Can one build a better (non) LS3/5A speaker based on T27s & B110s?

Another in the long list of 110/27 variants, in this case from Jim Rogers himself. Crazy crossover complexity but imaging monsters.

Very Charlin speakers like !

an another design with the t33 instead t27 but may certainly work without Dapolitto : Rewind Audio: KEF Reference 104/2 Floor Standing Speakers - BLACK (I insist but maybe Off topic... yes !)

Notice the little tweeter horn maid in the front baflle as for the b110 as well. Should work without the MTM and complex 104/2 ref crossover design (both b110 have not the same crossover shematic). The edge bafle is just good. But the rear load of the mids is not not classic... very smal space between the bottom of the B110s and the cabinet !

like the soundstage of this speaker with almost same drivers of the OP :eek:
 
surv1v0r,

You forgot to make one important question to some:

"Have you ever auditioned some outdated, suitable only for playing in a van, LS3/5a variant?"

I'm saying this, because two 35yo "imaging monsters" of post#83 are smiling at me right now. And quite a lot of speakers paraded in front of me all these years, but the JR149's remained.

So, are their drivers outdated? You bet!
Is the speaker as a whole outdated? on it, DON'T bet!

Using the same drivers, can you better an LS3/5a?
VERY improbable, sorry. Cause this is a pretty sophisticated design, to better it (hmm...), sound understanding of the principles involved and metering equipment is required, as the helpful and mostly the unhelpful posts have already indicated.

What you should do IMO?

1) Check if I'm talking nonsense: go find a pair some way, and listen to it.
2) Go to Falcon's - they really know their stuff.
3) Build the crossover designed for a given enclosure exactly as proposed. Build the enclosure exactly as proposed.

What is important, understand the limitations: these are speakers that don't stand a lot of power, and will deliver compressed sound at very high levels. I would not go for a vented enclosure, and would be quite skeptical with a TL (rather than Atkinson's I would seek for help with a modern MLTL in this forum).
And of course, there is this spec-shift problem due to aging. You need equipment and experience to address this. So, go for a closed box, it is very forgiving.
As for the lack of bass, use a sub (I do) and cross them low! Or even better, B139's in separate closed (that's adequate, don't worry) boxes. 70-90Hz is ideal to relieve the strain of the B110 and preserve it's fabulous midrange, at the same time providing low extension that no BR design will. I know, that's becoming expensive, but I do think that this is the way to go. Done right, man these LS3/5a's DO play music
 
Last edited:
Oh yes, they do indeed! And can go much louder that way.
But then you need two quality subs, fast and musical at 150Hz, and must be very careful with the emphasized 140Hz response of some variants to achieve a seamless matching

Oh, I forgot!
surv1v0r please forget the Crestas, Codas or whatever they call them
 
Last edited:
surv1v0r,

You forgot to make one important question to some:

"Have you ever auditioned some outdated, suitable only for playing in a van, LS3/5a variant?"

I'm saying this, because two 35yo "imaging monsters" of post#83 are smiling at me right now. And quite a lot of speakers paraded in front of me all these years, but the JR149's remained.

Those were pretty great speakers for the time, a real classic and my first foray into great sound many years ago. However, IMO, my redesign of the 149 with the Scan 9000 in the same cabinet using the B110 and some additional reflection control smokes the old JRs (sound and measured) and loses none of their good qualities. Not to knock the JRs, it did take many hours of measure-sim-listen-repeat to do this.

But as other posters mention, there are many drivers out now that provide cleaner sound, more dynamics, more even spectral balance in small boxes (neutrality) than the B110. And no way around it, Toulou is right, the T27 just won't play very loud. At least to my ear. Best example that I would consider in the lineage of the B110 but just bests it in every way possible is the Tangband W4-1720

Anyway, its just my opinion, but I've done the time, designed, measured and listened over countless hours but I know I'm probably just stirring up a debate so I'll cut it short.
 
...
So, are their drivers outdated? You bet!
Is the speaker as a whole outdated? on it, DON'T bet!

Using the same drivers, can you better an LS3/5a?
VERY improbable, sorry. Cause this is a pretty sophisticated design, to better it (hmm...), sound understanding of the principles involved and metering equipment is required...


DDF,

Hey, no debate here, I am pretty conscious of reality, see above.

But the TO makes it clear that his understanding of speaker design is not very advanced, asks how enclosure and crossover correlate, finds software like edge difficult to understand, and asks for advice:
"I have a set of drivers from a KEF Kit3, can I achieve something better than an LS3/5a with them?"
Well... very improbable.

Besides, you say it all:
...it did take many hours of measure-sim-listen-repeat to do this...
...and you used a Scanspeak 9000 to achieve it.

Well, surv1v0r will not repeat this, not today.
Instead of telling him "sell everything and buy this or that fantastic driver and design something from scratch", some proven design better than the LS3/5a family using the same drivers (if such a thing exists) should be proposed to him.
Or at least, "sell everything and build THIS proven design buying this or that fantastic driver".
Or... "build the best LS3/5a clone you can, following advice from people who know this stuff". The end result will be maybe not the best, bet still a very very listenable speaker, even by today's standards, take my word for it.

But "just good for listening in a broadcasting van" ???
Give me a break!
 
Last edited:
Toulou, well said.

I have the design info for the JR149s (stock). The front grill you can reproduce using metal gutter block from Home Depot, its the same curvature and construction as the original JR149, but the cabinet would be a trick to try and reproduce, I wouldn't know where to reliably start.
 
I have the design info for the JR149s (stock). The front grill you can reproduce using metal gutter block from Home Depot, its the same curvature and construction as the original JR149, but the cabinet would be a trick to try and reproduce, I wouldn't know where to reliably start.

