As have I. And the Kappa 9 is certainly a faulty design.
I realise that speakers can be a personal choice and that not all B&W speakers are excellent, but to suggest that the 800D and 802D are anything less than a stellar design is worrisome. Both are extremely accurate speakers. In fact, every incarnation of the 800 and 802 has, arguably, been as good as any speaker in their respective price classes. I accept that you may not like them, but that does not mean that their design is anything less than excellent. As is their sound quality, IMO.
My point exactly. Done right, a valve amp can sound pretty decent, but for ultimate sound quality, SS is the way to go.
I don't believe you can make that claim.
Nope. The ONLY way to design a system is to start with the room. Then choose the speakers. After choosing speakers, then an appropriate amplifier to drive those speaker can be selected. If the load happens to be "stupid", then a suitable amplifier to match those speakers can be chosen.
As a matter of fact, I am certainly not. I merely cited some speakers that sound very, very good indeed, but present difficult load impedances. Nothing more. If I cared, I could probably locate a whole lot more crap sounding speakers, that present a difficult load, just as I could find an equivalent number that presented an easy load. In my position, I have spent a good deal of my career dealing with speakers that present difficult loads.
Sure, that would be nice. Although speaker designs are tending to easier load impedances, thanks to CAD, there are still some designs which are being manufactured that will be difficult to drive.
Yeah, right. Never gonna happen. Not on this planet anyway. Would it be nice? Probably. You and I don't live in North Korea.
Most speaker manufacturers have budget designs that probably never should have seen the light of day. However, to tar the 800D and 802D with the same brush as the model you cited is, at best, disingenuous. At worst, it smacks of ignorance.
I have NEVER seen a mid-fi or better speaker than is inferior to Bose. In fact, I was servicing a Bose product a few years back and was experiencing some difficulties in working out how to pull the thing apart. I desperation, I called the Bose service department. The tech I spoke to laughed, as he said: "It's held together with double-sided tape!" As for consumers being ripped off by speaker purchases, I cannot say. I've never manufactured speakers commercially, though I have done some designs for manufacturers. I, too, have seen some ordinary designs over the years, but I can't say if they represent a rip-off, because I don't know what the costs of manufacture, distribution and warranty are.
I'd like to remind you that you two are 'repair guys' not design engineers, and you don't know everything about audio design. (none of us do, of course) but the learning curve of a circuit designer is different from a repair guy, although a good repair guy can be worth his weight in gold, his opinions might be subject to scrutiny by designers on what is important in amplifier design.
I feel that a 'good' amplifier should be able to drive just about any load, whether or not the designer overlooked the speaker impedance. This of course happened with Dave Wilson's WATT 1 speaker, a really good sounding speaker with an adequate power amp. However, he did make a design error, that apparently improved the sound of the speaker, but caused a near short at 2KHz, but only in that region. The Krells or whatever he used just ignored it. My Electrocompaniet power amp, (the original Otala design) only drove it when I inserted a 0.5 ohm resistor in series with the speaker. Still, it sounded wonderful as a combination.
Of course, Dave has attempted to eliminate this problem in his later models, but almost any Wilson speaker is difficult to drive. It seems to go with making the speaker sound its best.
I feel that a 'good' amplifier should be able to drive just about any load, whether or not the designer overlooked the speaker impedance. This of course happened with Dave Wilson's WATT 1 speaker, a really good sounding speaker with an adequate power amp. However, he did make a design error, that apparently improved the sound of the speaker, but caused a near short at 2KHz, but only in that region. The Krells or whatever he used just ignored it. My Electrocompaniet power amp, (the original Otala design) only drove it when I inserted a 0.5 ohm resistor in series with the speaker. Still, it sounded wonderful as a combination.
Of course, Dave has attempted to eliminate this problem in his later models, but almost any Wilson speaker is difficult to drive. It seems to go with making the speaker sound its best.
And this 'repair guy' would like to remind you that for all your condescension, you design engineers can be remarkably clueless about how the world actually works down here on the ground. A speaker that shows an impedance of 0.2 ohms at any point in its curve is a defective design, period. The availability of Krell or similar arc-welders does not negate this fact.
