Building a big two way system, a homage to the KEF BBC LS5/1A

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Always up for a challenge have recently been considering this proposition

Here is the original which used a Goodmans C129/15pr in the A version and two Celestion HF1300 tweeters.

145824mywmyyd3oyoudl0h.jpg


For a clone one could use a pair of Audax TW034 X0 tweeters, but for the Bass? which 15" unit can operate up to 2K?
 
Last edited:
For the LS5/1 and LS5/1A the following information is given in the BBC document ENG INF No. 43, from which I quote:

"LS5/1, and LS5/1A
The studio versions of the LS3/1 and A, designed to replace the LSU/10. The cabinets were larger than the LS3/1 and the tweeters were mounted above the woofer, rather than in front of it.
The following units were used:
Plessey (LS5/1) or Goodmans C129/15pr (LS5/1A), 15-inch paper cone woofer
2 x Rola-Celestion HF1300 tweeters. "



KEF /BBC DESCRIPTION FOR LS 5/1AC:
BBC MONITOR SPEAKERS
The monitor loudspeaker used by the BBC to judge the quality of its radio and television programmes has been developed by a highly experienced team of research workers and studio operations engineers over many years. Infact the model in current use is the result of thirty years continuous experience in operational broadcasting in which period the best brains and ears in Britain have contributed to its evolution. The basic conception of the BBC Monitor loudspeaker is quite different from most domestic high fidelity speakers in that the latter are usually only intended to give pleasing reproduction of a diversity of available programme material. To this end reasonable variations in tonal balance and colouration are considered admissible or even desirable if they produce aesthetic pleasure and it is not generally possible to control these effects sufficiently to ensure consistent performance between samples.

On the other hand a monitor speaker must produce the programme signal with the greatest possible accuracy without modification of tonal quality and devoid of colouration. Broadcasting and recording organisations use high quality speakers to ***** the aesthetic and technical merits of their programmes and guide them in obtaining suitable microphone placing and studio conditions. It has been found by experienced engineers around the world that monitoring on mediocre loudspeakers can result in poorly balanced programmes due to faulty microphone technique as well as to allow technical faults to go unnoticed.
The BBC monitor Loudspeaker represents the best possible compromise between all the various requirements in the present state of the art added to which it has been tested at every stage of its development by subjective comparison between the reproduced sound and the original. In this regard this loudspeaker is virtually unique since no other organisation has access to such splendid facilities for both subjective assessment and technical analysis.

DESCRIPTION
The speaker system employs three units, which together cover the audio range available from present day records and broadcast programmes. The enclosure is constructed from high-grade veneered chipboard of a quality, which has been found by experiment to possess the smallest degree of self resonance of all suitable construction materials. The structure is reinforced by internal partitions which restrict the vibration of the back and sides and also by a metal strut between the front and the back panels. Mudocel damping is applied to the top panel.
The internal volume of the enclosure is 4.7 cubic feet and a small vent resonating with the volume of about 50c/s is used to give a slight boost to the low frequency output. Standing wave effects are damped by internal pads of absorbent material.

LF UNITS
The lowest frequency unit is a 15 inch paper cone type with a 3 inch diameter voice coil and PVC roll surround. Its axial frequency response extends smoothly up to about 2 Kc/s and it is relatively free of colouration effects. The sound output for this unit radiates through a vertical slot 7 and a half inches wide to ensure good horizontal dispersion up to the crossover point with the high frequency unit s at 1750 c/s.
HF UNITS
Two identical HF units are fitted each having a phenolic impregnated diaphragm. The frequency response extends smoothly up until 13kc/s above, which it dies away in a regular manner. Tests have shown that the output to be free from prominent low damped resonances. The two HF units are mounted in a vertical line above the LF unit in such a way that the separation between high and low frequency sound sources is not noticeable to listeners located over four feet away. Both units are operated in parallel at middle frequencies to increase power-handling capacity in the region close to the crossover point.
Above 3Kc/s the input to the upper HF unit is reduced to induce dispersion in the vertical plane and the phase is retarded as to direct the combined radiation pattern upwards

CROSSOVER NETWORK
A thirteen element printed circuit crossover network incorporates facilities for equalising minor irregularities in the LF unit as well as providing for adjustment for relative levels of LF and HF units in 1 DB steps. An additional network corrects the diffraction effect at low frequencies due to the finite size of the cabinet, as well as the low frequency loss due to the motional impedance.

ALTERNATIVE CABINETS
There are two versions, one a floor standing cabinet for studio and control room work and a suspended model for use in television control rooms.
The floor standing speaker model LS5/ 1A is intended to be operated on a 15 inch plinth above the floor which positions the HF unit at optimum listening height.
The plinth is designed to accommodate the equalised power amplifier conveniently beneath the speaker enclosure.
The suspended model is equipped with metal suspension gear, which enables the entire speaker to be tilted downwards at the required angle The LS5/2A requires additional equalisation at low frequencies to compensate for the loss of floor reflections. A suitable equaliser is available which can be incorporated with the power amplifier if required.
Both cabinets are available in a choice of oiled teak veneer or hammer grey lacquer.

