On 2013-06-29, Phil Hobbs wrote:
Phil Hobbs wrote:

On 6/28/2013 8:13 PM, John Larkin wrote:
On Fri, 28 Jun 2013 20:02:29 -0400, Phil Hobbs
wrote:

On 6/28/2013 12:34 PM, John Larkin wrote:
On Fri, 28 Jun 2013 08:44:57 -0700, John Larkin
wrote:

On Fri, 28 Jun 2013 10:33:35 -0400, JW wrote:

On Fri, 28 Jun 2013 06:51:11 -0700 John Larkin
wrote in Message id:
:

On Fri, 28 Jun 2013 08:45:19 -0400, JW wrote:

Looking for a suitable cross-reference for a National PF5301 TO-92 N
channel JFET as I can't find the original part anywhere. Hell, I can't
even find an original datasheet. The critical parameter for the sub is
Igss which for the National part is 1pA. The only thing I've been able to
find so far is a Linear Systems LS5301 which has a Ig of 1pA, but at $40 a
pop (!), I'd like to find an alternate.

Does anyone know of any possible substitutes?

Thanks.

What's the application? Do you really need a jfet?

I'm pretty sure I do, as it looks to be a fairly sensitive circuit. It's
for a repair on a Keithley 220 programmable current source.
http://www.google.com/url?sa=t&rct=j...eEEYYk4sluspuA

The schematic is on page 6-31/6-32 (4 pages from the end) location Q302
near the top of the page. It's part of circuit which uses a AD515K
FET-Input Electrometer Op-amp.

Thanks.

Yikes, it's a jfet darlington! Really a weird circuit, overall.

I bet a 2N7000 would work. It's just an AC amplifier. Gate currents on a 2N7000
are femtoamps. You can turn one halfway on and leave the gate open, and it will
stay halfway on for hours. Fun to play with.

https://dl.dropboxusercontent.com/u/...rts/2N7000.jpg


Hmmm, biasing might be tricky. It might be better to replace both jfets with one
2N7000.

Q302 has a lot of drain-source voltage. I'm wondering why it doesn't have a lot
of gate current.


Maybe it's a cascode internally?

Cheers

Phil Hobbs

Was there such a thing, a 3-pin jfet cascode? There were some cute
dual-gate gaasfets that worked great as sorta-cascodes.

I only recently discovered the jfet impact ionization gate current
thing, while working on one of your circuits actually. See AoE p 136.

Yup, I remember learning about it there. It really only happens with
N-channel devices, and only at V_DS above about 5V.

I recall some old National app notes referencing an integrated JFET
cascode, but they're at the lab and I'm at home. Older datasheets and
app notes on the web have been becoming inaccessible alarmingly fast
lately, due to all the parasites. Hang on to those old National Linear
Applications books, guys!

On another topic We're having a barbecue tomorrow, so I just picked up a
couple of torpedo kegs from Captain Lawrence, a first rate local
brewery. Their stuff is amazing, especially "Liquid Gold", which is a
clove-scented ale, and "Imperial IPA", which is nice and hoppy without
being one-dimensional like Dogfish.

If we can pry you and Mo off the Left Coast sometime, we'll take you to
their tasting room. (Picture a Napa Valley tasting room on the New
Jersey Turnpike, and you get the idea. Mindbending.)


Found it. National NDF9401 through NDF9410 monolithic JFET cascodes.
The text of the National 1977 FET databook is on archive.org (no images
unfortunately).

Hmm, JFET, so depletion; it was probably like this.

|
|
|--'
|
.---|--.
| |
| |
| |--'
| |
----|---|--.
| |
`-------+
|
|

JFET cascode

--
š‚šƒ 100% natural

--- news://freenews.netfront.net/ - complaints: ---