I sometimes wish more designers would consult with repair guys. I've had more than my fair share of high end designs that can take many hours to dismantle. My background is that of service manager for Marantz (Australia), back in the 1970s. Back then, the designs were mostly delightful to service, with easy and fast access to all the stuff that typically fails. Except, of course, for the disaster known as the Model 500 Power Amplifier. I used to budget 10 ~ 12 hours to repair one. I still have mine. One of 300 built in Chatsworth Cal. I recall a recent repair job on a high end amp. The output stage had failed and the job turned out to be quite challenging. As part of the repair I noticed two small electros in the DC servo system that required replacement. Easy enough and I replaced those parts. I performed a visual examination on the other channel and noted that the same caps in the good channel also required replacement. Since every single output device, driver device, pre-driver and bias transistors had to be de-soldered, before access could be gained, the cost of replacing those two, $1.00 capacitors was horrendous. I felt terrible when I informed my client. $500.00 labour to replace $2.00 worth of parts. Bad design. Really bad design.
I can also relate a time when I advised a major sound reinforcement company on the shortcomings of the design of one of their most popular (but unreliable) amplifiers. I would dearly love to meet the genius who designed the Tapco CP500 amplifier. Hideous design, with a raft of really dumb design decisions. They told me I was wrong in my re-design suggestions. Six months later, the company issued a modification sheet, which incorporated all but one of my re-design suggestions. I also realise that my final recommendation was impractical, but it would have made the amp very reliable.
Emphatically agreed. I love lots of output devices and massive power supplies in an amp, along with very high VI limit points.
I heard a pair in Hong Kong back in 1987. Stunning sounding speaker. Running with ML2s, as I recall.
Yep. As would any good amp.
In truth (and this is just my opinion), after examining the original WAMMs (internally and externally) and the original Watts - Puppies, I gained the impression that David Wilson was not a classically trained speaker designer. Rather, a well-funded amateur, who lacked a serious understanding of electronic principles, but did have a good grasp of mechanics (nicely damped enclosures) and a desire to use the highest quality components throughout. Obviously, as time has moved on, Mr Wilson has likely been able to employ properly trained engineers to refine his ideas and designs, such that they do not display the clear faults of the original designs. Like the impedance dip of the original WATTs and the horrible electronic crossover used for the Puppies.
Like I said: Just my opinion, based on what I have seen and heard. I would be happy to own almost any Wilson speaker system.
Hi John,
With all due respect, and I really mean that, you can't possibly know what experience and capability we have in the world of design. Don't forget that we are also hobbyists who have studied a great deal more than the average technician has. In addition we have observed some of the best, and worst designs out there. In that regard my own experience probably exceeds your own. After all, you don't spend time examining other designs. I'd like to think we have learned a great deal about successful design through that process alone. I know I have.
In order to keep costs within a reasonable range, amplifiers should be designed to drive reasonable loads. If your desire is to drive a stupid load, then buy one of your designs I guess, but excellent performance is available with products that drive reasonable loads.
Let me tell you a truth. A speaker designer is aware what sucks efficiency and creates a difficult load. If, as you're about to claim, cost is no object to obtain peak performance, then it is a stark reality that you will have an active system with electronic crossover. The amplifiers will each have a driver or two to run and the loads presented to the amplifiers will be reasonable, easy in fact.
Back to comprise #1, single amplifier through a crossover. No longer performance at any cost, right? Okay. Assuming an intelligent designer (not one who is all ego), that person is fully aware that if they start with the right drivers and install them into correctly sized enclosures (whatever type), they do not need to do heroic things with the crossover - do they? As you are aware, a crossover is a system of loss elements designed to add impedance in series with the driver to the amplifier in order to control the power delivered to that driver. The power loss inside these networks is high, and if the driver picked has bad habits with impedance in series with it, more has to be done with the crossover to control it. Well, no. The driver should have been replaced with one more suitable, or the driver design should have been modified to correct those issues.
