Humidity

I guess summer in the mid-Atlantic region is also bad for computers... Not to mention the Gulf coast and many other places in the US.

--

Rick

It is typically only bad for those folks and IT idiots who for some reason fail to keep the PC innards clean. The humidity gums them up even further and faster and the failures start to mount.

The fix for the idiots is NOT to attack the symptoms instead of the cause. They ban humidity and PC laifespans increase. How quaint.

Keep 'em clean and they do not gum up with crud.

With humidity under 30% you'll start seeing ESD events and failures. Especially if none of you are smocked and/or strapped.

Dry is bad for ESD reasons. I don't know of any reason that wetter is worse.

40% to 60% is pretty standard when you have a lot of rackmount gear around. This is hardly esoteric, either.

--
Les Cargill

I would have guessed the other way around. When it gets really cold and the indoor humidity drops down near zero, you can generate amazing amounts of ESD.

** Low temperatures cause condensation.

If the PCs are turned off at night or on weekends, along with any air conditioning, then a film of water can easily form on the external and internal surfaces of the machines.

The higher the humidity, the more water condenses.

I see lots of electronics badly affected by the phenomenon.

... Phil

Hopefully, it's not -20C indoors.

The problem is the dew point. If the inside temperature is fairly normal (i.e. 23C), and you artificially increase the indoor humidity, when the office is closed at night, and they let the temperature drop, moisture is going to condense on everything. High humidity can also promote mold growth, corrosion, electrolysis, metal migration, and component value changes in film caps. Most commodity electronics is not designed to tolerate either high or low humidity. As others have mentioned, low humidity can product some impressive electrostatic voltages that are quite good at blowing things up when touched.

Big servers and mainframes also have limited humidity specifications. For example: Even turned off, it doesn't want to see anything less than 8% RH due to EST, or greater than 80% RH due to condensation. Most data centers try to keep the RH between 45% and 65%.

In my past life as a marine radio designer, dealing with the effects of moisture was a major problem. It wasn't just clean water condensed out of the air. It was condensed salt water, salt spray, immersion, and tropical high humidity. Besides the usual water damage problem, high humidity produced an environment similar to a terrarium, complete with mold and mildew growth. You wouldn't believe the botanical nightmare that arrived for warranty service.

Presumably, your humidifier isn't going to turn the office into a tropical swamp. That's a bit extreme and unlikely. However, as Decadent Linux users #1 mentioned, it's not so much the water that causes problem, but the mixture of water plus dirt, dust, filth, spores, goo, and crud. Pure water is not very conductive. Dirty mud, is quite conductive.

The problem is that most electronics is not designed for harsh environmental conditions. Certainly not a desktop computah, that is expected to spend it's life in a controlled environment. Portable devices are a bit more tolerant, but not spectacularly so. If the computah has passed MIL-STD-507.5 tests for humidity tolerance, it might tolerate your humidifier. However, few do.

Incidentally, speaking of condensation and portable devices, I once had to deal with an elastometric keyboard problem. The device was installed in skool buses, that were normally parked outdoors. The keypad was part of an alarm/paging/location system. When a button was pressed, a transmitter in the bus would transmit touch tones and data as long as the button remained pressed. At night, the buses would all cool down to ambient, which was fairly close to freezing. In the morning, the drivers would prepare for their daily route, start the diesels, turn on the heater, and about 150 transmitters would be stuck on the air at the same time. It was a truly impressive cacophony. What was happening was that moisture was condensing inside the elastometric keyboard, and keying the transmitter. The cheapest and fastest solution was new keyboards, with a larger contact gap, and better tolerance to humidity.

If you really must have a humidifier in the office, I think you might be able to justify it on the basis of controlling the upper limit on the humidity. Search for the acceptable humidity range for your unspecified maker and model computer. Get a few cheap RH meters, and promise to keep the RH between 45% and 65% (suitable for a data center) and I think you'll be ok.

--
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

This is not my spitzen kompentanze, but if you heat up the air, then the relative humidity is lower, but the absolute is still the same (the same amount of water per liter air)

I have my lab in my garage, at low temperature, with a dehumidifier set to 50% to protect my beloved instruments

I am thinking about adding more intelligent system, one that calculates the absolute humidity of both the inside and the outside, and when the outside is lower, pumping air in

Cheers

Klaus

%<

It sounds like you've seen 50 Shades of Grey.........

I have a humidifier running here.

With a new wick and reverse osmosis filtered water it never goes above 30 while the outside temperature is below -10C. I have a central forced air heating system. The temperature in the house is 20C.

The office in question has electric heating but no other form of air conditioning.

I doubt that use of a humidifier like the one above would get the humidity over 40, except with the outside temperature above 0C but then it wouldn't be used anyway. It could be turned off at night.

I think that the best way forward would be to measure the humidity in the office and see how it varies over a 24 hour period at specific outside temperatures. A sensible decision could then be taken about whether a humidifier would be beneficial to the people but not detrimental to the computers while the outside temperature is below a certain point.

Actually I think no one has yet touched on the real problem, and that is th e white dust created by the typical humidifers using regular water. This m ineral white dust CAN be harmful to electroncis.

I suggest you remind the powers that be that low humidity allows ESD and is harmful for humans and the well being of humans should be at least as impo rtant as the computers. Work with them to find a type of humidifer that doe s not create the harmful white dust.

