Myth busting solder-flux cleaning with isopropanol alcohol

[rhb]

I had a Harbor Freight PIR sensor night light that had stopped working and been tossed in the junk bin.  I pulled it out yesterday to see if I could scavenge a AA batter holder and in the process took it apart.

It's a pretty simple device, so I thought I should be able to diagnose the fault.  In looking at it I noticed some flux residue.  So I cleaned it with a spray of isopropyl and an old toothbrush.

Now it works.  So I'll have to put it back together instead of removing the PIR to play with.

The surprising part was how little flux residue there was and that what I noticed was not close to the sensor or op amp, just the on-off-auto switch area.  I cleaned the whole board, so there may have been some I did not notice

[T3sl4co1l]

PIR detectors are pyroelectric, i.e., the electric field on a crystal changes with temperature.  Electric field means very high impedances, low leakage -- exceptionally so at the rates of change used here (low Hz).

If that's seen at board level or not, depends on if the sensor is internally buffered.  I think that's not uncommon, so you get a PIR module that looks very much like an electret mic: an internal JFET follower on the capacitive (charge based) element, giving reasonable circuit values (continuous current (fractional mA?) instead of charge).  Board contamination could still be a problem, but it would have to be much worse to kill operation.  Or the circuit is just that badly designed (you can always make a circuit worse; you can't always make a circuit better!).

Tim

[SeanB]

No clean flux, or water soluble flux poorly cleaned, will leave a residue that affects high impedance circuits. as the detector typically use 100k or higher value resistors to get gain, any board leakage on either top or bottom will cause incorrect operation. Might work when new, but after a few months of absorbing ambient humidity and rain there will be conductive paths that are not intended. Clean board and give a coat of some conformal coating, or just mask off the pyro detector top with painters tape and spray board top and bottom, avoiding the variable resistors using more tape, and they will last for years.

[rhb]

The circuit is very similar to the example in the datasheet for the PIR D203B sensor.  But it's driving an LED and timer circuit rather than a relay and uses an LP324 quad op amp.  It also has a sensor to keep it from turning on if the room lights are on

Putting it back together was a huge pain.  Every time I soldered a wire back on, another one broke off.  But I did eventually get it back together and it works just fine.  So I washed the "NG" written on the back off and put it to use.

I've not had trouble once the muck was cleaned off.  I did have to reclean a CD player after a few months, but that was several years ago.  It's never had a problem since.

[TheNewLab]

I did a little video on common alternatives

TLDR version:  Use real flux cleaner (or isopropyl if you don't have it)

You really should have included brake cleaner in your test. Cleans at least as well as the MGC product with zero board or component damage, and MUCH cheaper. I swear by it now.

I am also about to buy an ultrasonic cleaner and wonder to bath solution to use.
1st, some auto parts, and Iknow what to use,
2nd electronic parts and boards, and that I do not know.
Does Brake cleanerreally  work that well and is safe on circuit boards?
IF so, that would be ideal. Do you dilute it?

[TheNewLab]

Also, regarding ISO alcohol. I use 99% and I am not always happy. Some have mentioned how bought flux cleaner has additional chemicals
Can I make up a solution combining alcohol with other chemicals to make flux cleaning and PCB cleaning MORE effective?


or, some special blend that will not blow up or poison me?

I know acetone works on some stuff, I don't think PCBs though
ANd Naphtha (lighter fluid is great for plastics..but only for plastics as I understand.

So many questions?
one more:  any ultrasonic  homebrew solutions for electronics ?

[chickenHeadKnob]

I did a little video on common alternatives

TLDR version:  Use real flux cleaner (or isopropyl if you don't have it)

You really should have included brake cleaner in your test. Cleans at least as well as the MGC product with zero board or component damage, and MUCH cheaper. I swear by it now.

Well I just gave this stuff: (Certified Brake Cleaner non-chlorinated) a try - https://www.canadiantire.ca/en/pdp/certified-non-chlorinated-brake-cleaner-390-g-0381430p.html
Certified appears to be a Canadian Tire house brand, it is just Heptane and Carbon Dioxide. First impression is that it works very well!, not egregiously smelly or toxic and cheap and locally abundant. Winner!

