Rogers LS3/5a (BBC) Speaker Review

[DSJR]

I couldn't find the exact damping factor of Quad 303, however many mention that it was very low. Probably, it was still better than the tube amp...

The 303 is capacitor coupled to the speaker and I believe it goes to pieces below 20Hz which may help explain why the matching 33 preamp was rolled off quite quickly. HiFi News did an appreciation with measurements and the reaction with the 57 electrostatics.

 

[Waxx]

The Quad 303 is a 1968 class AB design with bipolar transistors, so it's output impendance is relative high and damping factor relative low to modern amps. This is the electric schematic i got from a repair I did long ago:

It will enchange the bass a bit, not like tube or class A amps, but neutral sounding is certainly not the right description for this little amp. It does sound well altough, and the 15R nominal impendance of the LS3/5A will reduce that bass boost a bit.

edit: btw: I just found this: https://rubli.net/classic_amps/files/amps/quad/33and303/test.htm

 

[MaxwellsEq]

I may find it difficult to be impartial here, but it might be useful to put the cards on the table first. I worked in the BBC for 35 years as a sound mixer, sound editor etc. A large part of my life was in Outside Broadcasts trucks (small vans really - Commer FC Van ) using the supplied 3/5a as my daily speaker. The ear to speaker distance was about 3 feet max and the speakers were normally mounted with their backs towards the rear doors of the vehicle with the mixer/operator facing the rear - there was no sound deadening, only a simple cloth curtain to pull across the doors at the back.

The design brief of the LS 3/5 was to match where possible the mid-range sound of the BBC's larger 'studio' speakers such as the LS 5/5 and later LS 5/8 and the larger Outside Broadcast speaker in use, the LS 3/7. As DSJR has pointed out the amplification was either Quad 50D, Quad 303 or the HH mono amps. It's important to note that the equipment was left in the van in all conditions, never removed, no matter what the temperature or humidity was doing and travelled on the wall mounts in position (never buy a secondhand BBC 3/5a). There is some thought that the decision to use a plastic (Bextrene) for the B110 cone was the alleviate some of the problems with humidity and temperature found in a paper cone based driver. Because swapping out a spare from stock was a possibility, keeping minimum variation in production samples was essential as it was impractical to rely on 'matched pairs' coming from stores when out on the road. For that reason, production tolerance was part of the BBC Spec for the 3rd party companies.

Its primary role was to monitor speech or dialogue based audio intended for FM reproduction (15kHz bandwidth) with a restricted dynamic range of about 20dB - most speech if manually mixed was only 10dB or so. I was told informally that the reason for the bass 'bump' was to make it easier to identify mic plosives, bumps & wind noise rather than give an illusion of extended low frequencies on music. So yes, they were 'posh' Auratones.

The bextrene and surround will have aged with time - there is a known problem with cone sag, the BBC used to have a maintenance schedule to rotate the drivers through 180º every few years. The dampening dope applied to the cone also ages - can change colour to white too. Who knows was the caps in the crossover measure now.

In the case of the reviewed 3/5a, there is no way it is still going to be near it's original designed spec, so it has to be reviewed like buying a secondhand car - not even a restomod. It also very likely that the original design spec does not tally with today's requirements or tastes too.

What @amirm is doing is reviewing a 40 year old speaker and seeing how it stacks up against modern competition - a bit like comparing a late 1970's GP car against today's Red Bull of Max Verstappen. Both have four wheels and an engine and go around a race track. You would still have to pay a significant amount of money for Jody Scheckter's Ferrari 312T even though the Red Bull RB19 will be significantly quicker.

I'm sure the original members of the BBC Loudspeaker Design committee would be having a chuckle at this discussion though what they did at the time was very serious and ground breaking. I was on the distribution list for the minutes of the committee and some of the minutiae and detail was bewildering at times. The only design tools they had was simulation, scale modelling and listening tests and a slide rule or two. The general quality and consistency of off the shelf loudspeakers at the time was pretty dire. The commercial sales of the 3/5a potential was never a thought - I don't think.

I do have a set of 3/5a kept original and not used and a modified pair with the crossovers removed and the T27 tweeters replaced by SonAudax HD12d25, actively driven via a MiniDSP and Musical Fidelity power amps (inspiration from the Siegfried Linkwitz 1977 Wireless World system) the bottom end is provided by an original Rogers LSB1 subwoofer. I have no intention of changing things as it suits my very small living space.

Your post is the most useful here. There's a surprising number of people who are looking at the current (daft) prices and not spotting the device being tested is a very old loudspeaker that will absolutely not measure or sound like it did when it left the factory.

You are the exact "customer" for the speaker as it was designed. The designers would be astonished at what has happened since the 70s!

 

[SteveC]

nearfield for voice in a van, isn't that their original purpose ?

Yes, and not for full range monitoring, but for voice. The spec was not so much for sound, but for ensuring that you could pull out a bad one, put in a new one, and it would sound basically the same as the old one. I owned a pair for a while many years ago and I always liked them best up against a wall, but not well enough to keep them forever. I'd love to see a second set of measurements taken with them on a listening desk and backed up against the wall, but I know that's not what Klippel is all about.

