Hi all, I have some computer keyboards which got submerged in water due to a flood. After 2 weeks I opened the keyboards to find that parts of the pcb components have corroded and effected the circuits. I have attached some pictures as reference to the corroded parts. I have some questions regarding these:
1. How long does it takes for a component to corrode to that level? 3-5days?
2. Would drying it after it got submerged help prevent the damage and corrosion?

 

You might be able to get some of that crud cleaned off by using an old toothbrush, then rinse with isopropyl alcohol (91% if you can get it).

 

what kind of water was it submerged in. It looks like it was in salt for there to be that much rust? Or perhaps it was outside for a while. To truly restore stuff like that you would need to wash it in an alcohol bath and then access each circuit. To asses what component is bad if any. You will have bad components. For the cost a keyboard is it worth the time. I mean sometimes if it is a side project to learn and your just gonna zone out to tunes or something while you tinkering at it with your multi meter sure. Is it like your magic keyboard makes you program like the GOAT. I guess i get it.

As far as time goes that could be three minutes or three hours or three days of submersion. What happens is it is a tight space so the water evaporates very slowly. Now add the humidity after a flood and its just sitting there clinging to the board.

The only way to accurately know is is collect all three variables. Chemistry of the water, component metallurgy (for each component and the solder) and air humidity. Then you might be able to figure out a time of submersion or exposure but again why?

I guess your question leads to a question? Why is that important. Is this about insurance or busting some one in some columbo style aha moment. Unless it is murder I would say let it go.

The proper way to handle any dunking for any electronics is exactly what metermannd said wash with the highest % alcohol you can get and dry rapidly.

 

I called some engineers from the keyboard and field engineering groups to a meeting at one company at which I worked. I wanted to know what the main causes of failures in keyboards were.

They all had the same answer -liquids spilled on the keyboard. The drink that gets knocked over, the built-up accumulation of water droplets on an ice-cold mug... it doesn't take much water more than an instant to destroy a keyboard. It often works its way into the laminated buttons and they never work again.

If you have high hopes I recommend that you go ahead and try to save some. At the very least you will learn something, but even after fixing all of the visible damage, there is likely to be contamination inside the laminated buttons.

Good luck with your project.

 

The green caste to some of the white salts on the board indicates copper. This means some of the traces may be gone underneath. Cleaning could restore operation, if the PCB is still intact. I have rehabilitated PCB that looked much like that but in some cases had to repair traces.

It may seem counterintuitive but starting out with a soak in distilled water would probably make things easier. The crud on the board will be largely water soluble. When usage a brush, be gentle, the traces may be intact but vulnerable. I've found that standard pattern M16 rifle cleaning brushes are excellent for this sort of thing. The nylon type, not the brass or steel. They are cheap and easy to get.

​

 

If a piece of electronics gets dunked the first thing to do as soon as possible is to drain it, and if it was not clean water, then rinse it with clean water right away, and then drain it and dry it right away! Corrosion starts as soon as air hits wet parts, and any dirt will hold the water for a longer time. So acting fast is important. Also removing all power right away is vital. After that, drying as fast as is reasonable, before applying any power. Drying in a hot car is a safe option.

 

It's interesting that people recommend different washes but the book end are the same. Get the power off ASAP and and dry ASAP.

I have never considered it before but what about vinegar? Like white distilled? From my knowledge about the dive industry simple green might be an interesting option as well. You forever deal with green crap like that an simple green was a choice of allot of techs.

Fore renewal purposes hyper sonic scrubbers are another excellent road.

 

Couple different jobs. One was working with trade system computers. The really big stuff that occupied entire rooms. We'd get returns from the field, monitors, switchers, power racks - the list goes on. But the most common thing we got back was keyboards. I don't recall the manufacturer of these KB's but they'd come back with "Spilled coffee", "Spilled tea", "Spilled water", "Glass from a broken light bulb fell inside", "Spilled soda" - and that was the interesting one. Pepsi, Dr. Pepper, Root-beer, 7uP, all meant you opened it up and washed it out, then put it back together. 95 times out of a hundred they still worked. But when you spilled "Coca-Cola" in it - you just tossed it in the trash.