SPECIFICATION

SIZE: LS5/1AC 33 X 19x 17 inches
Weight: 82 Lbs
Nominal Impedance 25 Ohms
Maximum input: 35 watts R.M.S
Frequency response : 40 –13 K/cs +_ 5db
Directivity index: Better than 54 db up to 3 Kc/s reaching 7db at 10 Kc/s
Calibration Accuracy: The axial frequency response checked against a BBC approved standard sample is adjusted so that the curve of the loudspeaker under test does not differ from that of the reference loudspeaker by more than the following amounts:
50 c/s - 200 c/s +_ 1.5 db
200 c/s – 400 c/s 0db
400 c/s – 13000 c/s +_ 1.5 db
13000 c/s- 15000 c/s +_ 2 db - infinity
The reproduction of the Loudspeaker under test is also compared with that from the reference speaker using high quality programme sources and white noise.

LF Unit: 15 inch Heavy paper cone with plasticised P.V.C. Roll surround.
Fundamental resonance: 25 c/s
Flux density: 9000 oersted on a 3 inch dia centre pole
Total Flux : 16000 maxwells

HF Unit: 1.5 inch diameter direct radiator with Phenotic impregnated - fabric diaphram
Flux density : 10,000 oersted
Total Flux : 25,500 maxwells

Crossover Network 13 Element printed circuit network employing Radiometal - - cored inductors and closed terminal metallised foil capacitors. - Completely enclosed and shielded in a steel case.

Input Connections : Canon XLR3 . Other connectors can be fitted to suit req.

Plinth: PLS5/ 1A Constricted from heavy plated steel finished in - - satin chrome with amplifier section brackets and heat shield.

Equalised Power Amplifier: Type MA40/LS5/1a . Fixed bias push pull amplifier delivering - 35 watts output over the entire audio range . The input
circuit is arranged for bridging 600 0hms balanced line at
- zero level and incorporates equalisation for the HF loss due - to roll off in the upper high frequency above 3 kc/s An input - attenuator is fitted giving infinitely variable control of - sensitivity. Output impedance of 25 ohms.

Input Connections P.O . Jack No. 500B6 in parallel with Painton 4 pin socket
Type 310035

Output Connection : Terminals

AC Mains Supply 110-140V or 220-250 V 50 c/s adjustable


Speaker units
LF unit : 15 inch (Unidentified)

HF Unit: Very powerful KEF T52. Same as fitted to the KEF KM1 studio monitors.
 
Last edited:
The Volt driver looks like it's very adequate. It's good for a BR speaker ~86L@34 Hz (as a starting point/alignment). 2π: 95.6 dB/2.83V/m, 94.6 dB/W/m; 4π: 91.6 dB/2.83V/m, 90.6 dB/W/m.
Volt RV3813 15" 500W 8 Ohm Radial Chassis Loudspeaker Driver 316.99
Volt Loudspeakers - R3813
It's going to be a difficult project if not using the same drivers. Why? Vintage looks vs. nostalgic vintage sound ??? Does it have anything you classify as good besides the design problems and intrinsic no-no's.
What tweeters are you looking at vintage or recent? Some links/ideas.
Seas T35C002 E0055-06 Tweeter - Excel Series
E0055-06 T35C002
Seas 27TFFNC/G H1396-04 Linkwitz LX521 tweeter
H1396-04 27TFFNC/G
SB Acoustics SB26STCN-C000-4 Tweeter, Free UK Delivery.
SB Acoustics :: SB26STCN-C000-4
 
With such a large enclosure, the Eminence kappalite 3015 would be my choice to 2khz. It will need a notch filter between 1-2khz but the OEM crossover count at 15 pcs indicates there's a notch in there somewhere as well. The low pass after the notch could be as simple as a 2nd order electric to work well with the natural acoustic rolloff.

IMO, the tweeter limitations of the day indicated the need for two tweeters......and given what we know now about combing, I think a single modern day 1 1/8" dome would be better served. It won't be a clone, but I believe the system would perform better overall.
 
Hi,

Just a very bad idea to attempt to clone nowadays, speaker
design has moved on leaps and bounds since that very self
serving description of the speakers merits was written *.

FR : 40Hz –13 KHz +/- 5dB : is poor by any standards.

This will utterly slaughter it in all respects :
http://www.zaphaudio.com/SB12.3/
your looking at 40Hz to > 30KHz +/- 2dB.

As will this :
https://sites.google.com/site/undefinition/tarkus
and this :
http://www.zaphaudio.com/ZDT3.5.html

rgds, sreten.