 

[amirm]

Your post is the most useful here. There's a surprising number of people who are looking at the current (daft) prices and not spotting the device being tested is a very old loudspeaker that will absolutely not measure or sound like it did when it left the factory.

You are the exact "customer" for the speaker as it was designed. The designers would be astonished at what has happened since the 70s!

That is a hypothesis, not fact. The fact that the companies that produced these never made any anechoic measurements of them says that you also don't know if any of this new were a) like the BBC and b) there was no variability in them. If you are going to clone a speaker, these are the mistakes you don't want to make. Every video I watch of these companies talks about components/parts, etc. There is little emphasis on measurements and proper ones of that.

Before publishing this review, the owner and I scrutinized a lot of measurements to build confidence in this sample being representative. While this is hard given the less than ideal protocol uses by others, the correlation remains good. Here is Hi-Fi news gated measurements from 2019 of Rogers LS3/5a:

It is cut off unfortunately at 200 Hz due to gating but you can see that broad dip. You also see the resonance peak around 1.1 kHz. And tweeter level that is higher than that bass dip. Overall, it looks poor so the conclusions would be the same as one thinks of my measurements.

This is stereophiles gated/stitched measurements of a 1978 sample: https://www.stereophile.com/content/bbc-ls35a-loudspeaker-1989-measurements

Again, we see good correlation with the same dip and elevated tweeter response.

So what you see is likely similar to what you buy today, or what you bought new then. The specific sample here is in pristine shape btw.

 

[amirm]

As to BBC, who knows what their production samples looked like. In their paper they show an anechoic response of a prototype that is very different than what we are seeing: https://downloads.bbc.co.uk/rd/pubs/reports/1976-29.pdf

Given that the scale of the measurements is only 30 dB (vs my 50 dB), the response they thought they were getting was quite flat. So all this justification of bass dip, etc. does not belong to their target. They have manifested themselves after the fact. Indeed, there is zero talk of accommodating vans, wall installation, etc. Fascinating that they paid attention to off-axis response indicating they wanted the reflections to correlate well with on-axis -- something we cherish today.

If folks were cloning that response, they would have garnered far more positive response from me. Instead we have a speaker with clearly wrong bass level response which is trivial to get right, only if they paid attention and didn't just go by folklore and faulty history.

BTW, someone talked about vans being dead. Same doc talks about listening tests assuming an RT60 of 0.35:

 

[Endibol]

See below

 

[Endibol]

This is how they look without grill.

 

[Mr. Widget]

Thank you Amir for testing such an iconic design. Even though the results align with what many of us would have expected from such a light and lively speaker and enclosure, it is great to see thorough measurements of the FR, distortion, cabinet resonances etc.

 

[dfuller]

This is how they look without grill.
View attachment 328618

Wow, there's a lot I would not do there... So many points for diffraction. Certainly explains its absolutely bizarre off-axis behavior.

 

[Mr. Widget]

Wow, there's a lot I would not do there... So many points for diffraction. Certainly explains its absolutely bizarre off-axis behavior.

Very true, and most of us will say that, but we are basing that on the 48+ years of experience that we have gained since the very competent designers first penciled this design.

 

[Deleted member 21219]

I'm sure the original members of the BBC Loudspeaker Design committee would be having a chuckle at this discussion though what they did at the time was very serious and ground breaking. I was on the distribution list for the minutes of the committee and some of the minutiae and detail was bewildering at times. The only design tools they had was simulation, scale modelling and listening tests and a slide rule or two. The general quality and consistency of off the shelf loudspeakers at the time was pretty dire. The commercial sales of the 3/5a potential was never a thought - I don't think.

This is somewhat in line with the comments of Dick Gundry's son, Ken. The comment is here:

Well, of course having found this, I have to jump in. My father, Dick Gundry, who spent almost all his working life in the BBC and was for many years responsible for maintaining technical standards in BBC Radio (which have sadly gone down since his retirement in about 1971), and who was known behind his back as golden ears, would not have been pleased to have his name attached to a deliberate departure from a flat frequency response in loudspeakers. Has anyone any idea on how this term arose? It must have been much more recent than 1971.

One of my father's responsibilities back in the late 1950s and early 1960s was the development of stereo techniques in preparation for a means to broadcast it. (Some of those early experimental recordings have more recently been issued on CD by the BBC). At that time the BBC developed its own monitoring loudspeakers on the grounds that commercially available ones were generally not very good. I used to say that loudspeakers were either good or loud but not both! During early stereo experiments it became apparent that the best BBC monitoring speakers of the day did not perform well in pairs for stereo because they did not match each other closely enough, particularly in phase response, so central images tended to be diffuse. A major reason was that to accommodate variations in the drivers each and every cross-over network was adjusted for a flat amplitude response. A new range of speakers was developed, but it is possible that at least for those first ones, the uniformity was considered more important than perfect flatness, and thus the speakers may have shown the "Gundry dip". However it would not have been a design aim but a side-effect, and in any case my father would have had no input to the designs, which were developed at the BBC Research Department (Dudley Harwood, Spencer Hughes et al.)

