Hans Camenzind's Book, Designing Analog Chips

[Jim Thompson]

Chapter 5 made good sense to me right up to 5-12, "Self starting
current source without large value resistors". Alas, I am unable to
understand this circuit well enough to guess what the no-no is. He
says the purpose is to eliminate wasted power by replacing the
"primary current" from Vcc in 5-14 with current the circuit is
intended to generate, I1. If Vcc in fig 5-12 is replaced by I1 and Q4
eliminated entirely, the figure might seem to accomplish that. ???

He who prays for leakage will die from the lack of it. The real
gotcha is that beta falls like a rock at low currents, so a low, but
not zero, leakage will not start-up.

...Jim Thompson

 

[Jim Thompson]

It's just a pamphlet that he describes as an "overview." It's not meant
to be all encompassing. Camenzind already has two other mainstream
textbooks to his credit. The main purpose of this latest is to simply
convey a perspective and assumes a working knowledge of electronics. And
as for the alleged historical inaccuracies, if the work was proprietary
and not made available to the industry, it did not exist.

Unpublished indeed... most of my designs are. But the chips were
manufactured. When I was 24, I was green, believed in absolute faith
to my employer, keep everything a secret, don't publish, all that rot.
It hit me in late 70, I quit (actually laid myself off), and now keep
everything to MYSELF ;-)

An independent
discovery that did not rely on the ground breaking achievements you
claim is a discovery nonetheless, and, if it was published, the credit
for that discovery *rightfully* goes to that author. Therefore,
Camenzind's historical narratives are absolutely correct, accurate, and
without error; this is *the* history of electronics development, and you
have been left out. This book is a significant contribution and
resource, well worth the money.

...Jim Thompson

 

[Winfield Hill]

Jim Thompson wrote...

Well... Three things...

Now we know why the book is only $21.95 ;-)

Wonder how he "introduced the phase-locked loop concept to ICs"?
Motorola was making PLL chips designed by Ron Treadway and myself
before Hans left graduate school.

The Signetics chips certainly had more impact, widespread use,
second sources, etc., and can still be easily obtained today
from multiple sources.

"In February of 1964 David Hilbiber of Fairchild..." discovered the
concept of what would later develop into bandgaps. Six years later
(1970) "...Bob Widlar put in the missing pieces".

Except I was making bandgaps well before I left Motorola in 1970.
Guess that'll teach me to publish instead of just going for the big
bucks ;-)

That's two things. Anyway, the correct dates must be much earlier.
I remember standing with Bob Widlar and his constant side-kick, Bob
Dobkin (who strove to look the same as Widlar, with his beard, etc.),
outside a lecture hall at the Philadelphia ISSC (Widlar wasn't one
to sit listening to papers being delivered), talking about bandgap
references, and Widlar teased me with the assertion that one could
make a bandgap from two transistors rather than the original circuit
requiring three, but he wouldn't tell me how. Bandgaps were already
mature by that time - the year had to be 1968 or the Spring of 1969.
After the conference we met again at the airport, going down the same
isle, except I was going to Boston and he was alone headed to Mexico.

 

[John Larkin]

The Signetics chips certainly had more impact, widespread use,
second sources, etc., and can still be easily obtained today
from multiple sources.

What ever happened to Signetics?

John

 

[Robert]

On 23 Feb 2006 07:30:19 -0800, Winfield Hill


I figure I had a bandgap around 1966. I'm pouring thru my files for a
schematic.

...Jim Thompson
--

Sounds like another vote for that book you've mentioned writing once or
twice.

Robert

 

[Spehro Pefhany]

What ever happened to Signetics?

John

Apparently Philips bought them in '75 and killed off the name around
'92.

 

[Joerg]

Hello Jim,

He who prays for leakage will die from the lack of it. The real
gotcha is that beta falls like a rock at low currents, so a low, but
not zero, leakage will not start-up.

He write in the text next to it: "These currents may be small (pA),
but..." When I read that I got some goose pimples. I hope there is
nothing like this in the engine controllers of our cars.

