Electrical contacts for Femtoamps measurement

Hello. I would appreciate any response, opinion on the following.

I have to measure current between to electrodes made of thin metal foils. T he current is in Femtoampere range (~ 100 fA). The problem I am having is i n making electrical contacts on thin metal surfaces, they are aroun 0.5mm t o 1mm surfaces and tried using thin (0.15mm) Copper wire, but it was very n oisy.

Does anyone know of techniques of making such delicate contacts? (Please no te that any destructive treatment is allowed on the metal, so, welding/ sol dering all are invalid options, I tried using conductive glue but that also has a lot of noise)..

Thanks a lot :)

Reply to
PR
Loading thread data ...

any destructive treatment is NOT allowed on the metal, so, welding/ soldering all are invalid options..

Reply to
PR

Your problem is more probably in the insulation than contact resistance at that current level, unless you want also signals at microvolt level.

I'd look at insulation and shielding of the front end.

Been there - done that.

--

-TV
Reply to
Tauno Voipio

Yup. The sample, leads, and front end amplifier will need to be surrounded by a a continuous metal shield, or it'll talk to everything in the universe--cell phones, electric motors, fluorescent lights.... This doesn't have to be fancy--a metal cookie tin works fine, but you have to be sure to bypass the power leads really carefully, and solder the shields to the case without leaving pigtails. I use some Russian

33-nF feedthrough caps from eBay, and they work great.

Also, of course, you'll probably have full shot noise, so your theoretical SNR is limited to

SNR_shot = N/2B

where N is the electron arrival rate and B is the bandwidth.

A 100 fA current is about 600,000 electrons per second, so your SNR wil go to 0 dB in a bandwidth of at most

B = 300 kHz.

You need a SNR of at least 20 dB to measure anything usefully, and usually a lot more, so you're looking at a bandwidth of a kilohertz at most. Calculating this is simple, and you'll develop a lot of intuition about your measurement very quickly.

Cheers

Phil Hobbs

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

Your noise is actually interference, is it not? Mains hum perhaps, or slower seemingly random excursions? I doubt it's the connections that are the source of the noise.

Why is soldering to the foils destructive? What are these foils made of?

I could measure ~150aA currents in an ionization chamber with an active volume of a few cm^3, so it can be done, but I do think we need a bit more information to make more directly useful comments. For the moment, keep in mind that unshielded high-impedance nodes will pick-up any e-field in the vicinity.

Jeroen Belleman

Reply to
Jeroen Belleman

Electrochemists here use capacitive discharge welding to bond the wires to thin films.

Steve

Reply to
sroberts6328

The current is in Femtoampere range (~ 100 fA). The problem I am having is in making electrical contacts on thin metal surfaces, they are aroun 0.5mm to 1mm surfaces and tried using thin (0.15mm) Copper wire, but it was very noisy.

note that any destructive treatment is allowed on the metal, so, welding/ s oldering all are invalid options, I tried using conductive glue but that al so has a lot of noise)..

Pogo pins? I made a sample probe thing with some pogo pins and an optics slide stage.

formatting link
(pic is a bit blurry.. sorry)

George H.

Reply to
George Herold

One more thing: The multi-gigaohm glass tube resistors do not like fingerprints. Please use protective gloves when handling.

The active input node must not be on a PC board. Use Teflon standoffs or just in-the-air connection to not lose your femtoamps to leakage.

Teflon is good enough to be an insulator at the femtoampere level, but it is microphonic, be prepared.

--

-TV
Reply to
Tauno Voipio

Can you apply pressure to the foils? If so, some sort of spring or pogo contact should work. If not, a wire and a conductive adhesive or grease (EKG lotion?) should work. If a little pressure is OK, use a conductive elastomer. Again, EKG stick-on pads would be interesting, if you're careful about contamination.

Noise is more likely to be caused by things other than the contacts. Below 1 pA things get very tricky.

How are you measuring the current? What's the applied voltage? How are things shielded?

--
John Larkin         Highland Technology, Inc 
picosecond timing   laser drivers and controllers 
 Click to see the full signature
Reply to
John Larkin

. The current is in Femtoampere range (~ 100 fA). The problem I am having i s in making electrical contacts on thin metal surfaces, they are aroun 0.5m m to 1mm surfaces and tried using thin (0.15mm) Copper wire, but it was ver y noisy.

note that any destructive treatment is allowed on the metal, so, welding/ soldering all are invalid options, I tried using conductive glue but that a lso has a lot of noise)..