My test was not a full soldered board. Just recently I bought some Rose's metal from Russia. This is low temp. Bismuth/tin/lead that melts around 97 degrees Celsius, and it comes  from the seller in drops (small blobs). I wanted it in the form of wire so I can  hold it while using it to solder the lids on hermetic boxes I am making so I coated the drops with MG-835 liquid rosinRA flux and heated them on a hot plated griddle. I use a black and decker brand teflon surface 24 inch breakfast griddle as my hot plate. I got the drops warm enough to melt them and get them to coalesce into 10mm wide ribbons, which should be as good as wire. The flux never got to smoke temp. so this is not the same as hardened burned on flux typically seen on PCB's.

I let the hot plate cool down, peeled off the rose's metal and let the flux harden for a day. I scraped off as much as I could before hitting it with the brake cleaner. There was still a layer of hard powdery flux stuck on the teflon that came off like magic with the brake cleaner but admittedly this is not as a good test as a real PCB.

[rhb]

One of these might help:

https://www.seattlefindings.com/Draw-Plate-Round-10-32-Gauge_p_4067.html

The normal routine is to make a narrow mold, cast an ingot and then successively draw until you get the desired diameter.  That's what jewelers have done for thousands of years and still do regularly all around the world.

A piece of 10 AWG copper wire pulled straight and pressed into plaster of paris that is almost set should work for the mold.  You can probably melt the drops in the mold with a hot air gun.

[chickenHeadKnob]

The normal routine is to make a narrow mold, cast an ingot and then successively draw until you get the desired diameter.  That's what jewelers have done for thousands of years and still do regularly all around the world.

A piece of 10 AWG copper wire pulled straight and pressed into plaster of paris that is almost set should work for the mold.  You can probably melt the drops in the mold with a hot air gun.

Thanks, a few similar approaches occurred to me. Rose's metal is new to me as it is the first time handling it. It is surprisingly hard and non-ductile, I don't think it will draw well. Plaster of paris is a good idea. As it turned out the teflon prevented the molten metal from alloying with the underlying aluminum and I could steer the liquid around with a popsicle stick to get ribbons so my hasty approach gave quick functional results.

[rhb]

If it work hardens while drawing it will be really difficult to anneal without melting it.  It may be hard from being quenched in water to form the drops.

Jewelers commonly use solder in sheet form.  So another option is to cast a wire ingot and then either draw if out with a hammer and anvil or compressing it in a vise to make a thin ribbon that you can cut with nippers.

The plaster of paris mold should be baked in an oven at 250 F for several hours before use to make sure it is dry.

[T3sl4co1l]

A quick look at the literature suggests elongation (a measure of ductility) should be very good indeed.  You may have to go slowly, however (which implies the large elongation is due to creep).

Tim

[chickenHeadKnob]

A quick look at the literature suggests elongation (a measure of ductility) should be very good indeed.  You may have to go slowly, however (which implies the large elongation is due to creep).

Tim

I have to play some more. i don't know the purity of the stuff I have, I always have some doubt when buying from ebay. One of my goals was to use the griddle and a thermocouple to ascertain the melting point to get a gauge on the quality. Well I screwed that up, somehow the thermocouple via my Brymen 869 was giving wild readings, plausible one moment implausible the next. I cranked up the heat after getting impatient and over shot 110+ degrees, then things started melting too quickly.
 
 I had a small container of water and a sponge and was using the sponge to drip water onto the griddle as a sanity temp test and the water drops were boiling off quickly when the metal started to melt. The water and sponge were also intended as a quench once I had the metal in the form I wanted. As the metal cooled a network of crystals could be seen forming, indicating to me that one component was coming out of alloying solution first and that the final result may not be all that homogenous. Once solid I tried bending it. It feels like something inbetween pot metal/white metal and standard 63/37 solder.

I feel embarrassed derailing this thread from the flux removal topic

[grantb5]

I feel embarrassed derailing this thread from the flux removal topic

Start a new one or bring it back on topic when you sort your current issues.

[T3sl4co1l]

Eutectic shouldn't be forming visible crystals.  Maybe you have rather more pure bismuth than you were expecting?  It should still be on the soft side if the melting point is obviously not that high, though, AFAIK...

If you leave a strip of the stuff cantilevered out, does it bend slowly over time?

Tim

[Ian.M]

The traditional Rose's alloy is non-eutectic.   See http://www.che.uc.edu/jensen/w.%20b.%20jensen/reprints/178.%20Fusible%20Alloys.pdf

You should be able to melt it under water in a glass beaker, with anti-bumping granules added so you cant superheat it by more than a few deg C.   If not, it not what its supposed to be.

I'm curious how effective the rosin RA flux was at such a low temerature.