This would use distiled water in an ordinary humidifer or use regular water in a "hot mist" humidifer which actually boils the water and does not crea te white dust. The minerals are left inside the humdifier which needs to b e cleaned out every month or so.

Mark

This mineral white dust CAN be harmful to electronics.

least as important as the computers. Work with them to find a type of humidifier that does not create the harmful white dust.

does not create white dust. The minerals are left inside the humidifier which needs to be cleaned out every month or so.

Years ago I repaired VCRs, I had a few customer machines that had white dust following certain traces on circuit boards. It was from ultrasonic humidifiers. Mikek

--
This email has been checked for viruses by Avast antivirus software. 
http://www.avast.com

Well, they are imbeciles! VERY low humidity is REALLY bad for computers, due to ESD risk. Back in the old days of floppies and mag tape, it was also bad for them, but that technology isn't used much, now.

Humidity up to 50% is very good for computer gear, and it is unlikely that a small huimidifier could get above that in Canada in the winter.

You can look up the storage conditions for typical magnetic media if you need a reference. I doubt there will be quite so much info on the computer gear itself, but I'll be all Dell owner's guides, for instance, would have storage and operating conditions listed.

Jon

OK, as others have mentioned, if they turn the heat back SEVERELY during off-work hours, then there is indeed a possibility of condensation or just sharp rise in relative humidity as the place cools.

But, I doubt that, as letting the place even cool to 50F (10C or so) at night would not likely raise RH all that much. And, letting it cool more than that might be bad for a lot of things in the office, not just the computers.

Monitoring would be a great way to find out what is going on. But, without any humidifier at all in the place, the RH is likely to be astronomically low, possibly below 10%.

Jon

Yes, I have fought this problem. The only things that REALLY make the white dist are ultrasonic humidifiers. Wet pad units rarely do this, but may grow mold and bacteria, and that is another whole problem.

I used to have the typical forced air furnace, and a rolling foam belt type humidifier did a great job there.

Our new place (well, most recent is more accurate, been there 25 years) has hydronic (hot water baseboard) heat, so you can't use a furnace-attached humidifier. I have been through a range of commercial stnad-alone units, and didn't like any of them. So, I finally invented my own. it is a big aluminum pot, 6" in diameter, with a ring heatihg element clamped to the bottom. It is set up with an autotransformer to run at about 140W. There is a water level sensor, and it fills the pot to keep about 1/2" of water in it. (There is also a thermocoule safety control in case the sensor sticks and the pot boils dry.) This thing fills from the icemaker fill line in the kitchen, so other than occasionally sucking up the concentrated salts from the city water, it is maintenance-free. Works great, keeps the kitchen at around 40% RH.

Jon

Generally not as the dew point makes a difference.

The air in a room *where* such readings are taken is slightly different than the air actually in a cabinet too, UNLESS it is indeed being exchanged.

If you turn off air handling, the air in the cabinet at that time is likely a lot drier than that outside the cabinet. The dewpoint of that air means condensation will never happen.

And hell, if they are so cold that they even approach dew point, you may not need any air handling at all.

One would have to generate profiles for heat sinked machines with air handling on and with it off. That would be the only way to truly know what to expect from a machine.

I have a conduction cooled "cubox" ARM computer which I run idle and full tilt video process running, and observe wattage consumed and temperature rise.

I have it mounted on a peltier cooling surface and observe the benchmark differences and then with the cooling. It would matter on a PC, but on this thing, it barely gets hot, even full tilt on quad cores, so it is sort of a cheat, but you get the idea. I have never seen any condensation on the cooler plate, so I felt quite safe.

Pay the extra money and use worn out water filters.

I use my filter for drinking water, but when I change it out, the old one goes into another pitcher to collect water for other uses, like watering plants, pets, etc.

That way, I get additional use from the filters before I pitch 'em.

Setting up a small still is also an option, but I would recommend a solar powered variety.

Piezo or ultrasonic misters are likely far cheaper to operate.

Living in a similar climate (Finland 60+ N), I do not understand that restriction.

At low outdoor temperatures, such as -20 C to -50 C, the absolute amount of water the air can carry is quite low, still the outdoor Relative Humidity (RH) can be more than 90 %.

When the air condition blows that air into the office environment, it must be heated to +20 C or above. The absolute amount of water vapor remains the same, but the RH can drop below 10 %, since that warmer air would be able to carry much more water vapor.

As mentioned by others, the low RH is a problem due to static electricity (ESD) and in the old days, in very dry air, the paper did not flow smoothly through a line printer, if the air was too dry :-).

In practice, you need at least 30-50 % RH for proper system operation during normal daytime operation.

While it may make sense to drop the night time or weekend temperature to say +18 C to conserve energy, you should not go much below that, since the inner wall and other surfaces will take a long time each morning/Monday to reach comfortable temperatures at a high energy cost.

At low outdoor temperatures (below -20 C) it doesn't be a problem running daily at +20 .. 23 C indoor below 50 % RH and dropping to +18 C at night, we are still well below condensation.

It doesn't make sense to let the night time _air_ temperature to drop too low, since you would need to heat the surfaces the next morning anyway (possibly by IR-heaters).

Anyway, turn off the humidifier at 12:00 and turn of heating at 15:00 and the RH doesn't fluctuate too much.