Solvent for cleaning PCBs in a CNC machine in situ....

Status
Not open for further replies.

fastline

Member
We have a machine that was bought many months back. It ran but they were having some issues with the spindle VFD shutting down. DUH!!!! The machine was obviously close to cast iron dust and these guys did NOT take care of it the air filtration and there is likely conductive dust all over the interior PCBs.

We discussed removing boards, cleaning, etc, etc, but it will take a serious amount of time and most certainly some risk in that. What we discussed is simply using an automotive spray gun loaded with a "recommended solvent" that will be safe for everything. I suspect safe will not make things perfectly clean again but we also don't want to go rinsing off solder masks or otherwise damaging boards. I just don't think air by itself will be effective and manually cleaning just won't work.

I was thinking isopropyl but that stuff seems to rarely work around here for cleaning boards. I know there will be oil deposit, not just dust. I am hoping to feather between solvent spray to just straight air, then hoping the solvent will help break stuff down and we can finish with fine air to get the crud out.

the other mode of attack is to use a large and soft paint brush, dip in alcohol or other solvent, wash down the PCB, then finish with air. In all cases, there are just places that are almost impossible to access so either mass brushing or air will have to be it. I do have concerns getting stuff up under chips but I think we decided the potential pros out weight the cons as the machine already has reliability issues due to contamination and more shorting may zap the machine anyway.

Speaking of zap, I might ask for ways to minimize spark concerns. I know compressed air release can and does build a charge. I need to make certain we cover this base. My best friend has been a pro auto painter for decades and never had an issue and they are shooting stuff 10x more volatile than alcohol but...... they are shooting stuff wet and I want a fine mist of solvent with mostly air.
 
Last edited:
When you say oil deposits, do you mean from the coolant used in the CNC machine?

There is MG Chemicals Superwash and that stuff deals with oil (at least the kind of oil used to prevent shop equipment from rusting) and is PCB plastic safe. It's expensive though you probably still want to use cheaper isopropyl first and then use smaller amounts of Superwash later to get what is still behind. I don't know the scale you're thinking about doing this on though since Superwash only comes in a little 15oz aerosol can.

THere is also MG Chemicals Safetywash which comes in larger quantities in liquid form that you could put into your spray guns but I've never used it, but it's probably more effective on oil than alcohol. It has ethanol in it, which from my experience with the MG Flux Cleaner, is the magic ingredient that separates it from just plain isopropyl (as far as flux is concerned anyways).

If you do remove the boards, then maybe consider ultrasonic cleaning? Those requires special cleaners though since the ultrasonic action can ignite regular cleaners. I am also unsure if the ultrasonic action will do more harm than good with metallic dust embedded into everything.

About the zapping, wouldn't the conductivity of alcohol alleviate the spark concern from static buildup of compressed air? I'm just floating the concept. I have no idea.
 
Last edited:
Well, I ran a quick test and it was promising enough to go use some other solvents on a test board. My immediate concern is charged air particles damaging components. I was really shocked how little product I used. Maybe 1oz of 70% iso. I need something more aggressive though.

I really need to ensure I don't create static, whether that is a spark concern or component damage concern. I guess if the machine and gun are grounded, maybe that will help. My only experience with charged air is in a sand blast cabinet and let me tell ya, it will straight up zap ya! I need to install a band to ground myself.

I also have vodka in the house so I am going to go test with that now.....lol I bet something will work in the right proportions.
 
If you look on the MDSD datasheets o the MG CHemicals website for Superwash and PCB Safe FLux Cleaner, it will tell you the mixture used.
 
Getting closer. Tested with Lacquer thinner. That is a blend of acetone, methanol, and some keytones. I think I am pretty close with that one. Certainly better than isopropyl but I need to check how it reacts with wire insulation. I can control where I spray pretty well. I have my gun dialed way down to about 1/2" diam but if it can melt even one thing, I probably should not use it. I really want to turn it up though. I do notice that though lacquer thinner did not remove all residue, what was left was very easily removed with my finger, indicating it totally broke down the contaminate.
 
I don't recommend doing anything that involves evaporating or aerosolizing large amounts of solvent. Vapor cloud explosions and flash fires are far more expensive than remaking some pcbs.

Static is most dangerous with non-polar solvents but good grounding and bonding should always be used. I would simply track down a good ultrasonic cleaner and Dawn dish soap. It is amazing how well that decreases in an ultrasonic bath. Rinse them in tap water, then rinse with distilled water. Let air dry. Then put them in a 70°C oven over night.
 