There is absolutely no reason why a speaker engineer needs to create a disaster to foist onto the buying public (who are not experts). John, your designs perform better into reasonable loads than they do into stupid loads. So why protect designers who create products that do not bring out the best in your own designs? Don't even try to mention cost is no object designs, because we both know that is only valid for active, multi-amp systems. We are talking about broken designs here, not successful ones.
Just out of curiosity, did Dave Wilson go back to the owners of those speakers and replace the defective products with ones that were fixed? That takes character of course, but did he?
As well, taking the dismissive tone you tend to do with others is getting extremely old. There are lots of "repair guys" out there (you were one when you started), but they have varying degrees of ability. There are lots of things we have to say that are valid, no less valid if said by an engineer who is an expert in the field. BTW, this is audio and hardly qualifies as "important". Tell that to engineers who work with medical electronics, space related or military electronics. Their equipment is far more complicated than anything you have ever designed and they really do work with secret technologies. Even the guys who keep time and work with ion fountain clocks have far more design expertise than you do. Audio isn't that high up on the food chain, stop pretending that it is and maybe get the ego under control.
What audio is, is fun, satisfying and expensive to work in. I just got my RTX-6001. You should look into one of those. It isn't the best, but it's better than most things out there. It might be better than anything you have on your bench. Now I can see below the noise floor to see how my designs are working finally. Did I say designs? It appears as though anyone can design something and maybe even call themselves an audio designer. The horror!
-Chris
With all due respect, and I really mean that, you can't possibly know what experience and capability we have in the world of design. Don't forget that we are also hobbyists who have studied a great deal more than the average technician has. In addition we have observed some of the best, and worst designs out there. In that regard my own experience probably exceeds your own. After all, you don't spend time examining other designs. I'd like to think we have learned a great deal about successful design through that process alone. I know I have.
In order to keep costs within a reasonable range, amplifiers should be designed to drive reasonable loads. If your desire is to drive a stupid load, then buy one of your designs I guess, but excellent performance is available with products that drive reasonable loads.
Let me tell you a truth. A speaker designer is aware what sucks efficiency and creates a difficult load. If, as you're about to claim, cost is no object to obtain peak performance, then it is a stark reality that you will have an active system with electronic crossover. The amplifiers will each have a driver or two to run and the loads presented to the amplifiers will be reasonable, easy in fact.
Back to comprise #1, single amplifier through a crossover. No longer performance at any cost, right? Okay. Assuming an intelligent designer (not one who is all ego), that person is fully aware that if they start with the right drivers and install them into correctly sized enclosures (whatever type), they do not need to do heroic things with the crossover - do they? As you are aware, a crossover is a system of loss elements designed to add impedance in series with the driver to the amplifier in order to control the power delivered to that driver. The power loss inside these networks is high, and if the driver picked has bad habits with impedance in series with it, more has to be done with the crossover to control it. Well, no. The driver should have been replaced with one more suitable, or the driver design should have been modified to correct those issues.
There is absolutely no reason why a speaker engineer needs to create a disaster to foist onto the buying public (who are not experts). John, your designs perform better into reasonable loads than they do into stupid loads. So why protect designers who create products that do not bring out the best in your own designs? Don't even try to mention cost is no object designs, because we both know that is only valid for active, multi-amp systems. We are talking about broken designs here, not successful ones.
Just out of curiosity, did Dave Wilson go back to the owners of those speakers and replace the defective products with ones that were fixed? That takes character of course, but did he?
As well, taking the dismissive tone you tend to do with others is getting extremely old. There are lots of "repair guys" out there (you were one when you started), but they have varying degrees of ability. There are lots of things we have to say that are valid, no less valid if said by an engineer who is an expert in the field. BTW, this is audio and hardly qualifies as "important". Tell that to engineers who work with medical electronics, space related or military electronics. Their equipment is far more complicated than anything you have ever designed and they really do work with secret technologies. Even the guys who keep time and work with ion fountain clocks have far more design expertise than you do. Audio isn't that high up on the food chain, stop pretending that it is and maybe get the ego under control.