The first time I came across one with CC in it I started to open it. My work partner said "Just toss it. You're not going to fix it." Undaunted at the idea of fixing it I went ahead and opened it. Found components with no leads. Found traces that just went so far and then just disappeared. The CC ate the copper away. Likely the KB set wet with CC in it over night before the stuff dried out. And yes, it was truly a 'toss in the trash' affair.

Another job I had was working as a Micro-Section technician. I'd receive every shipment of bare PCB's and coupons. The coupons would be cut and mounted in epoxy then ground into the through-holes. Then etched with Amonium Hydroxide, Hydrogen Peroxide and water. The mix was 1 to 1 to 1. It would eat the very surface of the copper revealing its grain structure along with the interfaces of inner layers of the PCB.

A period of time went by where engineers were arguing about the hydroscopic nature of PCB's and the need for baking before wave soldering. I understood the argument but didn't believe baking was all that necessary. So I contacted one of our PCB suppliers and asked them to send me a bunch of failed PCB's for hydrostatic testing. I took 27 boards of three different types. Separated them into their types leaving 9, 9 and 9. Of each set of 9 I took 3 and dunked them into a tank of clean water and left them there for three days. 3 more were put into a baking oven for the same three days and the final 3 were left as packaged. After the three days I ran the 27 boards up the wave solder machine with no components, just bare PCB's. The examined results were as follows: Of the three different types of PCB's the ones that soldered the best were the ones that were left untouched, straight out of the shipping bags. The second best soldered boards were the ones that sat soaked in water for three days. The difference between the first two were almost undetectable. There were a very very few tiny blow holes. The ones that soldered the worst were the ones that soaked in the oven for three days. They were hard to solder and there was a lot of unsoldered pads and holes.

The conclusion was that PCB's are not hydroscopic. Others here may feel different on the subject, and I respect their opinions. I'm sharing MY results with you. The reason I bring this up is because sitting in water alone shouldn't cause much issue. Especially if the board was not powered at the time of submersion. Remember the first job experience mentioned; we'd clean them and put them back into service. YOUR KeyBoard has far more than just water on it. Likely something was introduced that is conductive and causing the KB to be non-functional. Rinsing it off and brushing it off with a toothbrush might remove all that gunk I believe I see on the board. What concerns me most is the switch. It likely has so much junk in it that it can't effectively switch one circuit on and another off. Even if it's merely the "ON" switch, it may either be stuck ON or enough contact to not let it shut off. Or possibly when switched ON the corrosion is so thick that the contacts don't make full contact.

One more experience, and yes, this is getting far longer than I like to see: The wife cleaned the microwave oven. Down inside the control board she washed cooking grease into it without knowing it. The MO didn't work after that. She was deeply upset thinking she ruined it. I took the board out and washed it off in the kitchen sink, using mild detergent to break up the grease and flush it away. I then took it outside and doused it with rubbing alcohol. The purpose of that was to rinse away the water. Then I blasted it with an air hose. That flushed away and dried out the alcohol. Put the board back into service and it's still working some seven years later.

The moral of all these stories is that 1) you have nothing to lose. 2) you might succeed. If you don't succeed you still have made the attempt. So give it a go. However, you may have to disassemble the switch to clean it. As for the corrosion - that looks more like something deposited than corrosion of the copper. If water could corrode copper then we wouldn't have copper pipes. And yes, water CAN corrode copper, but it has to be super ultra pure. So pure that it leaches the minerals out of the copper. I've even seen manufacturers who use ultra pure water to etch metals. But I don't think you encountered super ultra pure water. Worst case - it was salt water. Or sewage. In that case - ew!

 

Coke has phosphoric acid in it. It will eat the components without electrolysis to help. The only hope is to get to it very soon after the spill. The green color is clearly Cu in the salts. If power was applied for any time after the spill electrolysis will eat the traces and component leads.

 

I called some engineers from the keyboard and field engineering groups to a meeting at one company at which I worked. I wanted to know what the main causes of failures in keyboards were.

you would think that someone would design a different type of keyboard that would look and feel like normal one but it is immune to this... should design one that can function in sea water. clumsy people around the world would celebrate.

