Re: femtoampere

Mar 24, 2023 Last reply: 3 years ago 9 Replies


>


> > al from Technical University Berl> >>
> >> how are very low currents (fA range) measured?
> >
> > See:
formatting link
> Read the "Teflon" and "Femtoampere" articles.
> >
> > [edited] The best "Teflon Capacitor" we tried was a few
> > centimeters of RG-188 Teflon coax (outer braid to output
> > and center to the summing point) "trim" by snipping off
> > small pieces. 10-15pF is a convenient value for measuring
> > in the fA range. use the second channel of the 662 dual
> > for the integrator...
> > Paul Grohe (The "Paul" mentioned in the Teflon article).
> Yes, Paul, I remember your name in that fascinating account.
> One painful issue, as Pease discusses, is a suitable circuit
> to reset the integrator. Bernie Gottschalk, an engineer at
> the Harvard cyclotron, used NSC cmos opamps and had a clever
> solution: reverse the supply voltages. He used a resistor in
> series (perhaps 3.3k, etc.) with both opamp supply leads to
> limit the reversal current. The resistors with the opamp's
> protection diodes act as a bridge to zero the capacitor (the
> opamp's + input is grounded, although I'm not sure that's
> important). Significantly, when restoring the supply-voltage
> polarity the NSC opamp's protection-diode charge-transfer is
> balanced, leaving the integrating-capacitor voltage properly
> reset. Of course, in integrator applications like this, the
> exact capacitor voltage at reset isn't very critical.
> Very impressive, cute and elegant!
> A single HC4053 chip could be used to make a +/-5V supply-
> reversal circuit.
> The scheme is especially attractive with many channels, where
> one supply-reversal circuit resets all the channels at once,
> without any extra parts!
> I haven't tried this trick myself, but Bernie didn't mention
> any problems or special considerations (e.g., no additional
> bypass caps for the opamps after the series resistors, IIRC),
> so try it and let us know how it works out.
> Thanks,
> - Win

Just exhuming this very interesting thread from 20 years back. Recently we were talking about the trick of inverting the power supplies to reset an op amp integrator, and I think this is the first mention with attribution.


Cheers



Phil Hobbs


The best integrator is a single capacitor, with some sort of mechanical contact to connect it to the readout+reset circuit.

One could probably integrate a few electrons per second. Cosmic rays might be the biggest error source.

On 2023-03-24 13:35, John Larkin wrote:> On Fri, 24 Mar 2023 10:13:44

-0700 (PDT), Phil Hobbs > snipped-for-privacy@gmail.com wrote: >

For certain values of 'best', sure. It's often important to keep the terminal voltage of your integrator low, which is a lot easier to do with an active deice. >

Cheers

Phil Hobbs>

Anyone know if Win is keeping well? It's been a while since he's posted. Hope he's not upstairs arguing with Jim.

Hope he's not upstairs arguing with Jim.

Last time he disappeared, we were rewarded with AOE V3. And all the best to TUB, even if the reference is old.

Gerhard

I asked Paul Horowitz about Win, less than a year ago, and he said that Win was fine.

I e-mailed him about six months ago about a personal matter, and got a aimiable reply. At our age that doesn't mean that he is still okay.

Thank you Bill, I'm glad to hear that he has just decided that there's better things to do than post on SED. I can't say I blame him, it seems to have turned into an open sewer over the last few years.

It's not a fake name - I've got his e-mail address, and he's the guy who gave me a better .pdf of the Baxendall class-D oscillator paper than the one the publishers had sold me. Fred Bloggs *is* a fake name, so you are throwing rocks from inside a glass house.

And Flyguy, Commander Kinsey and a a are doing their level best to make s.e.d. an open sewer. All three post stuff that ought to be flushed away.

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required