interviews

Pentium divide bug did get out. _Only_the_Paranoid_Survive_ by Andrew Grove, has a lot to say about that.

That's not what I remember. The 617 electrometer, and 427 amplifier used jfets (and a very clever biasing scheme that basically kept Vds at a fraction of a volt). When did they change?

"very obscure"

Here's the manual:

Do you know why a man gets excited by a woman dressed in leather? Because it makes her smell like a new pickup truck.

Don't know. The 610 schematic is dated 1968.

whatever it is that gets the plane into the ocean really fast.

The F16 had the feature when it crossed the equator but at least they found it in simulation.

The MOSFETS are Q101,2 in schematic, but in the parts list, they are denoted by double-asterisk. (**) The ** is nowhere to be found.

So it's a mystery what the mosfets were. Their replacement by jfets may be a tacit admission of their instability or just part unavailability. My Keithley

jb

The mosfets are part of a subassembly. They were apparently selected parts so are not individually repleaceable. I assume they matched them for offset and selected them for leakage.

Were they ever replaced by jfets? Jfets have more leakage current.

I'll have to check.

I was interested in the mosfets because I had experimented with ion chamber circuits that measure small currents - fA range. Ion chambers operate at biases down to 5V, so are just collecting ion trails in air.

Glenn Knoll is the world's expert on ion chambers.

techlib.com has a nice circuit, which uses a picaxe to collect values via a

But the fact that mosfets are true field effect devices makes them sound interesting to me.

However, the Geiger counter is a better design, if you can get around the vacuum requirement.

Any idea of a circuit that would use them?

I designed the electronics for an interesting gadget. It looked like a shopping cart that you rolled around a floor looking for radioactive particles. The bottom was a roughly 1m square coarse pixel detector. The detector was a serpentine tungsten wire sandwiched between two metalized mylar sheets, filled with some exotic gas. There was a high voltage on the wire, enough to give a lot of ion chamber gain but not quite geiger mode. We could figure out the hit locations by comparing the peak amplitudes that came out the ends of the wire.

This was a test program that I wrote.

It worked, but never saw production. The company founder was old, and his son was flakey, and the business died before the system could be marketed. But I learned some useful stuff. Even projects that fail can teach useful stuff; in fact, they usually do.

ha!

what did the plane try to actually do?

It did a 180 roll (a little sign issue).

Den tirsdag den 24. juni 2014 18.17.44 UTC+2 skrev snipped-for-privacy@attt.bizz:

it's a nice story and all, but I'm not sure anyone know what it did if anything

The nearest thing to source for the story is a 30 year old usenet posting that begins with "a friend of mine .."

it has urban myth written all over

-Lasse

Perhaps but it was written about in the rags at the time as an example of things that hadn't been though about before FBW came on the scene.

reportedly killed a simulated pilot.

google can't find the original post.

Why was he interested in a radioactive particle detector on the floor?

So, what DO they know?

Sylvia.

Because it's a Bad Thing to have radioactive particles on the floor in any active or former nuclear facility, and scanning with a small detector is very inefficient. He also sold and rented people inspectors, things that employees would walk through coming to and leaving from work, to check for radiation. Some of those were mobile, shipping containers, for jobs like decontamination and disassembling nuclear stuff. Decomissioning was a big part of his business.

It was a good business, but he had no workable plan for succession.