After some thinking and review of the MG superwash, I realized mass air flow sensor cleaner is nearly exactly the same! MG is hinging on pentanes and butanes with some methanol likely to reduce evaporation. It is a sure bet that MAF spray will completely evaporate and leave no residue. So just pitching that to anyone that cares, that is something you can grab at any store.

For me, I found that Lacquer thinner may be better for my needs until I can find something better. It's increased evaporation time actually helps as I learned to spray once and let it sit for a while, then go again and follow with air. It seemed to work decent. I want to at least try acetone though. I know it can be aggressive with plastics but for a no touch situation, it might work.


Again, removing the boards is really not an option. hundreds of wires have to be marked and removed. Goal is to not do that.
 
Whenever I'm working on a board in situ I always take lots of pictures before hand so I can reconnect the wires that inevitably get disconnected.

Mike.
 
Try IPA - we use it for removing oily deposits and it is relatively inoffensive if it doesn't work.
 
Circuit boards covered in cast iron dust.
Blowing compressed air and assorted solvents around.
Don't want to remove the boards to do a proper job.

In my opinion this sounds like a really half baked idea to try and do a quick and cheap job.
Was this method suggested/insisted upon by "the management" ?

In my opinion this can only end in tears.

Do the job properly.
Take lots of pictures.
Make notes where necessary.
Remove the boards from the machine and clean them in a well ventilated area.
Use a vacuum cleaner and a soft paintbrush to remove accumulated dust and debris.
Use iso-propyl alcohol and/or de-ionised water with a mild detergent to wash the boards.
Dry them for a couple of hours in a warm place.

Re assemble the controller box and do something to improve its IP rating.

JimB
 
No biggy, plenty of good people here.

As to the comments of JimB, all I can say is this is a low value device and you may not quite understand the complexity in unwiring things. As well, there is certain risk in breaking something just taking it apart. Equipment is 20yrs old. You will just have to trust that we feel we are making the right decision and if PCB removable was practical, we most certainly would be on that train.

As to the soap, actually a soap would do precisely what I want, no doubt about that. the problem is knowing what that product is and how it reacts to various things, not to mention to use of water which could get inside something and stay there for quite some time. I am open to it though. We commonly use Alkaline cleaners such as royal purple, orange degreasers, etc I can pretty much guarantee a little royal purple would do the job but I have no way of knowing what it might attack if I cannot get it fully removed so submersion might be best with that stuff. That stuff gets right with it on grease and tears the hell out of your hands so just not sure.
 

Incidentally what make of control is it? Or is it a custom system?
Max.
 
We just use IPA.
We do remove boards and clean them individually - and strip & clean the chassis / backplane if that also needs doing.

Unless you can get straight on to the face of each board and spray it under pressure, I don't think you will have much success cleaning in situ. They often take quite a bit of scrubbing if the dust has mixed with oil or coolant vapour, it seems to gel and take rather more than just wetting.

You also have the risk of it running slightly and an even heavier deposit ending up around such as rows of IC pins, making things worse.


If the dust happens to be totally dry, a good blast over with an airline may do the trick - and it's a lot easier to get an angled air nozzle between boards than a brush.


Be extremely wary of acetone or thinners, that attacks many plastic and will likely remove printed component markings, even if it does not do any other harm.
 
Well........ we are done with the job for right now. It was determined that the majority of deposit on the boards that was of concern was actually just dust so nothing but compressed air was used on the boards, but the enclosure walls and around the fans had nasty oil deposit. Technically, that stuff would probably help catch dirt, but it did not look good and was all over the cables too. We used Citrus cleaner on all the cable markers, cables, walls, etc, and just air for the boards. I am not sure if we have all problems solved but noticed all plugs were oriented on the bottom side of the boards so any fluid cleaners would be running crap down into the plugs so the boards would at least need turned 90*. It was decided to run testing at this level of clean and monitor things and see where we are. The pictures we have would not do justice but it was 8hrs of nothing but wiping and blowing air. I feel we made a large improvement.

The VFD we were most concerned about has really good conformal coatings so that drove some of our decision to eliminate fluid cleaning for now.
 
Always be careful of cleaning PCBs- you don't want to remove the varnish off the PCB if it's their, you don't want to make traces come up, and you don't want to dissolve plastics or rubbers.
 
Status
Not open for further replies.
Cookies are required to use this site. You must accept them to continue using the site. Learn more…