Regards, Joerg

 

[Jim Thompson]

Hello Jim,


He write in the text next to it: "These currents may be small (pA),
but..." When I read that I got some goose pimples. I hope there is
nothing like this in the engine controllers of our cars.

Regards, Joerg

You'd be surprised how many instances I know of. I just fixed one
this past week.

It's particularly common amongst those users of HSpice-derived
simulators (HSpice, SmartSpice, Cadence, Mentor...). Those simulators
will find a solution without regard to how the currents got started OR
NOT (algorithms based on CMOS _logic_, NOT _analog_).

You'll know you've been "Thompsonized" if I ask, "How does it start
up" ?

There are also those "designers" who rely on the rising edge-rate of
the power supply voltage to start such loops or operate POR's (as
we've discussed here this past week).

It keeps business coming in.

There's nothing sweeter than the voice of a COO/CEO begging, "Can you
_please_ fix it this week?" ;-)

...Jim Thompson

 

[Joerg]

Hello Jim,

You'd be surprised how many instances I know of. I just fixed one
this past week.

It's particularly common amongst those users of HSpice-derived
simulators (HSpice, SmartSpice, Cadence, Mentor...). Those simulators
will find a solution without regard to how the currents got started OR
NOT (algorithms based on CMOS _logic_, NOT _analog_).

Considering the rather excellent book I was surprised Hans even
suggested it.

You'll know you've been "Thompsonized" if I ask, "How does it start
up" ?

There are also those "designers" who rely on the rising edge-rate of
the power supply voltage to start such loops or operate POR's (as
we've discussed here this past week).

It keeps business coming in.

There's nothing sweeter than the voice of a COO/CEO begging, "Can you
_please_ fix it this week?" ;-)

The thing is they have to find you. I had come across many instances
where a design could have been saved but they had no clue whom to call.
There simply is no suitable consultants database anywhere. The ones
there are such as the IEEE database aren't advertised and none of the
engineers I asked had ever heard of it.

The most sad story was a client that never became one. They wanted me
and one of my networking peers for something and then probably decided
we would be too expensive. Tried it on their own. Later I met the former
owner and talked shop over dinner. They had a huge mechanical/analog
issue, or at least it seemed huge to them. Together with an ME I could
have fixed it but by then the company hand been gone. Too late. Man,
this guy could have retired and bought an island or something.

Regards, Joerg

 

[Jim Thompson]

On Sat, 25 Feb 2006 01:15:43 GMT, Joerg

[snip]

The thing is they have to find you.

[snip]

Yep. Word-of-mouth has its pros and cons :-(

I also run into a lot of "Cadence snobbery". Since I don't have
Cadence tools they often just walk away. Too bad. I've seen some who
got dumped by their boards/investors for failure to deliver.

...Jim Thompson

 

[Joerg]

Hello Jim,

Yep. Word-of-mouth has its pros and cons :-(

Thing is, the word has to find its way. The only path would be the
occasional dinner talk between big shots about some problem, and one of
them scratching his head and digging through his address book.

I also run into a lot of "Cadence snobbery". Since I don't have
Cadence tools they often just walk away. Too bad. I've seen some who
got dumped by their boards/investors for failure to deliver.

You can also see that in HW design. Not so much with respect to CAD
tools but lab equipment. Once a guy said that my HP3577 is kind of old,
ain't it? Turns out I had to schlepp it down there because their fancy
new one couldn't do it. Then there was the 20+ year old Rhode&Schwarz
UVM analog RF meter. The minute something modulated all their newfangled
stuff went beserk but not my UVM.

The best was a top of the line spectrum analyzer that had cost a client
the equivalent of a new Lexus. I hung up on it and came back with my
comm receiver, also top of the line but these things are under $3k.
Started at 9:00am, earned lots of frowns, found the noise source at
9:10am, determined a fix by 9:15am. They had set up a 10:00am meeting to
discuss the strategy to chase that problem. So I went to the boss and
told him we could still hold that meeting but there wasn't much to say
about noise anymore, other than where it comes from and what we did to
fix it. Dropped his jaw.

Regards, Joerg