatten its behaviour nicely!
All the best & I hope some of this helps a bit
Scott
And here's the response of a straight, Mass Loaded Transmission Line. For more on these, Look at Martin's FE208Sigma and Lowther MLTL projects. They look like bass-reflex cabinets, but they don't behave like them!
45" internal height.
Internal width 11.5"
Internal depth: 14.35" (So & Sm=3Sd)
Driver at 33.75" up from internal base. (0.25 of line length from the top)
3" wide by 2" long port 3" up from the internal base.
0.25lbs ft^3 of stuffing from the top to just below the driver.
Note: with both of these designs, you can put the assumed midrange and tweeter wherever you like, within reason. They shouldn't affect the cabinet etc. In the case of the MLTL, I'd place them above the B139. Actually, if you did that, and were using a midrange unit, you could lower the B139 to 22.5" from the internal base of the cabinet, which would remove the touch of ripple you can see. Again, I was aiming deliberately for a slightly drooping response to compensate for room-gain.
In the ML TQWT, unfolded, I'd probably place the mid and tweeter below the B139. Folded -your call. These MathCad sheets of Martin's are staggeringly accurate by the way. This is exactly how these designs will perform in anechoic conditions, assuming they are built correctly and the drivers are working as they should. However, as our friend has above mentioned, the bit you want to be paying attention to in both cabinets and graphs is below around 200Hz: above that, you'll need a decent midrange unit. That's fine: place it in a location like those I suggested above where there is some stuffing and that should flatten its behaviour nicely!
All the best & I hope some of this helps a bit
Scott
Attachments
Hi Scott.
Just a quick question, as you're the Worksheet Wiz.
Would the simulated behaviour of the line change at all in the TQWT if the driver was mounted on the tapered side, and can the software even differentiate?
I'm sure it wouldn't make a difference, but MJK's work has caused me to question my assumptions on TLs many times before, so I'm now back to asking beginner questions!
Just a quick question, as you're the Worksheet Wiz.
Would the simulated behaviour of the line change at all in the TQWT if the driver was mounted on the tapered side, and can the software even differentiate?
I'm sure it wouldn't make a difference, but MJK's work has caused me to question my assumptions on TLs many times before, so I'm now back to asking beginner questions!
pinkmouse said:Hi Scott.
Just a quick question, as you're the Worksheet Wiz.
Would the simulated behaviour of the line change at all in the TQWT if the driver was mounted on the tapered side, and can the software even differentiate?
I'm sure it wouldn't make a difference, but MJK's work has caused me to question my assumptions on TLs many times before, so I'm now back to asking beginner questions!
Hmm. A very good question. Not in the current worksheets is the answer. That will change a bit around March though when the updated sheets emerge. Until then... Well, I'll do my best, though I'm still learning myself.
I wouldn't expect the behaviour of the line itself to alter much, whichever face you put the driver
Mounting a driver on the trapezoidal face like in Martin's Project 2 ML TQWT (and the layout I think would be best in this case) is a much better, albeit more 'interesting' idea than sticking it on the sloping side in my view, and I think your main query, correct? Now that will almost certainly have an effect on the speaker's response, not because of internal changes but because of differences in baffle-step. As it is a different size and shape to the other panels, it's probably going to need different values in the correction circuit.
Best
Scott
Hi Freo,
Thanks for the tip on the Harry Manx cd, I'm a big fan and have caught his shows here in Australia a couple of times now. I used the B139 driver back in the 70's in a transmission line that used the old 'Concerto' kit with the B139 flat on the top of the speaker facing straight up.........they sounded pretty good at the time from memory. Drop me an email and I can give you some info on this design if you're interested.
Cheers,
Thanks for the tip on the Harry Manx cd, I'm a big fan and have caught his shows here in Australia a couple of times now. I used the B139 driver back in the 70's in a transmission line that used the old 'Concerto' kit with the B139 flat on the top of the speaker facing straight up.........they sounded pretty good at the time from memory. Drop me an email and I can give you some info on this design if you're interested.
Cheers,
That could work really well. Nice idea! You'd just need to juggle the mid-dome pod enclosure so it could get down low enough. That should be easy to do. Actually, now I've done some checking, I think that the 139 could be OK up to around 800Hz or so, which should make life easier for whatever it's crossed over to. And maybe even allow the midrange to be dipolar too!
Cheers
Scott
Cheers
Scott
Scottmoose said:
My thoughts exactly. I have some nice Fostex mids I might use, I was going to give two away, but now I'm having second thoughts. Now all I need to do is blag some tweeters from somewhere, and the drivers for this project will have cost me nowt!
I'm not sure about going up to 800Hz, but if I can make it up to 500Hz, then I'm sorted.
Dimensions of the tower are exactly as stated in Scott's ML TQWT design on the last page. I was thinking of hardwood for the front curvy baffle but thick MDF or ply would be cheaper and easier to work with. The dimensions of that part will change depending on exactly which drivers you work with. I haven't run any sims yet, just sketchpad ideas at the moment.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.