Sound Quality Vs. Measurements

Status
Not open for further replies.
This is the tip of the ice burg . Some look as commercial as Revox . After the war we had all the industries and engineers to do such things . What we didn't have was the willingness to have a fair society, all were to blame . I was talking to some of the chief engineers of the old Rover company . To hear them talk it was a period of great success with them at the head of it . I am sure the rules of this forum would not allow me to say more as the words I must use are not the nicest .

A GUIDE TO BRITISH TAPE-RECORDERS
 
Vintage recorders

This is where Julian Vereker ( Naim ) started in electronics ( cars before that ) . His minimalist amplifiers sounding more like the tape deck compared with phones I suspect ? Julian's point being that somehow when playing back a tape recording the reality of it was lost via hi fi . Julian's belief that the tape recorder had preserved the experience . I would go further . What Julian hadn't understood is that real sounding recordings seldom sound good on ordinary equipment . The old TV's being the worst . The Beatles and Pink Floyd seem to sound good in all ways and all systems . That's when the recording engineer needs praise . I listened recently to Dark Side of The Moon as original unmixed single tracks ( second generation 8 tracks , not 16 , that's never allowed out ) . Even better than you can imagine . So clean and deep .

Would anyone like to comment on that . I feel Julian was right to eliminate the tone controls . When set to neutral they still degrade the sound ( softer , less snap , out of focus ) . Doubtless much of that is poor design . What I would like to try is tone controls inside the power amp feedback loop ( or passive , or active bass and passive top ) . I can see no reason why they should degrade the sound . Not least if switchable . They need to be tailored to the job . For example an OB speaker rolling off at 160 Hz is better served by a 160 Hz control than 100 Hz . What people might not understand is that early recordings often are transformed by standard tone controls . Leak Varislope doing a great job with the actual EQ curves for most types back to 1927 .
 
Yes Andrew, I did.

Here's why. At the time I seriously took my tape recording, which was in 1970, I knew of quite a few companies making tape decks around the world. From the UK, by far the most widely known was Ferrograph, and, speaking from limited experience with them, they were excellent machines with unfortunately small popular appeal, I believe because of its rudimentary external design (i.e. looks).

Neal was more into small industrial purpose cassette decks, and as far as I know, they are used to this day by say the police when taking statements, etc.

The fact that it seems there were many more is of little relevance, as few had even heard of them, let alone owned one, outside the UK.

I believe many had gone prematurely because of the British inability to do what the Japanese were doing wholesale, and the Germans to a lesser extent, was to pool resources. By way of example, most German made tape decks, from people like Braun (the famous TG1000), Uher, Grundig, ASC and others used head assembly blocks from the same manufacturer (forgot their name), which were furthermore interchangeable, allowing you to switch from 2 to 4 track in less than 60 seconds, no tools required, to have a fourth head for synchro or reverse play, etc.

I have long forgotten the name of the Japanese company specializing in tape magnetic heads, which serviced the likes of Akai, TEAC, Otari, JVC, Sansui and generally most who ever made any tape decks (though some dropped out of the game early on, like Sansui and JVC).

Not at all unlike the British Leyland story, within which everybody and their dog made their own engines and mechanical subassemblies. Since the development of an engine is a lenghty and damn expensive business, companies like Triumph soon had outdated engines which could not keep pace with modern industrial standards of the day, while Rover fared better, dropped their ancient pushrod engines and introduced modern units in their model 2000, for example. Why that 2 litre engine could not be made centrally and spread around the BL group, with the usual designation change, remains a mystery. Made centrally, many more could be made, thus reducing the price per unit, not to even mention much cheaper and more efficient quality control.

I suspect the UK tape industry fared about the same. Thus, it fared about the same as the UK auto industry, it sort of remained frozen on the semi-manual manufacturing level.

As I said, a great pity.
 
On tone controls.

I believe the mass production audio industry did them in when they started the "specs wars". All of a sudden, the most important thing were your boost and cut capabilities.

They forgot what the tone controls were for. I believe they were most useful in helping marry the speakers to a specific room.