I was a bit miffed when the JR149 came out because I had DIYed a similar speaker a year or two earlier and the Rogers speaker rather reduced its impact. Mine had a large radius on the top and bottom panels rather than sharp edges since reducing diffraction was a significant motivation in the design. Although advised against it (which of course spurred me on) I made the cylinder by casting resin between two thin metal sheets to allow for a bit of movement when going off. It worked reasonably well but I would have had better control using fibreglass but I wanted to do something different. A rubber sheet between two layers of fibreglass might have been a good next step had I not dropped home audio as an interest soon after.

Before the OP asks, I used an external active crossover which made no sense in those days in terms of cost but I wanted to play around with op-amps as many did in the 70s. After a couple of failed power amps and getting in dispute with the component supplier I never got the electrical side properly sorted before dropping interest in home audio when the audiophile thing kicked off in the late 70s.
 
No B110 and no T27, but perhaps you like: WA LS3 Bookshelf Loudspeaker
I'm sure that that speaker is great. However, it is pretty much off-topic.

I happen to have a spare pair of 40 year old T27s & B110s; I would like to build them into small(ish) cabinets. I suspect that whilst the LS3/5a design may have been ideally suited for its original application it might not be the best possible solution in a domestic environment. If I thought otherwise I would certainly buy the respected Falcon Acoustics LS3/5a kit immediately :)

With help from Colin, I have discovered and am now entirely focused on the P Atkinson Mini-Line design and all I am interested in is the design of a suitable crossover.

  • The well braced 12 cms ply baffle dimensions are approx. 40 cms x 20 cms.
  • The T27 is mounted centrally 7.5 cms from the top.
  • The B110 is mounted centrally 20 cms from the top.
  • My desired crossover point is at 3,500 Hz.
The original design uses the B110 so the transmission line should be suitable for my bass/mid drivers. I do not have the experience or skills to use one of the many crossover design tools about which people write so glowingly.

Once again, I accept that this is not the best place from which to start. However, it is where I am.
 
Go ahead.
If you end up with a listenable small speaker, then you have won.
But what are the chances of ending up with a listenable speaker never mind the stated objective of better than the LS3/5A?

I can only go by the first page of the article since the scribd site will not show me the second and third pages. Does anyone have an alternative link? The Atkinson mini-line transmission line is an odd looking thing. Perhaps there are reasons others have not followed this "state of the art" design.

"(a) Cabinet rigidity is fundamental to low colouration." Clearly this dictates 6 mm thick walls doubled up to 12mm in places.

"(c) The relationship between cross-sectional area and length of the line (assuming rigidity of cabinet) can affect the need for damping in the line." Damping in the line is part of how a transmission line speaker works but damping is not needed in the port of a bass reflex speaker.

"It can also be argued that the use of laminations of different densities will improve damping". I would like to see such an argument given the role of mass in damping. Perhaps it is at the top of page 2?

By not damping the line/port the design would seem to want to use the resonance due to the mass of the incompressible slug of air in the line/port and the spring due to the volume of air trapped in the main enclosure. Does anyone know the effective port tuning for this design? The compressible quarter wave resonance has presumably been pulled down low enough in frequency to also do something useful for the small 5" driver at low frequencies.

A significant concern would be the loud harmonic resonances due to not damping the line which will radiate from both the cone and the line/port. Another might be the relatively poor transient response due to using resonances to boost output. Are there any measurements to suggest things might be under control?
 
But what are the chances of ending up with a listenable speaker never mind the stated objective of better than the LS3/5A?<snip>
Hard questions to answer but worth a try and thanks for your comments which to be honest I don't understand but perhaps they will elicit feedback from some of the knowledgeable posters here? :)

My start point for this project was the LS3/5a. However, reading around suggested that that design would probably have benefited from greater volume, a path that was probably eschewed because of size / space considerations - the LS3/5a is a small speaker.

Someone here suggested the Fris Daline + B110 which I have seen described as an "LS3/5a with bass", unfortunately at nearly three feet tall, they were larger than I wanted.

I was also pointed in the direction of the Atkinson Mini-Line which at 18" is much nearer what I am looking for.

  • All the walls are 12 mm thick (2 x 6mm) and they are extensively braced with 6mm internal walls and with dowel rods.
  • Approx. two ozs. of long haired wool is used in each decoupling chamber but not in the transmission line, I guess that one could experiment here.
  • There is no attempt to justify the assertion that "the use of laminations of different densities will improve damping" but I get the feeling that this technique has been tried in other designs? I have considered taking this idea further with 6mm MDF internally and 6mm ply externally for the cabinet but will probably stick to the original plan.
Maybe the whole thing will be a disaster; who knows? If it is, I guess I'll just have to resort to the LS3/5a after all.
 
You'd need 10" PVC pipe... not really easy to get -- at least in small quantity and cheaply unless you can score scraps from a pipe laying job near at hand.

dave

One of the UK magazines produced a design using similar Dalesford units. For the cyclinder, they used 10 inch diameter (or thereabouts) concrete pipe. The drivers were mounted on a wood baffle but I forget how they cut the pipe to fit it. (PS The designer cut the pipe in half - the cross-section was a D shape.)

Concrete pipes have been an old standby of the UK DIY scene since Hi-Fi News ran a design in the late 50s.

Returning to Mini-Line and Daline, the latter had pretty minimal damping in the line and there are some writers who consider the transmission line to be closer to the reflex speaker than the labyrinth designs of the 50s. PMC seem to use profiled foam to line the walls of their TLs but nothing in the centre of the line - at least from the couple of cross-sectional photographs I've seen.
 
Last edited: