I had a problem somewhat like this, a few decades ago. Our Macintosh computer sat in the "den" (third bedroom), the house had little insulation and a very inefficient heating system, and we let it the house temperature drop down to the lower 50s overnight.
If my wife took a shower in the morning, just down the hall from the den, the Macintosh would occasionally power itself on and boot without human intervention a few minutes later.
Warm moist air traveling down the hall, infiltrating the Mac ADB keyboard, condensing out as "dew" on the PC board, and creating a small current leak across the high-Z "power on" button terminals was what I eventually figured out was the culprit.
Cleaning the board with alcohol and then coating the on-key terminals with acrylic nail polish fixed it.
Electronics gear really shouldn't be operated under "condensing" conditions, due either to "adding humidity to the air" or "dropping the temperature". I've seen warnings about this in the ops manuals for some test gear... if you bring it in from a cold area, let it warm up by itself before powering it on (ideally inside a plastic wrapper to keep out humidity).
Except that in a lot of areas, turning the heat or AC completely OFF at night has near no effect as the temp changes little. A few degrees.
Note too that the recovery in the morning almost costs as much as was saved, but it is still worth it.
Very doubtful condensation becomes a problem in such instances as ALL of the air slowly changes temp, as does ALL of the gear as well.
If it is driven gear (heat source), it may not exhibit any change.
Folks like forced air because it is moving air. Turning off air handling and temp maintenance machinery at night saves the air handling cost, and not really much more.
in the "good old days" when you could rent a video player with the videos from the local shop, there was a warning to leave it inside for 20 minutes before using and it had a "dew light"
Good point. The ultrasonic variety are called nebulizers[1]. The white dust is mostly calcium carbonate.
So, how is this stuff harmful to electronics? I can see it acting as an abrasive for moving parts. It's also ionic in water and can cause the water to conduct. However, once you have condensed water on a PCB, the damage is already done, and the calcium carbonate only makes it conduct some more. AFAIK, if the white dust is dry, nothing is going to happen (unless it gets wet).
[1] In college, I built a piezoelectric nebulizer that was used mostly to vaporize alcoholic beverages and creative concoctions that were suppose to act as incense. The big advantage is that it will produce fine mists and vapors without getting them too hot to inhale. It's amazing how quickly one can get drunk just by inhaling alcohol vapor. Amazing. It even has it's own Wikipedia entry:
--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558
Some consumer devices still have a humidity sensor. Most Brother printers have this feature. I discovered this the hard way. I had a new Brother MFC-7360n printer in my car, which I parked outside in the rain. I was scheduled to deliver the printer first thing in the morning. When unpacking the cold printer, in a warm office, moisture condensed all over the printer. I powered up the printer anyway, hoping that the heat from the electronics and fuser might warm up the printer enough to evaporate the condensation. Instead, the printer came on with an error message, which I later decoded to mean that the humidity was too high. I left the printer turned on and returned about eight hours later, when most of the water had evaporated. It then operated normally. Also, high humidity makes the pages stick together on a laser printer, which then causes paper jams.
--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558
Agreed. However, PCB's aren't the cleanest things on the planet. In a past life, I designed marine radios, which use low impedance circuitry to reduce the effects of PCB leakage. They are also designed to have the PCB's mounted vertically, to drain off any accumulated water. Whenever some clever person decided to use a high impedance circuit, or leave unconnected CMOS inputs floating, board leakage would appear to ruin their day. My standard sanity test was to spray the board with a thin mist of distilled water, which is allegedly non-conductive. We also had a small test pattern etched onto both sides of the PCB. The test was mostly for leakage, but also for contamination. Of course, we found plenty. We did our own board assembly and soldering, which sometimes produced some rather contaminated boards. Unpopulated test boards, untouched by our wave solder machine, were just as bad, which suggests a PCB vendor problem. The eventual solutions were to move to lower impedance circuits, wider trace clearances, conformal coatings, overkill cleaning, better control over the soldering process, etc.
