It's usually pretty humid here in SF. Not right this instant, as we're having our usual October hot/dry spell, with the wind from the east. It's up to 74F, unbearable.
Breathing on my electrometer test board doesn't un-pin the meter.
That's interesting. Thanks Jon. If you have ~> 100M ohm and you breathe on it and it doesn't spike down, then it's not working. (at least my circuits.) I've had all sorts of problems with plastics.... mixing IC sockets and high impedance is risky...
I was thinking about my particular problem and realized that the tech who got fired a few years ago was the guy who took care of the cleaning bath... It may be filthy. Now I've got a monitor. :^)
George H.
What's your solvent? I've got acetone, 91% IPA and ethanol (denatured.)
I mostly just want to make the boards look nice.
George H.
OK my "spike down" may be a change of ~10%. George H.
We use a mix especially designed to clean boards.
That stuff costs more than good bourbon, but it's recycled (constantly distilled) so we don't lose much of it.
Probably tastes better too.
Cheers
Phil "not a whisky fan" Hobbs
That's a mix designed as an azeotrope (so you can evaporate/condense/rinse/repeat without composition change), and dissolves your particular flux (it wouldn't clean water-soluble very well). It targets flux, not boards.
The wetting-agent-added recipes rely on sheeting action, to drip-dry the inner crevices, and ought to work well on water-soluble fluxes. Wetting agents aren't generally part of an azeotrope (though alcohol does lower surface tension a bit).
The devil is in the details, though: what mix of oxides, organics, and salts is in a particular board after reflow? It depends critically on the (mainly-unknown) chemistry of the solder paste. It also depends on the components on the board (a humidity sensor often requires no-wash processing, for instance).
The ADA4530-1 datasheet (Analog Devices:Femtoampere Input Bias Current Elec trometer Amplifier) offers cleaning guidance (p38) and solder paste selecti on for this extreme application. Though it does not refer to SMD resistors the SOIC amplifier chip has small clearance from the PCB. My 100Tohm transi mpedance amplifier feedback resistors are more conveniently in large glass enclosures and so are easier to clean. Regards, Scott Hamilton.
ectrometer Amplifier) offers cleaning guidance (p38) and solder paste selec tion for this extreme application. Though it does not refer to SMD resistor s the SOIC amplifier chip has small clearance from the PCB. My 100Tohm tran simpedance amplifier feedback resistors are more conveniently in large glas s enclosures and so are easier to clean.
Nice thanks...
Everything got cleaned with IPA.
Wow.. 100T ohm. (and I thought 1 Gig was living dangerously :^) Can I ask what sort of time constant you get with 100 T ohm? Do you use a bootstrap to get the capacitance down...?
George H.
I have some samples of a 1T 0805 surface-mount resistor.
Fun. I bet you just got out the Windex bottle to give you something to do while you wait for the TIA to settle. ;)
I've used 50G in 0805 in a noncontact surface potential tool for the late lamented Qcept Technologies:
RES 50G OHM 1/8W 30% 1000ppm 0805 1625862-4 TE $3 Digikey
That was in a resin flux SnPb system with solvent wash.
1206es and 1210s are better, because you can put a slot underneath.Cheers
Phil Hobbs
Hmmmm...Victoreen resistors from the '50s?
Electrometer Amplifier) offers cleaning guidance (p38) and solder paste sel ection for this extreme application. Though it does not refer to SMD resist ors the SOIC amplifier chip has small clearance from the PCB. My 100Tohm tr ansimpedance amplifier feedback resistors are more conveniently in large gl ass enclosures and so are easier to clean.
0-1.pdfTime to within 1/e of final ~1sec, settle to within noise level ~5sec. Not clear what form of bootstrap you refer to. I have often thought I was invol ved in what John Larkin refers to as 'lunatic fringe electronics'. Phil Hobbs: on power up recovers in about 20sec. Robert Baer: I have some venerable Victoreen resistors but unfortunately th ey no longer manufacture above 10Tohm. Presently use Welwyn 3812 100T resis tors.
Dare to be stupid!
I don't think the glass tubes are necessary. Digikey sells multi-Gohm epoxy-dipped radial resistors and they work fine.
Agressive fingerprinting didn't change the indicated 1G value.
Here's a 100G 0805:
and a 1G Welwyn
Gigohm stuff isn't hard provided you keep it away from water.
100T and 1 pF is a 100 second tau!
t Electrometer Amplifier) offers cleaning guidance (p38) and solder paste s election for this extreme application. Though it does not refer to SMD resi stors the SOIC amplifier chip has small clearance from the PCB. My 100Tohm transimpedance amplifier feedback resistors are more conveniently in large glass enclosures and so are easier to clean.
530-1.pdft clear what form of bootstrap you refer to. I have often thought I was inv olved in what John Larkin refers to as 'lunatic fringe electronics'. Grin... I'm not sure what type of boot strap I meant either... but you've got the capacitance of the resistor that has to be charged. One second sou nds amazingly fast!
George H.
they no longer manufacture above 10Tohm. Presently use Welwyn 3812 100T res istors.
nt Electrometer Amplifier) offers cleaning guidance (p38) and solder paste selection for this extreme application. Though it does not refer to SMD res istors the SOIC amplifier chip has small clearance from the PCB. My 100Tohm transimpedance amplifier feedback resistors are more conveniently in large glass enclosures and so are easier to clean.
4530-1.pdfot clear what form of bootstrap you refer to. I have often thought I was in volved in what John Larkin refers to as 'lunatic fringe electronics'.
they no longer manufacture above 10Tohm. Presently use Welwyn 3812 100T re sistors.
JPG
hley.jpg
(oops I slipped a few decimal places.. I was getting 10,000 seconds.)
100T is as far away from 1G as 1G is from 10k.George H.
You're a physicist, right? ;)
A one-second TC with 100T ohms? Hmm. There's no way that resistor has as little as 10 femtofarads of self-capacitance, let alone whatever you have it attached to, so I'm gathering that you're using a resistive guard tube around it (sort of a distributed bootstrap).
The noise gain of that TIA must be something pretty terrific even so.
Cheers
Phil "also a physicist" Hobbs
On Thursday, October 13, 2016 at 8:41:49 AM UTC-7, John Larkin wrote: [about gigohm resistors]
You can fill with inert gas, though: a carbon film resistor can oxidize easily to a new value, and high voltages create ozone to make that happen FAST. Glass and inert gas, or silicone, are good for device lifetime.
An acquaintance was building things for use in vacuum, and got the factory to sell him uncoated resistors. They were happy to skip the high-tech 'paint' process.
The modern resistors are thick-films. Maybe the old Victoreens were carbon and needed the glass for protection.
Hrm, the datasheet says "The guard band is described, with application notes, in a Product Information sheet, available on request."
Ah, here it is:
Have something to add? Share your thoughts — no account required.
Ask the community — no account required