What audio is, is fun, satisfying and expensive to work in. I just got my RTX-6001. You should look into one of those. It isn't the best, but it's better than most things out there. It might be better than anything you have on your bench. Now I can see below the noise floor to see how my designs are working finally. Did I say designs? It appears as though anyone can design something and maybe even call themselves an audio designer. The horror!
-Chris
Hey Zaphod,
I actually like working on the 500. I just rebuilt another that killed every other amplifier this group had. One person is a sales rep for Bryston.
I also cut my teeth on Marantz and had a fair amount of training from them. I did overflow work for them and was designated a full service shop (we could fix everything Marantz made under warranty) and we got the 500's to repair.
My own system is a 300DC and 3650 (if those switches would ever stop being noisy!). I bailed on Marantz after Philips destroyed the line and the distributorship was bought by a Quebec company, TC Electronics (terrible to work with).
I'm going to bet we are more similar than different.
-Chris
I actually like working on the 500. I just rebuilt another that killed every other amplifier this group had. One person is a sales rep for Bryston.
I also cut my teeth on Marantz and had a fair amount of training from them. I did overflow work for them and was designated a full service shop (we could fix everything Marantz made under warranty) and we got the 500's to repair.
My own system is a 300DC and 3650 (if those switches would ever stop being noisy!). I bailed on Marantz after Philips destroyed the line and the distributorship was bought by a Quebec company, TC Electronics (terrible to work with).
I'm going to bet we are more similar than different.
-Chris
I retired my Model 500 back in 1984. I swapped it out for a local, zero global NFB design. It blows the 500 away. Easily.
Probably. The 300 and 3650 were OK, but, as you have found, the lack of gold plating on the switches is a fatal flaw. For me, the highlight of the 70s Marantz stuff (apart from the 500) was the 1200b/240/250/250M/3300 products. Fabulous sound quality. The 300 (and the smaller 170) were brilliantly designed to be serviced (which was pretty rare).
Also, FWIW: I had (I wish I still had it) a confidential list of reliability data for all the Marantz models up to the late 1970s. Standout for reliability was the mighty 1070. Within the 3 year warranty period, failure rate was a stunning 0.5%. And that included the first series of 1070 amps which had crappy protection relays fitted. Worst (apart from the disastrous Model 500) was the 4230 receiver. It enjoyed a 63% failure rate within the warranty period. All thanks to those early, horrible TO220 output devices.
[ASIDE] How about those Motorola, TO3, aluminium (aluminum to you) output devices in the 140 and 2270 products? The vapourising silicon would burn tiny holes in the cases when they failed. Shoulda used RCA outputs. MUCH more reliable, with their steel cases. Motorola finally learned their lesson.
As for Philips: I disagree. Towards the end of the 1970s, Superscope was burning cash at a phenomenal rate (thanks to the Pianocorder) and the quality began to suffer. Compare the internals of the 2285 with the 2285b, for instance. Then we saw plastic faceplates on some models and some really horrible design choices. The Imperial range seems to have disappeared off the face of the Earth. Thankfully. It was garbage. When Philips acquired Marantz, they gave the brilliant Ken Ishiwata free reign to indulge some pretty nice designs. Models like the PM64 were much better than the Superscope designed stuff. I left the company in 1980, but I stayed a service agent for many years and watched some nice things come from the Philips years.
Probably. The 300 and 3650 were OK, but, as you have found, the lack of gold plating on the switches is a fatal flaw. For me, the highlight of the 70s Marantz stuff (apart from the 500) was the 1200b/240/250/250M/3300 products. Fabulous sound quality. The 300 (and the smaller 170) were brilliantly designed to be serviced (which was pretty rare).
Also, FWIW: I had (I wish I still had it) a confidential list of reliability data for all the Marantz models up to the late 1970s. Standout for reliability was the mighty 1070. Within the 3 year warranty period, failure rate was a stunning 0.5%. And that included the first series of 1070 amps which had crappy protection relays fitted. Worst (apart from the disastrous Model 500) was the 4230 receiver. It enjoyed a 63% failure rate within the warranty period. All thanks to those early, horrible TO220 output devices.