Boards from the field were worse. Since they were used in a marine atmosphere, they were encrusted in salt. Add distilled water and instant board leakage. One day, I was wondering just how much salt was in such a radio. So, I dumped it in a bucket of hot water (standard procedure for cleaning the radio), saved the runoff, evaporated the water, and was left with a con about 1/2 inch high of sea salt. Most of this salt was not on the PCB's, but it's still a fairly large amount for a small radio. Don't ask how it got inside. That's another horror story.
I'm not sure if condensed water really is as pure as you suggest. Methinks it's much the same as rain, which is nothing more than massive condensation, which can have various impurities, such as "acid rain".
--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558
Yes, and then the whole house is full of white dust. I always wondered if you could then put a dust filter on the outlet of the humidifier.
As far as energy efficiency, the water has to become evaporated no matter how you get the process started. Gas heat would be cheaper than electric, but this is not a huge energy sink.
Since the thing is totally automatic, no big tanks to fill twice a day in the bathtub, etc. it is worth it to me to have a little more energy consumption.
Yup, exactly! And, any board in consumer-grade stuff is usually not real clean new, and gets a lot dirtier after a few years of use. Get it wet, and it is a horrible soup. (But you already said that!)
As soon as you talk about draining the water off a board I think we are "out there".
I seem to recall an employer talking about some government spec that required a test that a unit would not retain water. The item he was overseeing the test for was some sort of mounting bracket that had no top or bottom. The guy doing the test had to pour a cup of water through the unit into the bucket below, lol.
CMOS is a whole 'nother thing. You only need a few MegOhms of leakage to pull an input one way or the other. I seem to recall one data sheet talking about the impedance of a CMOS input as being 100's of MegOhms. I think that is in the range of an electrometer input impedance.
I can't think why you would expect bare PCBs to have 100's of MegOhms impedance between traces. They aren't make in a clean room and such levels of cleanliness are not typically needed. Such levels of isolation typically use coatings to achieve.
Yeah, I've heard all sorts of similar stories. Sea environment is pretty tough, even if you aren't shipboard, just being within some distance of the sea gets a lot of salt in the are.
That all depends on the environment. Acid rain is formed mostly from the exhaust of industrial processes or natural processes like fumaroles. How much exposure to industrial process exhaust or fumaroles do your boards receive?
I can't say for sure how clean any PCBs are. So perhaps condensation on your boards would be very conductive. It depends on how clean they are I suppose. But it's not from the condensation itself. That would be very clean unless your air is grossly contaminated.
Our customer had a removable 14" disk pack transported by car from the other side of town and moved directly into the computer room with RH around 50 % and later inserted to a disk drive.
A spectacular head crash occurred, requiring the disk pack, R/W-heads and air filter to be replaced. With such low read head flying high hitting even a small droplet is like a human hitting a tsunami :-).
Later on, the convention was changed, requiring a cold disk pack to be kept at normal office space with RH between 10-20 % for several hours, before moving it into the RH 50 % computer room. With this convention, no further head crashes occurred.
On a sunny day (Sat, 07 Mar 2015 09:17:57 +0200) it happened snipped-for-privacy@downunder.com wrote in :
I remember in Rome, Italy, we were on a reporting trip from some convention, brought video equipment in a truck, left there overnight.. In the morning it was brought into the convention hall, and we tried to start it. Moisture alarm on the Umatic recorder, did not want to run. One guy found an electric heater, opened up the Umatic, blew at it with the electric heater for a minute, and it recorded just on time... 'What is that guy doing with that electric heater at the video desk over there????'
That's pretty difficult to do in the US, particularly in this situation. Unless the employer were doing something illegal (according to OSHA, etc.), workman's comp would take care of it. Maybe.
I've heard of them costing their owners millionS, though.
Not really that far fetched. Vertical mounting of PCB's is strictly a preventive measure. Given the right conditions, there will be water inside the enclosure, and it will tend to pool and puddle where it can do the most damage. Vertical PCB board mounting, drilling drain holes in horizontal motherboards, providing drainage paths for enclosures, and other measure are preventive measures.
I found some photos online of one such radio that I helped design. Intech M3600: All the PCB's are vertical, except the motherboard, which has drain holes. Note the wide trace clearances. No conformal coating on this version.
Yeah, that won't work. However, resistance to a pressurized water blast is a very real test as in IP6x specs: and MIL-STD-810G method 506.5 (rain) and 509.5 (salt fog). Oddly, it's not the lack of a proper enclosure seal that causes problems with water incursion. It's the drop in internal temperature, causing a partial vacuum, when the enclosure is hit with cold water. The partial vacuum sucks any pooled water into the enclosure.
(...)
I never suggested that the PCB's need to be that good. The solution is to tie the unused CMOS inputs to either ground or power through a
10K resistor, not to improve board leakage. As I vaguely recall, we were looking for surface conductivity greater than about 100Kohms per square. That was fairly easy with conformal coatings. Without coatings, I guess about 10Kohms per square was considered good.
As for clean PCB's, we gave up and did our own cleaning. We couldn't get clean PCB's from any vendor that we tried. They didn't care because they knew that the subsequent soldering would leave a similar amount of chemical residues. So, we did a really thorough water wash and solvent rinse after soldering and before coating.
We're currently installing a 3 meter C-band fiberglass dish on the coast, above a brackish lagoon: We setup the previous dish about a year ago, when it was in the last stages of collapse by corrosion. At the time, I installed a new LNB. I was stuck with installing a painted aluminum scalar ring, to copper struts, on a galvanized steel plate, using stainless steel screws. I know a galvanic nightmare when I create one. A year later, it didn't quite crumble into dust, but it was trying. Fortunately, the old dish was destroyed moving it to a new location, therefore, the new fiberglass dish.
I also used to own a nearby rental house. It was well within the fog belt. Everything made of metal in the house was rusted, including the nails holding the house together.
That's why I said that I wasn't sure. My guess(tm) is that the exhaust fumes from the nearby freeway would be fairly good at locally simulating the acid rain effect. Some litmus paper and a water spray should be tolerable test.
I wish I had photos of some of the radios and boards that were returned for (lifetime) warranty service. The working environment aboard small fishing boat is far from pristine. Radios smelled from fish, diesel exhaust, mold, rot, mouse droppings, dead insects, or combinations thereof. Mouse urine is the worst, since it corrodes copper traces.
A few years ago, I inherited a Kenwood TS-50 HF radio that had been installed in a remote mountain top blockhouse that was powered by a diesel generator. Somehow, the exhaust from the generator was pumped directly into the radio. It took about a year, but it finally accumulated sufficient soot inside to short something. Someone was nice enough to clean the outside before sending it to me, but the inside was almost solid black soot. Plenty of soap and water eventually partially raised the radio from the dead. Some day I'll finish the job. Let's just say that you don't need acid rain to cause problems with electronics, when soot will do equally well.
I can, with a test pattern: Ours weren't quite this complicated, and were far from accurate, but did give approximate sheet resistance values.
Yep and that's what I don't know or understand at this time. It's not the water that's doing the conducting. It's the ions from the contaminants that doing most of the conducting. You're probably correct, but I still suspect that condensed water might be slightly conductive. It's easy enough to test if I can find a Megger or equivalent.
--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558
Water's dissociation constant is 10**-14, so if it's buffered at a pH of 7, it has minimum conductivity. The presence of a very small amount of slightly-soluble material can act as a buffer to give very different values. Salt adds other ions, acids or bases free up the ions in the water, both ways you get some sheet conductivity.
Pure water (pH = 7) conductivity is 0.0550 microSiemens/cm Copper conducitvity is 360 megaSiemens/cm, if I haven't bungled the conversion...
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.