PMT pulse amp

What comic books? I bought electronics magazines.

If they're so great in parallel, would they be even better in the form of a distributed amp, or wouldn't it be worth the bother?

Thanks, Rich

They have a unique combination of low voltage noise, about 0.7 nV/sqrt(Hz), high g_M (~30 mmho), and decent f_max, around 700 MHz. You can put a bunch in parallel to get amplifier noise and bandwidth performance you can't touch any other way that I know of. And they were designed for AM radios, so the price is right too.

I feel the urge to buy another reel. ;)

Cheers

Phil Hobbs

That's all? I have toobz with more than that! ;-)

Four in the back, only one visible glowing:

Tim

The 531/535 got 15 MHz without distributed amps. The 545 (I have a couple) got 30 MHz using distributed amps.

John

"John Larkin" wrote in message news: snipped-for-privacy@4ax.com...

Aha, 531 used 6BQ7 driving 6197, but 541 got the distributed amp!

Of course, the sweep amp is just a few 6DJ8s, but reduced horizontal bandwidth is typical.

Interesting. I should tack on some more 6DJ8s then. There's enough room below the CRT for two pairs of cathode followers.

Tim

What types of special things can you do with them?

Clifford Heath.

The original 541 made do with 6AK5's, which have a rotten Gm/Cp figure of merit. Some modern frame-grid pentodes would work wonders.

Tek was especially skilled at using t-coils for peaking. Regular peaking gives numbers like 1.5:1 improvement, but a t-coil can hit

I don't suppose I could interest you in trying some GaN fets...

John

If I weren't too old and decrepit to look after it, I'd kill for one of those.

Didn't they use (several) of the Packard engines in PT boats?

Don't know about that, but it'd do >100MPH on the road :-) ...Jim Thompson

-- | James E.Thompson, CTO | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona 85048 Skype: Contacts Only | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at

"John Larkin" wrote in message news: snipped-for-privacy@4ax.com...

I wonder if anyone has since tried it on monolithic InP. I recall distributed amps reaching 80GHz. They can probably do Dc-1THz directly today...

Tim

Several people make GaAs distributed amps, used mostly to drive e/o modulators in telecom apps. The upper end is something like 10 volts p-p into 50 ohms, wideband to 30 or 40 GHz maybe. Tricky to bias.

Discrete GaN fets have a Gm/Co figure of merit that's thousands of times better than any toob... a few siemens and a few pF. And they have enough breakdown voltage, 50 and up, that they could drive deflection plates. I wish I'd had some when I was a kid, instead of gassy tubes from old TV sets. The exotica, nuvistors and frame-grid tubes, were late to the party, just as transistors were starting to get good.

John

"John Larkin" wrote in message news: snipped-for-privacy@4ax.com...

But bias is just DC, i.e., under 100MHz. Nothing exciting down there...

Speaking of volts, GaAs has plenty of bandgap; wasn't it supposed to be a magic power device, like they're working on making in SiC now? I know 50V is convienient for RF, but only to a point... those big fat power amp transistors draw upwards of 10A, which is a *lot* of dI/dt for 900MHz or what have you. You don't have any choice at that point but to wire a million in parallel to make your power output. And that gets even more expensive.

I suppose this might be one reason why IOTs are still used for TV?

Some of the transitional Tek equipment had Nuvistor front ends and a few transistors inbetween, amidst mostly tube circuits. Man, >100MHz transistors from the early 1960s...no wonder Tek was so expensive...

Tim

A Packard engine, but not the same as used in cars (a bit heavy ;-).

You've got to actively adjust two or three nodes, in the right sequence, to get them to work and not blow up. At numbers like $200 a chip, one does this carefully.

The Japanese GaN company, I forget the name, has some 150 volt parts. Nytronex, whose parts we played with, are modtly in the 50 volt range. GaN fets have absurdly low pinchoff voltages, and enhance nicely up to maybe +0.4 volts. The SiC parts I've seen were more like toobs, -20 pinchoff and no enhancement to speak of. Mut maybe recent parts are better. In my business, switching, I like the GaN parts because I can drive them directly from a fast opamp or even an ECL gate.

My 547s (don't ask how many I have) are weird mixes of tubes, transistors, and tunnel diodes.

John

If

So how does one go about wringing things like the trr of the body diode out of the S-parameters? :-)

demo

box.

more than

Or they get rid of us and then, together with the bankers, try to live off of the leverage that was already leveraged :-)

Oh yeah! I've had people banging on the door to give them my last five but I need those as spares for my network analyzer jig.

LH0033s are also nice but can't pack quite the same punch.

Yep. But put them in a nitrogen cabinet.

a demo

box.

more than

cheap,

As everything around them collapses.