"Arny Krueger" wrote in message
...
"William Sommerwerck" wrote in
message
* In this case, I wouldn't have caught the problem. It
turned out that the Dayton-Wright preamp used switches
that were very good at generating RF when opened or
closed. The RF caused the triple-diffused output devices
in the Lux 5M50 to melt from tertiary breakdown, a known
problem with this power amp. (Ironically, Mike Wright was
aware of the RF problem with the preamp, and had added
holes to the PC board for suppression caps.
Unfortunately, the holes were not populated.) This was
not the first nor the last time this happened; the next
time was my own stupid fault.
The real problem is that both the DW preamp and the Lux power
amp were just more examples of poorly-designed high end junk.
For their time, none of the Laboratory Reference Series products -- except
the 5T50 tuner -- was particularly expensive. * They weren't cheap, but they
weren't Mark Levinson, either.
Junk is in the eye of the beholder. All my LRS stuff continues to work,
though it's more than 30 years old.
* According to the guy who fouded Kinergetics, the 5T50 was designed to
"sound like" the Marantz 10B. I don't know if it did or didn't, but several
reviewers felt that the 5T50 was the first digital tuner that "sounded
good", supposedly because Lux did a good job keeping the digital "junk" out
of the audio circuits. This ought to be a trivial engineering exercise, so
if the 5T50's sound was surperior, it was likely for another reason.
The world is full of fine-sounding preamps that don't create RF-rich
transients when you operate their controls. Many of them create no
audible transients at all! The world is similarly full of fine-sounding
power amps that don't melt down every time you drive them with a
few transients, or even a lot of transients.
I don't know why Willaim Watson Michael Dayton-Wright chose those particular
switches. If you want to damn him, damn him for not installing the caps.
As for the power amp... At the time it was designed, there was a lot of
arguing about slewing-induced distortion, TIM, and the like. (There were
tube power amps that suffered from these problems. Dig through your
early-80s JAES issues for an article about one.) The assumption was that,
the wider the open-loop bandwidth, the less likely TIM would be a problem.
So Lux used triple-diffused RF power transistors -- despite the common
knowledge that they could be blown by RF transients. That was the cause in
three cases where the output transistors blew -- not audio transisents.
It would be interesting to know what fraction of the total production of
Dayton-Wright audio products are still in service. It's my recollection
that DW briefly rose to fame on the strength of an interesting idea with
a lot of potential [sic!] - sealing electrostatic speaker elements in thin
plastic envelopes full of an insulating gas. Unfortunately, they were also
the reactive loads from #&!! and could and did fail frequently in actual
use.
The Dayton-Wright electrostatic speakers could, with a suitable amplifier
(such as Crown M-300) play at near-earsplitting levels in a large, dead
room, cleanly. I do not recollect the panels having a reputation for
failure, though I could be wrong. The use of SF6 as the dielectric gas
reduced arcing, which is the principal (thought not only) cause of panel
failure.
These speakers were among the hardest-to-drive loads, ever. There was a bass
impedance peak of around 100 ohms, while the impedance fell to less than 2
ohms in the upper midrange. If the amp couldn't pump the required current
(and some couldn't, even at moderate levels), you'd hear "current clipping".
One might equally ask "What fraction of the total production of Company X's
products are still in service"? The switches and electrolytic caps are the
quickest components to deteriorate. I have (and have had) stuff over 40
years old, and am amazed that it continues to work.
As far as the fail-o-matic Lux power amp went...
Arny, you have a good vocabulary, so why not use it well? "Fail-o-matic"
suggests self-destruction, which this amplifer was /not/ prone to. (I could
name another brand of amplifer which, in my experience, repeatedly
self-destructed for no obvious reason.) It was damaged when subjected to a
form of electrical abuse that Lux naively assumed would never occur. Lux
should have added low-pass RF filtering outside the feedback loop.
...Lux had been around for a long time even way back then. They should
have known better. Apparently their transition from tubes to solid state
was not smooth. Lots of people built very durable power amps with triple-
diffused output devices.
Who? My understanding was that designers generally refused to use these
devices, precisely because of the known problems.
No doubt Lux [never?] figured it out, or simply went out of
business due to their technical incompetence.
Lux is still in business (I think), though a shadow of its former self.
And before we go... I have to acknowledge Arny's real motivation in posting
this. He just loves kicking people who own not-cheap equipment with
less-than-perfect reliability. "If only you'd bought Grommes, you wouldn't
have had these problems."
http://www.grommesprecision.com/precisionelectronics