PCB

Hello David,

When hiring analog engineers I only needed to ask one question. Gave them a single-BJT RF amp schematic and asked what they would do to stretch the high frequency gain a little. If they couldn't answer that this was pretty much the end of the interview.

Regards, Joerg

Were you looking for suggestions of passives to add/change or was "I'd buy a better transistor" considered a sufficient answer? :-)

WEll, man, don't keep us in suspenders. Pass the schematic on and let us take a crack at it. Other than rebiasing it for max GBW or tossing in a peaking coil of some sort, I'd be hard pressed to stretch something that is unstretchable. Doesn't the device itself set the limits on high frequency gain? Unless you've purposely done some sort of feedback to limit the high end.

Jim

Hello Joel,

It was more along the line "... for an additional 2 cents" :-)

Regards, Joerg

Hello Jim,

It was pretty simple. Hotshot RF transistor, biased at 8mA or so, collector resistor, emitter resistor (that's where they'd have to put a series RC across). IOW the stuff in the "Basics" section of AoE. Believe it or not, most wouldn't know. And I am pretty sure they didn't know what AoE is either.

The candidate that passed I asked what he'd do if he had to drive something capacitive really hard and fast. Then he began to sketch how he'd done that to drive inkjet cartridge piezos. Well, I hired him and he still works there :-)

Regards, Joerg

Good interview question:

What's the last thing you worked on? Sketch some schematics.

If they mutter something about it being proprietary, subtract 5 points and try another approach.

John

Hello John,

Oh, I told them right up front not to devulge anything about previous jobs. This stuff had been published. But it wouldn't have mattered whether he was the originator or not. What matters to me is whether they are able to find out. If someone would have said "Page xyz in AoE" that answer would have been good enough.

Regards, Joerg

I did a software interview like that once. "Here's a computer. Write me a tic tac toe program using XYZ. You have 30 minutes. Begin". 30 minutes later, I knew how well he knew that software.

That's like asking a carpenter to hammer in a nail.

Depends what XYZ is. More like asking a carpenter to *design* a house, or plan how to build it (like a GC). In my case, XYZ was GUI toolkit, and it took most of the 30 minutes just to plan the whole thing. Watching how he approached the problem told me a lot.

Hello Homer,

Which doesn't mean everybody can do it. When I repaired part of a stair case that had fallen apart I quickly realized why it had: Six out of six nails had missed the beams underneath. They were just kind of dangling there.

Regards, Joerg

Perhaps his test was "saw a bit of wood"!

I can't understand it! I've cut this damn board THREE TIMES and it's _still_ too short!

I figure the first time you hit your heard on an exposed roofing nail in your attic, while you're sitting there contemplating what a lovely mess of red your hair's going to be (since scalps like to bleed so profusely) and when you had your last tentanus shot -- that's a good time to take a good long look at the roof, and see how often the roofers missed the framing... and mentally review the likelihood of hurricanes or tornados where you live.

:-)

And if (s)he was REALLY on the ball, they'd take lead length of the capacitor into account as an inductor and give you a series resonant short circuit at the frequency of interest across the emitter resistor.

Jim

Hello Jim,

I wish. What I found a lot instead was that there were a great many "engineers" who didn't even understand the basics of a single transistor amplifier. I mean, they didn't seem to grasp how that thing would work in the first place.

Regards, Joerg

Try this interview question:

What is the base voltage? (seriously!)

What is the emitter voltage?

What are the base, collector, and emitter currents?

Any other comments?

This one usually befuddles resume-heavy types. Maybe 10% of applicants for ee or tech positions understand this circuit!

(Extra credit if he mentions oscillation)

I get great comments, like "the base voltage is 5, so the transistor must be saturated, so the emitter voltage is 10" and "the base voltage is 0.6"

If they can't handle that, try this

or

John

Or, you've got a dark closet underneath a staricase, and you crawl in on your knees to explore it, and discover the carpet strip the hard way. =:-O

Cheers! Rich

Hello John,

Like, how many popes named Pius were there before Pope Pius the 12th?

These usually didn't even make it into my office. Credentials are very low on my radar scope when hiring. With very few execptions I found that excellent credentials are in no way an indication of true performance. Much less so than callusses.

Considering what comes out of academia these days that would almost warrant a Nobel prize nomination.

Regards, Joerg