[ASIDE] How about those Motorola, TO3, aluminium (aluminum to you) output devices in the 140 and 2270 products? The vapourising silicon would burn tiny holes in the cases when they failed. Shoulda used RCA outputs. MUCH more reliable, with their steel cases. Motorola finally learned their lesson.
As for Philips: I disagree. Towards the end of the 1970s, Superscope was burning cash at a phenomenal rate (thanks to the Pianocorder) and the quality began to suffer. Compare the internals of the 2285 with the 2285b, for instance. Then we saw plastic faceplates on some models and some really horrible design choices. The Imperial range seems to have disappeared off the face of the Earth. Thankfully. It was garbage. When Philips acquired Marantz, they gave the brilliant Ken Ishiwata free reign to indulge some pretty nice designs. Models like the PM64 were much better than the Superscope designed stuff. I left the company in 1980, but I stayed a service agent for many years and watched some nice things come from the Philips years.
Last edited:
Zaphod, I also had access to a Marantz 500 and a Marantz 250 when working with John Meyer, in our Switzerland lab in 1974-5. We designed a 3 amp driven all horn loaded 'time alighed' PA speaker with a derived 3 way xover and a time delayed tweeter.
The Marantz amps were OK up to 4KHz or so, but I then made a small Class A amp with a 100V/us slew rate that was originally called the JC-3, later I gave the design to Mark Levinson and he ultimately made the ML-2 from the original JC-3 design. It was my first really successful amp, using complementary differential jfets on the input.
Now what about repair? I don't know how to make my amps easier to fix. When I can, like with the JC-1 power amp, my former Vendetta Research partner does the board layouts for my best designs. He has to fit the board(s) in a specific place and quality thru-path is his main concern. Later, others, outside my influence add all the accessories that seems to make up 1/2 the amp, and they just fit them about anywhere. Heck, I can barely get a bias set measurement without special probes. I agree that things should be easier to work on, because I have to work on them as prototypes as well as you guys when they break.
As far as output devices from the 70's and 80's, Motorola made the best COMPLEMENTARY power transistors, RCA made pretty rugged NPN power transistors, but I never have seen large complementary parts from RCA. Do you know of any? They DID make some complementary T0-5 transistors that I originally used for the JC-2 line amp. They were exceptional for complementary T0-5 devices, better than Motorola at the time.
The Marantz amps were OK up to 4KHz or so, but I then made a small Class A amp with a 100V/us slew rate that was originally called the JC-3, later I gave the design to Mark Levinson and he ultimately made the ML-2 from the original JC-3 design. It was my first really successful amp, using complementary differential jfets on the input.
Now what about repair? I don't know how to make my amps easier to fix. When I can, like with the JC-1 power amp, my former Vendetta Research partner does the board layouts for my best designs. He has to fit the board(s) in a specific place and quality thru-path is his main concern. Later, others, outside my influence add all the accessories that seems to make up 1/2 the amp, and they just fit them about anywhere. Heck, I can barely get a bias set measurement without special probes. I agree that things should be easier to work on, because I have to work on them as prototypes as well as you guys when they break.
As far as output devices from the 70's and 80's, Motorola made the best COMPLEMENTARY power transistors, RCA made pretty rugged NPN power transistors, but I never have seen large complementary parts from RCA. Do you know of any? They DID make some complementary T0-5 transistors that I originally used for the JC-2 line amp. They were exceptional for complementary T0-5 devices, better than Motorola at the time.
Last edited:
I agree with you about the old Marantz amps. The 500 was a beast. Arguably the best big amp of it's time (1973). I used to demo it against Phase Linear 700s and the like. There was no comparison. That said, when I retired mine, it sounded noticeably 'dirty' compared with it's zero global NFB replacement.
And yeah, that ML2 was a Hell of an amp. I recall a night in Melbourne, where the owner was running a pair of Martin Logan Monoliths with a pair. Pretty impressive. Had to keep reminding myself it was 30 Watts. When I heard the WAMMs in Hong Kong, it was with bi-amped ML2s. Again, very impressive.