 

you would think that someone would design a different type of keyboard that would look and feel like normal one but it is immune to this... should design one that can function in sea water. clumsy people around the world would celebrate.

They do, a lot of them. Here’s one example.

 

you would think that someone would design a different type of keyboard that would look and feel like normal one

The industry has been trying for decades. The problem is that nothing feels as "springy" as an actual coil spring, and a completely sealed, completely waterproof springy switch is expensive.

ak

 

They do, a lot of them. Here’s one example.

Is it just me, or does the keyboard description have a typo in it?

ak

 

you would think that someone would design a different type of keyboard that would look and feel like normal one but it is immune to this... should design one that can function in sea water. clumsy people around the world would celebrate.

As the manager of Architecture and Technology for keyboards (and several other products) that is why I called the meeting. Then I started taking keyboards apart and realized that all of the relatively easy fixes had already been incorporated. Anything beyond what was already being done would have been too expensive for our markets. Having had to replace a few keyboard since retiring I see that for consumer products and most business lines, they have not progressed an inch in the last 20+ years, probably for the same reason I couldn't fix the problem for our customers back in the 1990's.

Now that decent keyboards (I like Logitech) are around US$12 each (where I live) the issue doesn't seem so urgent.

 

A keyboard would have to be very special to be worth the effort to repair. There's usually a stack of keyboards at local thrift stores (or computer recyclers) for a few bucks each. I replaced my worn-out K400 Plus (wireless with touch pad) for about $3.50 (same model) (1/2 price electronics day). Used wireless keyboards (and mice) often lack the necessary USB receiver, but Logitech's "Unifying Software" would have let me marry the new keyboard to my old USB receiver dongle (up to 6 devices per receiver). (Microsoft's wireless keyboards and mice don't have that ability, so if the dongle is missing, they're doorstops or hockey pucks. Microsoft calls that a security feature.)

 

some keyboards are not even PCB's anymore I recently went into mine to do a clean up of a sticky liquid and it is actually two layers of transparency with conductive ink and a thin rubbery gasket on each key.

 

About 30 years ago I was on the road teaching two-day seminars on networking. I carried a demo system comprising two notebook computers, one as server and one as workstation/presentation tool.

Just before lunch, I spilled water into my workstation keyboard. Suddenly half the keys didn’t work. I had another day of teaching to do, and another city to be in that week for another two days. Without that computer I was lost.

Of course I pulled the battery immediately. At lunch break I went to my hotel room and disassembled the computer. The keyboard was modular, which helped, but there are a plastic sheet over the PCB that had retained a lot of water, and between the two PCBs that made up the keyboard there was water as well.

The boards were attached to each other with 27(!) 00 Phillips screws about 3mm long. Fortunately, I always carried tools and so I had a screwdriver that fit, but the prospect of removing all those fiddly screws and the resulting who-knows-what pile of parts was daunting to say the least.

I removed all the screws carefully setting them aside (I was sure I would lose at least one) and then carefully and with great trepidation removed the lower PCB to reveal a lot of switches each with a spring that wasn’t otherwise secured and a swimming pool (or so it looked to me).

With the clock on lunch running out, I grabbed the hair dryer from the bathroom and proceeded to very carefully dry the keyboard with it. The little sprigs dangerously waving in the air currents threatening to disappear and become invisible in the carpet. After a few minutes it was apparently dry and I was nearly out of time.

I put the lower PCB back in place and replaced all 27 screws. Amazingly none were lost, none cross threaded, and none had their heads stripped. It all went back together. I powered the computer on, it booted and the keyboard worked! I wasn’t sure if it was a permanent repair, but it was working and lunch was over.

I made it back to the room in time and while I didn’t get to have my lunch I had a working notebook computer. I traveled to about 40 cities with that computer after the incident and never had a problem. The water (which at that time came in pitchers with ice To pour into glasses) would never, ever be on the same table as my computers again.

[EDIT: Changed the springs from “inviolable” to ”invisible”, not sure just how a spring could become inviolable in carpet…]

 

not sure just how a spring could become inviolable in carpet

You've never seen a spring violated before? Here in the house of Tony, there've been lots of springs that have been - um - violated - when fallen into the carpet.