[donperryjm]

For flux removal i tried IPA, but didn't hae 99%
So i got my GF's Acetone and added 1 part WD-40 to 10 parts acetone.

The acetone breaks up the stuff and the WD-40 makes it impossible to continue sticking to the board. The acetone will evaporate and leave behind WD-40 so clean up the rest with pure acetone. You won't believe the removal quality.

[helius]

That's the second instance of this thread being revived by necromancy...

Acetone will work well if you can stand the smell. (It's sold as "Heavy Duty Flux Remover" by several companies.)
What you do have to watch for is that it is absorbed by most plastics including the epoxy in IC packages, which quickly erases their markings and could cause delayed failure problems. Stick to the side of the board with no components and you should be okay.

[Electro Fan]

Ok, resurrecting an old thread; could have started a new one but there was a bunch of good discussion here; I learned a lot especially from this post by free_electron (Thanks):

https://www.eevblog.com/forum/suggestions/myth-busting-solder-flux-cleaning-with-isopropanol-alcohol/msg131044/#msg131044

So the question remains, what's the best way to clean a flux off the board - in a few respects....what to use?

(Background:  recently got a microscope and not only have I moved further from through hole into SMD but pretty much everything I thought I knew has been re-arranged now that I can see what is happening with much clearer detail).

I've tried IPA but found that it needed some rubbing and scrubbing.  Two things happened: 

1) using q-tips (maybe the biggest offender) and various pad and cloth materials the result has been that there are plenty of fibers getting stuck on components and in residue on the board.  If anyone has any better cleaning device beyond a toothbrush and better drying tools/techniques beyond compressed air, please say. 

2) after trying IPA and eventually getting the boards sort of but not really clean and dry (but full of stuck fibers) I noticed that the boards seem dull - kind of like the IPA had eaten some of the board finish.  So I experimented with diluting the IPA with some (pure?) water (90 IPA / 10 water, 80/20, 70/30, 60/40) and found that the board dulling problem seemed to improve with the water mix but it's hard to say where the best % is, or if this is even a good solution. 

It seems like some version of MG Chemicals might make sense:

https://www.amazon.com/MG-Chemicals-Remover-Boards-Aerosol/dp/B005T8L4V2/ref=sr_1_2?dchild=1&keywords=mg+chemicals+flux+remover&qid=1605216055&sr=8-2

https://www.amazon.com/Remover-Boards-Improved-Formulation-Liquid/dp/B07XDCRLR6/ref=sr_1_4?dchild=1&keywords=mg+chemicals+flux+remover&qid=1605216949&sr=8-4

I haven't used the aerosol version but it would seem to be difficult to manage in terms of overspray vs. the liquid - but that's just a guess.

Long story short, what are the best chemicals, dispensing method, and cleaning and drying tools to best clean flux off boards without dulling boards, harming circuits, or harming users with non-safe substances.

If we have to pick 2 of 3 nicely clean and safe (for parts and people) are more important than price.

Thx

[thm_w]

Distilled water is fine, but you'll have to dry off the board after washing if its a significant percentage. Use compressed air (filtered) to get the bulk of it off then place in a drying oven (~60-80C).
If you don't want loose fibers you use an ESD safe brush.

70% isopropyl 30% water is available for cleaning: https://www.mgchemicals.com/products/electronic-cleaning-products/isopropyl-alcohol/70-isopropyl-alcohol/
But I don't know if cheap off the shelf 70% is clean enough for your use.

[grantb5]

Same issue here exactly. It seems you really need to go to town with a brush and IPA to get the residue off and there is that dullness or even a white residue left behind. I find it necessary to trim those brushes as they aren't stiff enough to do much good in the original state. Toothbrushes actually seem harsher (poor teeth and gums). In my case I'm using Kester 245 and only (attempt) to clean boards that people complain about the appearance. It's a limited improvement.

Here are Kester's words, but I have not followed this up...

The 245 flux residues are non-corrosive, non-conductive and do not require removal in most applications. IPA will not clean the residues off the surface of the circuit board after the soldering process. A saponifier or cleaning agent specifically designed to clean a no-clean flux is required to clean the residues.

So for my solder, I need to locate a saponifier that will do the job, whatever that is.

[helius]

1) using q-tips (maybe the biggest offender) and various pad and cloth materials the result has been that there are plenty of fibers getting stuck on components and in residue on the board.  If anyone has any better cleaning device beyond a toothbrush and better drying tools/techniques beyond compressed air, please say. 

