BL> He, he. I'm not fazed by that. I've designed sets with inherent
 BL> faults that no tech *ever* found. A tech's approach is to keep
 BL> things working; an engineer's approach is to make things fail.

 RG> You didn't happen to work for Akai by any chance? 

  (grin)

 BL> I see. I've just told you that it is *never* (read: never)
 BL> layout, based on extensive actual design experience of these
 BL> things and you choose to contradict me. Okay

 RG> I realise that this is slightly different to the audio amps
 RG> under discussion, and correct me if I'm wrong, but don't many
 RG> HF circuits actually use the PCB tracks themselves as part of a
 RG> resonant circuit?

  Yes... at 10-100MHz. Roy was talking about an oscillation at 1MHz in
the broadcast band (and so was I).

 RG> Surely layout would then be very important for stability.

  Not at 1 MHz. Pay attention. To make a power supply oscillate at
1MHz you would need Paul Edwards to wire it up.

 BL> I supppose you've confirmed this by fixing an oscillating amp
 BL> or DC-DC converter by changing the layout, and then changed the
 BL> layout of a good one to make it oscillate? Pig's arse, you
 BL> have!

 RG> I haven't, but I'm a service tech, not a design engineer. I
 RG> just fix whats there, rather than redesigning things and
 RG> changing the layout.

  Yes... which was the message I was trying to convey to Roy.

 BL> If I put a design into production (read: that thing you have
 BL> *no* experience of) and it fails, there is a sudden stockpile
 BL> of 100,000 repairs! Therefore, I

 RG> Sounds like Akai to me. 

  I've never understood Akai. It costs the same to make a good set as
a shit set (within $1), so why doesn't someone fix the awful design
and their bloody-awful factory? It must mean that the Akai sales group
simply doesn't care.
  
  I actually put a TV into production with a fatal flaw no one noticed
for 3 months. Jeeesuz! Fortunately, it was a 12", not the mainline
set, and fortunately they'd all been sold in Victoria... but 5,000 TV
sets is a hell of a lot! It tends to teach you a lesson.

 BL> Service techs just slap on a 0.01 and hope for the best - if
 BL> you can't fix it, modify it - and never find the real reason,
 BL> never

 RG> Crap... This is what failed engineers do when asked to
 RG> service/repair something that has broken down. Service techs
 RG> fix the fault.

  I was talking to Roy, who had just modified a small-run power
supply, and I started to explain that "mods" can be dangerous. He got
up my nose, so I let him have both barrels. 

  But he is right too. Roy deals with small-run stuff that is very
likely to have design faults built in, simply because no one ever put
the effort into it. There is no excuse for Akai. A large factory can
and should spend the time to get it right, which means that when a
tech sees a faulty set he's pretty sure it's faulty... not built that
way. 

  I see quite a few design faults and I get the factory to fix them,
but it's not easy to convince Japanese or Koreans!

 BL> Most of the time, all you guys have to do is fix something that
 BL> an engineer designed right,

 RG> That's the theory, but surely you are not implying that
 RG> everything designed by an engineer is flawless or couldn't be
 RG> improved upon?

  Most of it is, actually. The Japs are particularly good at making
each new design a development of the one before, so the breed slowly
becomes perfect... but no one is perfect . As for "improvement"
much of this is a design decision for one reason or another (mostly
price). Almost anything can be "hotted-up" by shaving the necessary
tolerances built into mass production units.

 BL> and when you find a bad one you use standard cures... slap a
 BL> 0.01 across the collector base (0.01 is bit high, btw).

 RG> Crap. I've repaired zillions of electronic faults, and not once
 RG> have I needed to shonk a repair by redesigning the circuit...
 RG> the exceptions being to perform 'standard modifications' from
 RG> manufacturers fault/data sheets.

  At some stage, some tech has found the fault that led to the "mod"
sheet, so there is a fatal flaw in your logic. All design faults come
back from the field, in spite of what the factory implies. What you
do, is develop a list of reliable techs who get it right, and when
they say they've found something "funny" you check it out - quick!

 BL> Unfortunately, design engineers can't work like that. We have
 BL> all the time in the world to get it right. I offered you a
 BL> little of that

 RG> Pity that more of them don't actually use this time then. Even
 RG> a lowly service tech like myself could do better than some of
 RG> the designs I've seen :-(

  No, you couldn't (grin). A good tech makes a lousy engineer.   

  As I said at the beginning, it's a totally different mindset. An
engineer takes a circuit and tries to make it fail even when it is
working perfectly, so that at the end he knows it's foolproof. A tech
will tip-toe away from something that is finally working, and often
never knows if he has found *the* fault. This is not to say that there
isn't crossover in the mindset. What you call a "failed engineer" is
just a fiddler, but there are nevertheless true techs who have the
true engineering attitude, and vice versa.

  In the early days of transistor radios when I was fixing them for
money (at 6 an hour), I was the best tech in Sydney at that. You
suspend all thought and just *go* to the fault. You "know" the set and
hardly ever use a multimeter, let alone the CRO an engineer can't live
without.

  In fact, that's the most obvious difference between a tech and an
engineer. The engineer has a CRO, the tech only has a soldering iron. 

Regards,
Bob
___ Blue Wave/QWK v2.12
@EOT:

---