* No doubt it was efficient and didn't need much power to
go loud, but its no better than modern good PA speakers.
 
Last edited:
By the way, (and I agree with the previous comments) I was only giving links in the UK for the OP. To design a speaker you need to know what you are doing. To clone... well chose your poison: 1. you use the same drivers (xover/alignment/net.vol/speakers) or 2. you are building a new one (speaker)! Bet "all" are happy now (more food for thought). :)

I don't disregard the original idea but I have serious doubts that OP wants to take it further... :D

To clone a problematic design is not a good idea unless you want to listen to the same obvious mistakes reproducing the sound and building the speaker. And it has a longer learning curve, reproducing/diy the bad (mistakes) first, as in a learning curve. :eek:

Look at the good examples of speakers instead (like the Kef's Concertos) for vintage, try to understand why they were good. Or to the good designs of this late years like Troels-DK for one good example (various), if not just build a kit.
Loudspeaker Kits: Stand Mounted Loudspeakers, Floorstanding Loudspeakers, Sub Woofer Kits
 
I have cloned the Spendor BC1 the "Dorspen" and kind of cloned the Decca Corner Horn. But all designs are set in their time and context. The BC1 is a unusal 3 way design in that it is a classical two way design crossed at about 3 kHz and then have a third driver above 10 kHz. At the time they could find tweeters they were happy with all the way from 3 kHz to 20 kHz, but later designs scrapped the supertweeter as it was no longer needed when better domes was avaible.

It is my understanding of the BC1 that the main "magic" component is the bass/midrange driver and crossover. The second thing is damped thin wall cabinet. The fact that the system happen to be a 3 way system with a supertweeter is not critical. Nor that it is a bass reflex box. Nor that the bass is rearmounted. Nor that the panels are detatchable.

The Decca Corner Horn claims that the cabinet flexing is critical for the damping of the system and that the addition of damping material spoil the system. My finding was that flexing was not the reason why a totaly empty cabinet fared quite well, it rather had to do with geometry of the nonparallel walls and various tapers. Adding damping material both sounded and measured better.

So to clone you have to have both experience and measurement tools and a good understanding of the different interacting factors. When I started building speakers a 20 Hz highpass filter was allways needed in the amp as all records has some warp and rumble and the fashion at the time was to have pickups with very soft suspensions so the tonearm resonanse was too low and too undamped. This was to protect the loudspeaker especially open systems like bass reflex or pipes. With the arrival of the CD in the 1980s the need for a rumble filter was gone. There is no way that I could properly clone the Spendor BC1. I can take some elements of the design and incorporate it in a new design but then I need to understand what I should try to pick and also understand the limitations of what I also can pick as opposed to what I wish (like the 8" driver).
 
To 'clone' per-se means that identical drivers would need to be used as part of the project
If not ones that very closely match the properties of the so called cloned speaker.

I like the challenge but as a paper exercise to see if it is possible to replicate a particular design. I achieved this in part with my X-1 project which was featured in Hi-Fi news in April 2002 page 58~

It is interesting that since I have looked at the LS5/1a design I have found a couple of drivers that come close nothing that can be used to clone so that design has morphed into something different.

My X1 took the original in a different direction so cannot be classed as a clone.
 
Hi - just joined!

I am about to embark on a similar project, but maybe from the other way round. I have the baffle from an LS5/1, complete with Plessey (I assume) 15" bass unit, 2 GEC tweeters, and an FL6/2 crossover network. What I lack is the box. The bits I have work and I am listening to it now. I have scaled the box from photos online, and the dimensions given in the KEF data for the LS5/1a, and they match the dimensions of the baffle.

The sad thing is that I had a complete working LS5/1 some years ago, from a skip at work, but after a couple of years sold it to someone who had another one because I assumed I wouldn't be able to get another.

What I don't remember from the complete one I had was whether the box was chipboard or plywood - the LS3/1s I have are plywood, and whether the panels were damped and braced?

I am about to buy several sheets of 18mm OSPB to make cupboards for the workshop and could knock up a box from this to try out.

I don't really need another large speaker but since the bits were a bargain from ebay, and close to home too, it seems bad mannered not to!
 
There are tons of arguments for MDF or plywood and plenty of heated arguments around. As far as published measuremnets goes, the difference soundwise is small. Bracing, damping and laminations plays a greater role than the acutal box material. Plywood is vastly superior in taking abuse in field, mobile use but for sitting still in a dry room...

Why not make a box in the OSB leftovers and take it from there.
 
Hi - just joined!.

What I don't remember from the complete one I had was whether the box was chipboard or plywood - the LS3/1s I have are plywood, and whether the panels were damped and braced?

Plywood. I had a pair I got from KEF and they were plywood. Fairly thin and heavily damped by a thick and heavy felt layer, if I recall. There might have been a pair of side braces vertical on the left and right sides.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.