lessons taught

I remember a particular problem once with HCT564's (flip-flop's).

Production 2nd-sourced what was supposedly an exact replacement part. Turns out, the 2nd-sourced part had slightly better static discharge performance than the original, but was otherwise identical in every respect (at least on paper).

You guessed it: IT SIMPLY WOULD NOT FUNCTION in the target hardware. (Not function weirdly or erratically, some of the time. Would NEVER work!!) It took a while to figure this out.

As to why so long? "Hey", it's a '564. How complicated can it possibly be?!

I remember asking Harris for a sample on an 74HC5xx part once. They refused saying these were not complex parts. lol

--

Rick

Meaning, I take it, "If you can't afford to buy a 50-cent chip from Digikey, you aren't a serious customer."

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

A 50 cent chip at a DOD contractor costs over $100 and takes a bunch of my time. You would think Harris would get that, but I guess different divisions.

I was actually told that it costs $100 per line item and that was in the

80's. They were thinking about direct charging that rather than adding purchasing's costs to overhead. I expect they do direct charge common services in DoD companies these days... *and* add overhead.

--

Rick

This would have been magnet wire, what was left over on a spool from MWS.

I'm not sure what hard drawn wire is, but I'm guessing more than the

25% length change.. :^)

George H.

Take the four bits out of your pocket?

I would think that if you're using Plutonium wires, conductivity isn't going to be your biggest problem.

PCB layout people have enough things to worry about already, without dealing with criticality.

--

John Larkin         Highland Technology, Inc 
picosecond timing   laser drivers and controllers 

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

I always let our layout people know what the critical nets are.

If they're already critical, everybody nearby is dead anyway.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

Measure, mark, cut 1st approximation, re-measure, mark, trim, verify measurement,

works pretty well IRL. "Measure twice cut once" is too open-loop--lots of mistakes that way.

Cheers, James Arthur

+10,000K

Cheers, James Arthur

On Saturday, May 30, 2015 at 5:41:29 PM UTC-4, Lasse Langwadt Christensen w rote:

imes

bad and

s "Don't

cut

.

Thanks Lasse, Though my computer had a hard time viewing that document.

So hard annealed is 2-3% higher in resistance.

George H.

Don't know about magnetic layers, but I worked on a machine with 18 layer boards with an omega layer for termination resistors. It saved a lot of space using ECL logic.

--

Rick

We have a guy working with us that came from a tyco division that makes boards, they go up to 64 layers. Lots of mil boards are done there.

Jamie

Embedded resistors and caps (especially power planes) were a big thing once. Do people still do that?

--

John Larkin         Highland Technology, Inc 
picosecond timing   laser drivers and controllers 

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

The problem past 10 or so layers is impedance control. To keep reasonable traces, even 50 ohms, either the traces become impossibly skinny, or the board thickness explodes.

We prefer 6 or 8 and have done a few at 10, required to get the signals out of a big BGA. I actually did a TWO layer board earlier this year!

--

John Larkin         Highland Technology, Inc 
picosecond timing   laser drivers and controllers 

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

And hard to bend.

--

John Larkin         Highland Technology, Inc 
picosecond timing   laser drivers and controllers 

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com