My old reViox A78 integrated amp had tone controls done as they should always have been done, but prices made sure that was not to be. First, it had separate tone controls for L and R, enabling them to do what they were supposed to do, make say L speaker, positioned near a large window not to shout and spit, or boost a bit of bass to the R speaker placed well away from the wall, unlike the L which simply had to nearer to a wall.

Secondly, they had dB steps, specifically -6, -4, -2, 0, +2, +4 and +6 dB. Thus, they were easily and reliably repeatable.

And third, these were discrete steps backed by discrete resistor networks, not dirt cheap pots made in the jungles of whichever Far Eastern country.

They did NOT colour the sound, but then, I never really had any need to ever go beyond +/- 4 dB steps, and I had a difficult to please room.

To the best of my knowledge, the mainstream industry produced the last of such integrated amps by around 1980, specifically Dual models CV 1600 and CV 1700. By then, audio was no longer "in" and the fight for customers had become very bloodthirsty, launching price wars. Price, and only price, is the key culprit here, because it made the bean counters superior to engineers.

As for Julian Vereker and NAIM, no doubt both deserve a full page each in any decent audio history book. However, to my mind, NAIM does have a specific sound which is recognizeable, and that fact to me is proof that the sound is not so much true as it is proprietary. I happen not to be a fan, though there are many who swear by it, and are very partisan about it.

Put in context of price, as everything is nowadays, I will readily agree that Naim remains a very good choice overall. I actually preferred Armstrong's products and to this day rue the fact that I never actually owned one.

I wonder how many American members here even remember a Californian company called Craig? The usual deal of the day, designed in USA, manufactured in Japan (not sure, but I suspect by Sanyo Marubeni). Made only receivers, only 3 models, and I heard the top model only because a friend of a friend brought one back home from USA. I don't think I have ever heard any receiver, by anybody including Marantz (given tha receivers were their star products), which could match that sound. Long gone by now, of course.

Tone controls were perverted by the audio industry, the fundamental idea by people like Baxendell was good, price cuts made them bad.

Just as the current vogue of having only the basic volume control is just as sick to me. There is no ideal tracking pot of ANY kind (thought some are better than others), NOBODY hears exactly the same on both ears, and NO TWO SPEAKERS ever have exactly the same working conditions in any room. Just as introducing remotes is a perversion to me - you minimize the design, and then make the signal go through a procesor with a built in resistor ladder, plus a couple of hundred other by-the-way transistors, and call this a "minimalist" design?
 
The strangest part of the Naim sound was it's ability to be better than it should be through Quad ESL 57 speakers . This is no mystery as J V liked them . What is weird is how the house sound of Naim vanished whilst retaining the informative nature of Naim amps when 57's .

The Naim was not a greatly different to even a Denon PMA 250 budget amp . It had larger power supplies than most . The design of the transformers being slightly odd . Tantalum coupling capacitors ( yikes ) . One thing they did do is listen to capacitors in power supplies . Not for quality so much as size . Naim believed some people used caps too large for a given transformer . A point of view supported by Douglas Self from a different perspective . My brother thought such amps sound bloated . Valve designers are forced to be more modest in CV values . Some believe it can be a positive . It should be said if it can not be shown by measurement why go larger ? The caps must be able to meet the current demands naturally . It will also make designing and earthing point more difficult if using more caps .

The sound of a Naim amp is incisive and seems to have less stereo than others . None of that can be easily explained . The designs themselves are remarkably standard in the majority of what that have . Neat and nicely done . Nothing like the RCA design people suggest . More like Sinclair .
 
Nige, don't forget that in the early 80ies, even Naim's smallest offering had fully electronically regulated power supplies, which few others ever did. Thta surely makes for a hell of a lot of difference.

Regarding caps - I agree one should be reasonable in that matter. However, I would say the vast majority throw in caps which are smaller than they should be, I have yet to see a standard product which I'd comment as having too large caps. The disproportion most probably comes from the fact that most had smaller transformers than they should have had, probably recitifiers as well, and when that happens, even modearte size caps will make the transformer heat up, even possibly start to hum. The caps will also tend to bloat.