Oh, silver paint or conductive epoxy. Maybe conductive epoxy, but don't add the hardener.. silver goop :^)

George H.

Reply to
George Herold

Pogos rock. I made a bunch of these little boards

formatting link

which can be used to bridge surface-mount parts for tuning and probing and such. The pogos can be bought cheap on ebay, by the pound, probably slightly used from bed-of-nails fixtures.

formatting link

--
John Larkin         Highland Technology, Inc 
picosecond timing   laser drivers and controllers 
 Click to see the full signature
Reply to
John Larkin

ls. The current is in Femtoampere range (~ 100 fA). The problem I am having is in making electrical contacts on thin metal surfaces, they are aroun 0.

5mm to 1mm surfaces and tried using thin (0.15mm) Copper wire, but it was v ery noisy.

se note that any destructive treatment is allowed on the metal, so, welding / soldering all are invalid options, I tried using conductive glue but that also has a lot of noise)..

They make micro manipulators for probe stations that cost big bucks. I've been toying with ideas to make a simple version with pogo pins and pcb. Doing the z-axis (up and down) is the hard part. A flexing "diving board" type thing, maybe? I don't like the motion y (or x) axis. That leads to a long board.

I am my own worse enemy :^)

George H.

Reply to
George Herold

Our big Rigol scope came with these

formatting link

which are third-hand probe holders. They are OK, but something better would be cool.

--
John Larkin         Highland Technology, Inc 
picosecond timing   laser drivers and controllers 
 Click to see the full signature
Reply to
John Larkin

oils. The current is in Femtoampere range (~ 100 fA). The problem I am havi ng is in making electrical contacts on thin metal surfaces, they are aroun

0.5mm to 1mm surfaces and tried using thin (0.15mm) Copper wire, but it was very noisy.

ease note that any destructive treatment is allowed on the metal, so, weldi ng/ soldering all are invalid options, I tried using conductive glue but th at also has a lot of noise)..

delta robot?

formatting link

-Lasse

Reply to
Lasse Langwadt Christensen

On Monday, June 22, 2015 at 1:06:39 PM UTC-4, Lasse Langwadt Christensen wr ote:

foils. The current is in Femtoampere range (~ 100 fA). The problem I am ha ving is in making electrical contacts on thin metal surfaces, they are arou n 0.5mm to 1mm surfaces and tried using thin (0.15mm) Copper wire, but it w as very noisy.

Please note that any destructive treatment is allowed on the metal, so, wel ding/ soldering all are invalid options, I tried using conductive glue but that also has a lot of noise)..

Grin, fun.. I saw a 3-D printing that works on the same principle. Easy for a computer, hard human.

(It's not an important problem..(manipulator) so it just "stews" on the back burner.)

George H.

Reply to
George Herold

The metal tin is a good idea, but only needed for shielding electric fields, usually at low frequencies. RF interference does not usually cause any special problems for femtoamp circuits, unless the offset voltages also have to be accurate to microvolts. Therefore there is no need to worry any more than usual about RF filtering, but be very careful of pieces of plastic inside the shielded enclosure that could get charged. This includes the insulation of wires.

Adding guarding around any input wiring is a good idea - drive the shield of the cable with the same voltage as on the internal conductors.

Here is a video of a buffer amplifier I built, with low leakage current:

formatting link

Reply to
Chris Jones

You're nuts. Rectification by the input amp will blow you out of the water. I build a lot of picoamp TIAs for people, trust me on this one.

Charging is one problem, and another is microphonics due to V*delta C (i.e. things with nonzero bias voltages vibrating.)

Not having any input wiring is a much better idea.

Cheers

Phil Hobbs

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

Here's a small PC board maximally trashed with rosin flux and fingerprints:

formatting link

The needle is pinned on the 1e14 ohm range.

And a couple of resistors measured, no special handling:

formatting link

formatting link

Polycarb seems fine.

formatting link

Those RatShack binding posts were very leaky.

--
John Larkin         Highland Technology, Inc 
picosecond timing   laser drivers and controllers 
 Click to see the full signature
Reply to
John Larkin

I have oppositely charged eletrode plates separated by thin alumina insulator plates. And after making all electrical connections i put it into an aluminum case which is grounded.

Reply to
PR

I am keep whole assembly inside an Aluminum box. I am using Keithley Picoammeter 6485. For connections i use bnc and coaxial cables.

Reply to
PR

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.