What's good a sub picoamp op amplifier

Hi all

I was looking a very low bias current op amp. Less than picoampere. What I found had either a huge maximum limit at bias current or they are discontinued, nonexisting or not available.

Supply voltages and so on are not so critical and frequency is low.

I am testing a couple sensors which give out (DC)current around picoampere.

Reply to
LM
Loading thread data ...

formatting link
oduct.html

Reply to
Fred Bloggs

Try chopper amplifiers.

Reply to
Sjouke Burry

LMC660C is 0.002pA typical, but 2pA at temperature extremes, slightly outside your spec but perhaps only at high temperature.

--

John Devereux
Reply to
John Devereux

OPA129? Not cheap.

George H.

Reply to
George Herold

OPA128LM max 75fA @ 25°C Ta (warmed up).

Or if that's too rich for your blood, buy a couple $2 LPV521 (10fA typical at 25°C) and test them yourself.

Most of the high numbers are either because they don't bother testing the parts (I think it takes some time and care to test to fA levels) and/or the limit is at high temperature (leakage current roughly doubles every 10°C).

Reply to
Spehro Pefhany

I recall National Semi having some "fA" parts...the LMP7721 comes to mind.

Reply to
gravpoet

Here's my little fA tester box.

ftp://jjlarkin.lmi.net/99A260A1.JPG

ftp://jjlarkin.lmi.net/99A260A3.JPG

ftp://jjlarkin.lmi.net/99S260A.JPG

I can resolve maybe 20 fA. With a 1T resistor as Z1, you get 1 volt per pA.

Or buy an old Keithley electrometer on ebay. Great to have around.

ftp://jjlarkin.lmi.net/Keithley_1gig.JPG

The low range is 1e-13 amps full-scale.

John

Reply to
John Larkin

Oh, a 2N7000 typically leaks electrons per second. You can have fun with that.

John

Reply to
John Larkin

I got a Keithley 410 rack-mount electrometer for $6 on eBay. Works perfectly, if you don't mind waiting half an hour for it to warm up. (I should wire a Christmas light bulb inside the box to keep it warm.)

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal
 Click to see the full signature
Reply to
Phil Hobbs

Between gate and source? What kind of fun?

John

Reply to
John - KD5YI

Thank you all for these interesting devices (fet too). It is getting late, so I will check them later, but they look good.

Reply to
LM

Battery, led, resistor, fet. Let the gate hang open.

ftp://jjlarkin.lmi.net/2N7000.jpg

Touch B+, then touch the gate.

Touch B-, then touch the gate.

Use a pencil or a small plastic-handle screwdriver to transfer small packets of charge here and there.

Touch the drain and the gate simultaneously, then let go the gate. Then wait.

Walk around the room, then wave your hand around the gate.

Really fun: make an RC lowpass filter with tau about a second, drain as input, cap to ground. Take a wire from the C and tap the gate. Play with that.

Stuff like that. We geeks are easily amused.

John

Reply to
John Larkin

OMG! You are a Wallace & Grommet fan? I have all their tapes (which I copied to disks). I also have a Wallace stuffed doll...er, figure still inside its plastic cover. Did you notice that the penguin appeared in more than the W & G 'toons?

John

Reply to
John - KD5YI

Shame their barn burnt down. Most of the catalogue was destroyed.

Cheers

Reply to
Martin Riddle

ftp://jjlarkin.lmi.net/elecdog.jpg

John

Reply to
John Larkin

Oh, c'mon, John. You stole that off the Yahoo Eagle groups?

John

Reply to
John - KD5YI

Measuring for such leakage is time consuming at final test for the manufacturer. I suspect datasheet limits are guardbanded big time, so your 1pA room temp part is probably well under 1PA. If you can characterize your setup, you will probably find the 1pA part is over a reasonable temperature range.

Reply to
miso

"Cheeesse Gromit"... shaking fists.

I'll have to try it.

George H.

Reply to
George Herold

It is not that they do not bother (testing every part),it takes too damn much time to measure that level with automated testing equipment (using custom interface for DUT). Time is money. How many seconds are there in an 8-hour work day? Round up to 30 thousand for a crude number. Lose 200mSec for handling (automated loading/unloading) [optimistic number], leaving 800mSec for testing ALL parameters. Say all other parameters are "easy" and take 300mSec total leaving

500mSec integration time for ONE current test (charge capacitor with that bias current to give some reasonable voltage to measure). This guesstimate of times may be optimistic; YOU tell us if this is way off.
Reply to
Robert Baer

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.