Low current input stages: which soldering flux is most appropriate?

[tggzzz]

I'm planning on building a front-end (for a scanning tunnelling microscope) that measures very low currents doen to 10pA, say. Hence leakage currents are somewhat important.

I can design and build the PCB for low leakage, but which soldering flux is most appropriate in terms of minimising leakage? Post-assembly cleaning is not a problem, provided the chemicals are obtainable.

Thanks

[krivx]

I would look at app notes for high input impedance parts, like the ADA4530-1.

This document has cleaning procedures for fA level measurements: http://www.analog.com/media/en/technical-documentation/user-guides/ADA4530-1R-EBZ_UG-865.pdf

This is also an interesting thread: https://www.eevblog.com/forum/projects/picoammeter-design/

[tggzzz]

I would look at app notes for high input impedance parts, like the ADA4530-1.
This document has cleaning procedures for fA level measurements: http://www.analog.com/media/en/technical-documentation/user-guides/ADA4530-1R-EBZ_UG-865.pdf
This is also an interesting thread: https://www.eevblog.com/forum/projects/picoammeter-design/

Thanks. Those are useful background information, but don't mention the best type of flux to use.

[tooki]

Well, assuming you do a thorough cleaning, such that no flux residue remains, then it really shouldn't matter what flux you use. Rosin fluxes are easily removed with common flux remover, or even just with alcohol. My hunch is that in your application, the quality of flux remover, and the skill with which it is used, is far more important than what type of flux it is.

[Kleinstein]

If you can clean the circuit afterwards, the choice of flux is not that important. It helps if you don't need to solder to hot and long, so less flux desintegrates - this is the part thats more difficult to remove.  It sometimes helps not to put sensitive nodes to the printed circuit board at all, but use air wiring, even if this may not look pretty.

[tggzzz]

Well, assuming you do a thorough cleaning, such that no flux residue remains, then it really shouldn't matter what flux you use. Rosin fluxes are easily removed with common flux remover, or even just with alcohol. My hunch is that in your application, the quality of flux remover, and the skill with which it is used, is far more important than what type of flux it is.

OK. Any recommendations for flux remover?

I've seen statements that some fluxes contain polymers. I don't know whether that is true nor what should remove it.

[Fungus]

It sometimes helps not to put sensitive nodes to the printed circuit board at all, but use air wiring, even if this may not look pretty.

Or cut slots in the PCB.

[tggzzz]

If you can clean the circuit afterwards, the choice of flux is not that important. It helps if you don't need to solder to hot and long, so less flux desintegrates - this is the part thats more difficult to remove.  It sometimes helps not to put sensitive nodes to the printed circuit board at all, but use air wiring, even if this may not look pretty.

Air mounting beneficially increases resistance and reduces capacitance, but even so, flux will have to be removed.

Which flux and remover combinations are particularly effective?

[tooki]

Any commercial flux remover will do a fine job on whatever (non-water-soluble) flux you throw at it. Seriously, you're way overthinking this. Find a flux you like first — that's the harder (but still easy) part. Then use whatever flux remover they sell at your electronics supplier.

If this is going to be mass-produced, you can use a flux remover bath to soak off the flux, but this MUST be followed by a rinse with virgin solvent or it'll leave a film of flux on everything. If it's just a one-off board, then buy a spray can of flux remover, douse the board, scrub with the attached brush if necessary, and keep applying fresh solvent to rinse off any traces of flux. NASA-style soldering uses flux remover (I think they use simple isopropyl alcohol) brushed on through a Kimwipe, repeated a few times to absorb all the flux.