I noticed that the boards seem dull - kind of like the IPA had eaten some of the board finish.

IPA does not "eat" any part of the circuit board. What is happening is that you are redepositing the salt content from the flux and it appears dull grayish white. These salts increase the activity of the flux, but they are conductive and corrosive when exposed to metal. This is an indicator that your cleaning is inadequate, and you are simply pushing around contaminants instead of flushing them away.

Nanofrog's reply (and hundreds of posts all over this board, frankly) notes the correct way to remove flux; also see IanB's reply. Q-tips can only absorb a small volume of material, so they must be changed very frequently. If you have a large amount of flux or a large area to clean, you should either use towels (Kimwipes) to absorb the contaminants, or flush them off with spray into a waste receptacle. Brushes do not absorb contaminants at all, so simply brushing never removes anything. The brushing is for loosening the grip of the soil to the board, and that loose soil must be flushed away before it dries again.

So for my solder, I need to locate a saponifier that will do the job, whatever that is.

Saponifiers are used in aqueous cleaning systems in PCB assembly houses. They react with flux residue to form a kind of soap that washes away in the water tank and is removed.

You're using the wrong solder for small-scale operations. Aqueous cleaning is not conveniently done on a small scale.

[grantb5]

Saponifiers are used in aqueous cleaning systems in PCB assembly houses. They react with flux residue to form a kind of soap that washes away in the water tank and is removed.

You're using the wrong solder for small-scale operations. Aqueous cleaning is not conveniently done on a small scale.

Thanks. Any recommendations? Most of my stuff I am happy to use No Clean Flux and leave it on, but some boards I would like to clean.

[helius]

From Kester, 232 and 285 RMA are good for assembly. For rework, sometimes a more active flux like 44 is useful. These all clean up pretty well with IPA, although MG 4050A is slightly more effective. You might try it on your 245 flux, too. If the commercial cleaners are too expensive, you can also add a small amount of heptane (Coleman fuel) and acetone (like 5% each?) to IPA and it will clean much faster.

There are lots of companies making similar products: for solder, Multicore, Ersin, Alpha, MG, Chemtronics, etc...
For cleaners try Techspray, Electrolube, Chemtronics, ACL, Microcare, etc...

[Electro Fan]

1) using q-tips (maybe the biggest offender) and various pad and cloth materials the result has been that there are plenty of fibers getting stuck on components and in residue on the board.  If anyone has any better cleaning device beyond a toothbrush and better drying tools/techniques beyond compressed air, please say. 

I noticed that the boards seem dull - kind of like the IPA had eaten some of the board finish.

IPA does not "eat" any part of the circuit board. What is happening is that you are redepositing the salt content from the flux and it appears dull grayish white. These salts increase the activity of the flux, but they are conductive and corrosive when exposed to metal. This is an indicator that your cleaning is inadequate, and you are simply pushing around contaminants instead of flushing them away.

Nanofrog's reply (and hundreds of posts all over this board, frankly) notes the correct way to remove flux; also see IanB's reply. Q-tips can only absorb a small volume of material, so they must be changed very frequently. If you have a large amount of flux or a large area to clean, you should either use towels (Kimwipes) to absorb the contaminants, or flush them off with spray into a waste receptacle. Brushes do not absorb contaminants at all, so simply brushing never removes anything. The brushing is for loosening the grip of the soil to the board, and that loose soil must be flushed away before it dries again.

So for my solder, I need to locate a saponifier that will do the job, whatever that is.

Saponifiers are used in aqueous cleaning systems in PCB assembly houses. They react with flux residue to form a kind of soap that washes away in the water tank and is removed.

You're using the wrong solder for small-scale operations. Aqueous cleaning is not conveniently done on a small scale.

Thanks.

So this is what you like?

https://www.chemtronics.com/flux-off-no-clean-plus

Any reason to consider this (or some other MG Chemicals) product?

https://www.amazon.com/MG-Chemicals-Heavy-Remover-Aerosol/dp/B005DNQWV0/ref=sr_1_3?dchild=1&keywords=chemtronics.com+flux-off-no-clean-plus&qid=1605331814&sr=8-3

Just to clarify, you not recommend saponifiers, correct?

Thx

[KaneTW]

I just clean boards in an ultrasonic cleaner with a cleaning agent, then rinse in DI water and let them dry (oven at 40-60C with some molecular sieves inside, but you can also use IPA to displace the water if you're in a rush)