I got away with my refreshed Marantz 170 DC, replacing a single dual concentric 2x12,000 uF/56V cap with two 22,000 uF/63V caps only beause I took the trouble to read what the main tranformer fuse said, and the main fuse says "3.15 Ah" for 220, 230 and 240 VAC, suggestiong a 690 VA transformer. Rectifier diodes discrete mammoth types. So it could take it. Whereas most can deal with one, possibly two sizes up larger caps, but this is +(22:12) 83.3%. If you've got it, flaunt it, baby. :D
 
Last edited:
Oh yes, ...

As for sounding "bloated", I agree that can happen when very large capacitors are used, like 33,000 uF and larger. Quite simply, they are slow.

That problem has been tackled by say Sony by using two large caps, say 22,000 uF, in prarallel with same make, same model but only 4,700 uF.

The other solution is to replace one 22,000 uF cap with two 10,000 uF caps in parallel. Never ever fails. My preferred method. Besides, two 10,000 uF in parallel will usually give more current and lower output impednace then the same type, same make 22,000 uF cap.

Either way, I also tend to use 100 uF // 3.3 uF // 100 nF // 1 Ohm+220nF right on the power amp board. Improves high frequency sound and gets rid of spurious inductance. I saw that first on the Otala TIM amp way back in 1974 (IEEE publications), tried it, LOVED it, stayed with it to this day.
 
I have been playing with open baffle speakers and very pleased with the results . Was asked to do some music for a 90 year old lady and offered to take them along to the village hall , how very authentic when playing 78's . This was strongly discouraged .

I have a pair of Scan Dyna ( Dynaco ) A25's . They sound so enclosed and cupped , a speaker of their time and not of now . Yet sometimes they are magical . So for the princely sum of $10 I added a pair of generic Motorola piezo tweeter clones ( Maplin UK code WF55 ) . I made an L-pad attenuator ( - 4dB ) and set the resistance total at 15R9 . I used 1uF 100 V polyester as a crossover . After bedding in I have to say as a pair of lab speakers I am rather pleased . The total cost of them about $50 . They need coffee cup stains removed to be perfect ( Danish oil ) . I was always told these piezo are very good if used as Super tweeters . The capacitance is about 110 nF if wondering . That means my 15R9 is about what the amplifier sees . The tweeters sit on the cabinet tops . If not required I turn them face down . Quickly the point is proved and they get reinstated . They don't honk as they aren't trying to do mid band . When I measured them they happily did 1 kHz which will be highly distorted .

My son was watching a rather rude Canadian comedy soap via the 25's . The F word used mostly as a subjunctive . A broken window sounding startlingly real . I will be happy to do that party now .Benny Goodman is the one I most will enjoy playing . 78's need treble more than people imagine. Putting a brick-wall filter on them often makes them sound even more enclosed . Dynaco's start to die at 10 kHz .

If I had some AR3's I suspect this would be an even better up grade . I dare say even to Quads ESL if crossed at 15 kHz .

BTW . One bass unit was locked solid when I got them home . I helped it by hand . It recovered . The people selling them were so genuine I thought it might for once be rust . Seems it was . As you might guess I had A25's in 1976 to 78 . I was offered silly money for them so sold them . I always missed them . LFC 824N my Triumph 750 cc motorcycle also . Apparently the last ever Triumph made by the real company . I was told that years after I sold it .
 
Did my first what is now called open baffle in 76, evolved them over the years by running the bass units in BR enclosures, I cant be bothered with OB bass issues..

SOTA OB setup circa 95/96...
 

Attachments

  • martinet1.jpg
    martinet1.jpg
    641.3 KB · Views: 196
Last edited:
Did my first what is now called open baffle in 76, evolved them over the years by running the bass units in BR enclosures, I cant be bothered with OB bass issues..

SOTA OB setup circa 95/96...

I am curious as to why the speakers are so close together and why they are firing straight ahead. Would this not narrow the effective stereo listening window, as compared to crossing the axes in front of the listener and spacing to achieve the standard ±30º from listener's perspective? I am guessing you have tried this and the configuration you have sounds better.
 
I am curious as to why the speakers are so close together and why they are firing straight ahead. Would this not narrow the effective stereo listening window, as compared to crossing the axes in front of the listener and spacing to achieve the standard ±30º from listener's perspective? I am guessing you have tried this and the configuration you have sounds better.