Kenneth Gundry, San Francisco

... and is derived form a thread with Ethan Winer on Hydrogen Audio:

Gundry Dip or BBC Dip?

Gundry Dip or BBC Dip?

hydrogenaud.io

 

[617]

Honestly I don’t know why but the majority of British audiophiles are stuck in a 70’s time warp, they use the same equipment, even listen to the same music the rest are stuck in the 1950’s.
Keith

 

[EL_PW]

I considered buying a pair of these a few years ago because I was looking for speakers that were designed to enhance the frequency band of the human voice. I don't think these were ever designed for music reproduction. I think the BBC wanted to make the news announcers voices delivering the propaganda to your ears as hard to ignore as possible. That's just my own weird little theory that I have come to based on "internet research" so it is probably 100% wrong and not something rational people should ascribe any weight to whatsoever. Certainly they didn't look at best practices for frequency reproduction in the MK Ultra "Mind Kontrol" program. Plus at Rocky Mtn Audio Fest (R.I.P.) the LS3's I heard sounded awful playing any music. And I wondered if the dental issues that our British cousins struggle with have an effect on their auditory impressions as well.

Wound up with Genelec's instead. AwWWWww yiiIIIiea!

 

[MattHooper]

I considered buying a pair of these a few years ago because I was looking for speakers that were designed to enhance the frequency band of the human voice. I don't think these were ever designed for music reproduction. I think the BBC wanted to make the news announcers voices delivering the propaganda to your ears as hard to ignore as possible. That's just my own weird little theory that I have come to based on "internet research" so it is probably 100% wrong and not something rational people should ascribe any weight to whatsoever. Certainly they didn't look at best practices for frequency reproduction in the MK Ultra "Mind Kontrol" program. Plus at Rocky Mtn Audio Fest (R.I.P.) the LS3's I heard sounded awful playing any music. And I wondered if the dental issues that our British cousins struggle with have an effect on their auditory impressions as well.

Wound up with Genelec's instead. AwWWWww yiiIIIiea!

I wonder, though, given the generally strict and particular set of measurements people here generally seek, what chance did a speaker like this ever have of being rated good, or being a remotely attractive option to this group? These measurements it seems were worse than expected, but I doubt there was any chance they could have tempted anyone away from Genelec...

 

[sktn77a]

I've owned LS3/5As for over 40 years. I've tried numerous alternatives in this time (ADS L810s, early Kans, LS5/12As, Aerial 5Bs, P3ESRs, M30s) but always go back to the LS3/5A. Yes, they are inaccurate but this inaccuracy is on the "euphonic" side. With a suitable subwoofer to a) fill out the bottom end and b) relieve the B110 from overloading bass excursions, they are just plain hard to beat. The review is a little harsh but informative if taken in the correct context.

 

[Mr. Widget]

These measurements it seems were worse than expected, but I doubt there was any chance they could have tempted anyone away from Genelec...

I suppose if you were assuming objective high quality due to their popularity they may be worse than expected. In the realm of audio, I learned a long time ago that popularity and public opinion are often poor indicators of quality.

That said, I actually think the LS3/5A is a very good speaker for its era and original purpose. The flimsy cabinets were notoriously live and colored the sound, the speakers were terribly inefficient and SPL limited, but all in all, as a known quantity they were quite suitable for their intended purpose.

 

[GXAlan]

As to BBC, who knows what their production samples looked like. In their paper they show an anechoic response of a prototype that is very different than what we are seeing

+1 and my understanding is that the paper reflects the true BBC prototype.

Then, the BBC made 20 pairs of “production” LS3/5 speakers which were used in vans.

Then, they licensed the design to 3rd parties and then components became unavailable which led to the BBC and licensees coming up with the LS3/5A.

We also have this confirming the scoop in the woofer from this 1972 of the LS3/5 non-A

Which looks pretty darn close to the NON-anechoic measurement with minor differences in tweeter axis fully explaining the differences above the crossover.

This further shows that this specimen is “close enough” to the original.

The FR isn’t very neutral, but we have to recognize that this was 51 years ago. To put things in perspective, 1972 is when Formula 1 decided that seat belts should be mandatory…

Along the same lines, the fact that the non-anechoic measurements of this 1980’s specimen in 2023 reasonably match the results from 1972 is also a testament to the component selection. They used high quality components which still seem to be within spec.

This is not a good speaker in 2023 for transparency but this is a great specimen representative of the LS3/5a.

This sounds better than you’d expect from the measurements, in stereo, but the main reason to own these is as a historical collectible or desire for period-accurate sound.

 

[Ron Texas]

I doubt these sold for anything like $5k, adjusted for inflation, when they first came out. Thank you @amirm

 

[MAB]

I doubt these sold for anything like $5k, adjusted for inflation, when they first came out. Thank you @amirm

They always seemed expensive to me considering the sound quality in a home.
Here is ASP and $430 1977 dollars adjusted for inflation for comparison. Sources are in the links.

Seems the ripoff started after 2007.

I had two pair at different times in the '80s, they cycled through our store so often and I had to try, but they just don't sound good for music listening, it seems a totally different design objective. Like listening to an old telephone, which isn't as bad as listening to a new cellphone I guess.