Yeah, I get the commercial realities can often get in the way of engineering. I'll post some photos sometime of the amps I work on. 4 bolts, 8 nuts and four plugs and I have two (mono) output stages in my hands. About 15 minutes work. Best of all, interconnections are really short, so the maximum distance between the output devices and the main filter caps is around 150mm. Power supply inductance is, therefore extremely low. An entire mono output stage occupies around 1 Litre (including heat sink - fan cooled).
RCA second sourced MJ15003/4 chips and put them in their own cases (steel). Their bonding was far superior to Motorola's and the reliability of the devices far better.
40409 & 40410? Wow! That takes me back. As you would know, the Japanese were quick to build very fast, very linear TO220 1 ~ 1.5 Amp devices back then. Easier to deal mechanically with, compared to those RCA things.
Dunno if you ever saw them, but on of the most interesting devices to come out of Japan in the 1970s were the Hi-Rel, triple diffused power transistors. I recall reading a Motorola paper, stating that it was impossible to make PNP and NPN triple diffused power transistors, so Motorola headed down the epitaxial base route. At the same time, I had a spec sheet and a couple of sample Hi-Rel devices. ED203 and EB203. 20 Amp, 30MHz, 200Volt, 150 Watt (or thereabouts). Very linear, very fast. Rare as hen's teeth today. I have couple stashed away for a rainy day.
At the very least, that designer is incompetent or lazy. An LC 'zuigkring'* is an easy way to get rid of peaks in the FR of a loudspeaker, but will introduce those severe dips in the impedance curve.
* Literally 'sucking circle' in English, which seems to be much more appropriate than 'tank circuit'. Where is the tank in that? Shorting the lines in a loudspeaker with an LC-circuit most certainly sucks.
Although the B&W 800 series has a euphonic quality, they don't sound right to me. Look at the polars. B&W 800 Diamond loudspeaker Measurements | Stereophile.com. Although relatively straight on axis, it has a huge dip in its sound power from 2-3kHz. You could call it the Brit-dip. Sounds easy, but to my ears gives them a somewhat muddy timbre. Only on axis they are sort of ok.
I could be wrong but it's almost as if they were so excited by the kevlar midrange they had designed that they made that the pride of place and just splatted a tweeter on top to look nice. Dunno, I was a celestion kind of guy before I was tempted to the dark side of strange American speakers.
Why to buy those extremely expensive loudspeaker when can buy the best one for honest amount of money, LX521. It is an open baffle construction and not those boxed or even worst bass reflex loudspeakers.
I own early Linkqitz model, Orion. It uses, of course, active crossovers.
LX521.4 Store
Damir
Seems to be interesting, could you please cite the test(s) you´re referring to?
Does that mean that you have designed them to be as easy to fix as possible?
Yes. There is no reason why a '8 Ohms' speaker that has 6 Ohms at DC has to go anywhere under this value.
I should add that is not good that no speaker designer seems to try to make the impedance curves of his products linearized. It is, for me, an important point, and I compensate my personal speakers since decades for they are 6(+-1) ohms from DC to 40KHz).
Both with active or passive filters.
All this can always be done with few LRCs and the little extra cost is worth the listening improvement.
I believe the reason of this improvement is the amp has no more to deal with a non linear voltage VS current situation, depending on fréquences. Currents for fundamentals and harmonics stay coherent .
So, the feedback has only to deal with the internal errors of the amp itself and the voltages/currents produced by the speaker's resonances.
Sincerely your,
Tryphon Tournesol.
Last edited:
Pretty bad even on axis, and when bad on axis, then there is NO cure. This is very interesting, that so many so called high-end speakers are so so bad. The only advantage usually is that you can play them at high volume level.
Some time ago I had a similar argument with an audio friend. I tried to find out why this would be so, but his replies did not get any further than 'it's much easier for the amp' and 'the amp doesn't like phase shift between voltage and current'.
For me this is a great illustration of anthropomorphism.
Jan
- Status
- Not open for further replies.