The picture is a bit misleading, the panels are not in their listening configuration, pushed back when not in use, panels are usually more forward and wider part when listening and there is some toe ( slight) but no, toeing in is not always necessary, unless you have a narrow window of dispersion, or require being on tweeter axis for best FR, plus this is not a point source, it's an Iso-diametric point source configuration OB from 150hz-5K there are also rear firing tweeters ..
 
Last edited:
My next open baffle experiment is to slightly fold the baffle sides to enhance baffle effects ( using 2 x 4 ft , 610 x 1220 mm , plenty of EQ ) . I have a feeling I will loose something . Maybe not if the angles are right . One guy advocated a 400 mm baffle , 570 mm one side and 620 mm the other ( on piano hinges ) . The height 1.7 metre . The drive unit central ( Eminence 12 Lta ) . He got the idea from A French design . Shame he didn't give the text as I would like to see the maths that led to 570 and 620 mm . Being French it will probably be the French way ( There is the right way , the wrong way and the French way ) . Before anyone says about racism I am the French way if anything . The French locomotives are probably are the best in the world . Better than Germany or Japan . They can do it if they try . By best I include being very safe .

BTW I don't find Qts for OB a big deal .The question mostly is will the drive unit take EQ without bottoming . As my friend said about Bose 901 . They bottom too easily to be a serious PA speaker . As a mid-range they are rather nice . Executed with modern drivers could change that . I have heard them sound good . Never in a house as far as I remember . Almost the same rules apply as OB . I think I will try to design very compact double sub woofers giving 20 to 100 Hz . That should cure the bottoming and not cause the usual sub woofer disconnect . The Celestion OB sub woofer looks too good to be true . Would be nice if is does work .
 
My recent research shows we often sweep EQ under the carpet . Many speaker designers delight in the vagaries of passive crossovers . That seems so odd to me . Often EQ is used without people knowing .

I see no reason for power to be a problem . Very high quality low cost power can be had . Doubtless not very good over 250 Hz . No problem as that's not required . I would shop in the Pro Audio market for that . Pro Audio thinks in terms of bass transients and timing . They might be better than Hi Fi in that respect . Turbo Sound has optional digital delay for their rigs . Considered essential by many . The Turbo sound will work OK without it in small venues . I am not a great expert on this , this is how I am told . One day I might have to digitize my LP's so as to have this .

Some might remember I showed a very stable tag strip amp to do bass duties . I really should use it as I see no reason to doubt it . The EQ was in the feedback loop . It was better than - 62 dB distortion at any level up to 20 kHz without EQ ( - 80 db at the intended frequencies ) . Not bad for 150 watts and 5 active devices . The slew rate was poor ( 17 V / us ) . As 250 Hz was the limit I think I might ignore that .

The > 100 Hz might be nicely served by a SE tube amp . The transformers also the crossover ( made ideal , possibly still going to 20 Hz intrinsic ) . A similar amp with ideal transformer for the super tweeter ( more like RF transformer ) . I am not a great fan of tube amps . SE an exception .

One thing I did note when visiting the USA . Rooms are usually much larger . This causes so many problems . One friends house had a large central kitchen ( Texas , wood construction for tornado's ) . Sound is not too good and seems typical . Another has a house in the Georgian style and that is like the UK and probably nicer than our houses ( Maryland , wood and brick ) .

Someone said the problem is you can not scale up speakers indefinitely . When reaching the limit answers to problems will make people see things very differently .

I was asked by someone if I found standard car stereo rather good ? Yes I had . The reasons are many . One is the shape and size is always the same . Much money is spent to get it right . This might not add $10 per unit . Sure like everyone I have heard the bad ones . The room is small . Has sealed doors . DC power source . Modest , defined design goals which are easily satisfied . And zero expectations on our side . Most so called better car systems sound less like live music to me . It is good cheap car stereo I like . People who live on boats often use them with hi fi speakers .
 
Member
Joined 2004
Paid Member
Have any of you actually heard the big Magico's? I have twice. While I like Alon and am really impressed by the enterprise he has assembled I have not been impressed that much by the horn speakers. They are really an interpretation of the Goto drivers. The horn is different and required some customization of the components to get it all working right. But horn's have never impressed me. Its the cloth ear thing I have. . .
 
Status
Not open for further replies.