Electronics in a hot climate

Completely insufficient data to answer the question... there are

*multiple* unknowns that you have not characterized:

- Actual spectral behavior of the yellow box, and whether it *must* remain yellow or could be painted a different color.

- Power dissipation of whatever is inside the box.

- Thickness of the box walls, and their thermal resistance (e.g. degrees C per watt).

- Whether any form of heat-sinking or heat-pumping can be used.

Without active cooling of some sort, the best results would probably come in situations where the device inside the box dissipates a negligible amount of power, and the exterior of the box can be painted silver or bright white. This would result in a very small temperature rise above the 50C ambient.

A big battery, discharging into a high-wattage amplifier, inside a box with thick walls of foamed plastic (high thermal resistance), pre-heated by the sun to significantly above the 50C ambient... well, "melts and catches fire" sums it up.

As far as the sunlight effects are concerned... I imagine that the effects are worse if the yellow plastic allows for substantial penetration of the sunlight, but has a low thermal conductivity (again, foamed plastic might be an example). This would create the equivalent of the "greenhouse effect", in which heat is admitted by one mechanism (solar radiation), converted to a different form (kinetic heat in the box interior), and trapped in that form.

Heat penetration via this mechanism might be minimized to some extent by undercoating the yellow exterior paint with a fully-reflective coating... bounce as much of the heat back off/out of the box before it penetrates the interior.

--
Dave Platt                                    AE6EO
Friends of Jade Warrior home page:  http://www.radagast.org/jade-warrior
  I do _not_ wish to receive unsolicited commercial email, and I will
     boycott any company which has the gall to send me such ads!

Thanks for the response.

Yellow is specified by the customer

Minimal, less than 0.25W

5mm, 0.2" walls. The material is PBT with glass fibre, 30% from memory. I'll have to set up a test to measure thermal resistance.

No

I think you're probably going to have to simulate in order to get a dependable estimate of the heat rise.

Since the power dissipation is so small you can probably neglect it, and just test the case under the correct temperature and irradiation levels. I think that's also going to minimize the impact of the thermal resistance of the walls, since only a tiny fraction of the heat the box is dealing with, will be internally generated. The final temperature is going to be dominated by externally-impressed heat flux (i.e. the sunlight) and this will have plenty of time to soak in through the walls during the day, unless you line the box with Space Shuttle carbon-carbon tiles :-)

You'd need a test chamber that you can pre-heat to 50C and hold at that temperature, and a floodlamp arrangement which approximates the irradiance of worst-case sunlight. Or, equivalently, use an internal heater to approximate the heating due to the amount of sunlight absorbed by the box. This would result in an internal temperature somewhat higher than in actual sunlight, as the heat would have to soak out through the walls' thermal resistance. Taking the average of inside-wall and outside-wall temperatures would probably be close enough.

Hmmm. WikiPedia states that sunlight has a luminous efficacy of about

93 lumens per watt, and that bright sunlight is about 100,000 lux (lumens per square meter), which works out to about 1000 watts per square meter. Naturally, not all of that would be absorbed by the box... I'm not sure what its albedo would be... you could probably use a value of .3 (approximate average albedo of land) as a starting point.

So, for a coarse test, you could figure that your box will probably absorb about 300 watts of heat, for every square meter of surface it presents to the sun, under near-worse-case conditions. It will re-radiate some of this, and the rest will have to conduct away to the ground (less, I imagine) and be convected away by the air (more, I think). Hot, still air will reduce convection.

I *think* the interior will stabilize at a temperature which falls short of the boiling point of water... but I'm not sure of that. It'll probably be hot enough to hurt if you touch it for more than a moment!

--
Dave Platt                                    AE6EO
Friends of Jade Warrior home page:  http://www.radagast.org/jade-warrior
  I do _not_ wish to receive unsolicited commercial email, and I will
     boycott any company which has the gall to send me such ads!

"Raveninghorde"

** Sure you can.

** The temperature of the surface of the box facing the sun is the highest and unlikely to exceed 70C in direct sunlight. At around 70C, the rate of heat absorption by the surface become equal to the art of re-radiation by that same surface. The colour of the surface matters too - black being the worst case. The outside air will also have a cooling effect on the box.

Take 70C as the worst case the electronics have to survive, long term.

The battery is gonna be your downfall - many of them will not take that sort of temperature without suffering deterioration or damage.

Better start looking for cell technologies that feature high storage temps in the specs.

..... Phil

Mechanical and aerospace guys are usually quite good at simulating this. But whatever you do, personally I would not use electrolytics in there. Even the 105C and 125C types would not be happy campers.

At to "anywhere in the world": If that includes places like Alaska you need to also consider ultra-low temperatures. For example, crystals can become like molasses when the temps go to -30C or below. Oscillators might squarely refuse to start oscillating. And the sun ain't gonna shine much of the time up there.

Then there's weather. Golf ball sized hail stones, ice pieces flying about, chunks of trees hitting the box. And in Iceland you might hear a gigantic *PHOOMP* upon which the box goes flying, along with a chunk of mountain, lava, smoke, ash, but I guess you can get an exemption from that :-)

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.

Don't forget mist blowing in any vents, sub-zero temperatures or blowing snow sandblasting effects.

Vents or no vents?

100% humidity? Damned if you do, damned if you don't.

Capacitors have an awful temperature coefficient. Make sure your compensate for that.

Cheers

"Martin Riddle"

** One should avoid emotional terms like "awful" when discussing electronics.

Only a few cap type have poor tempcos - notably electros and high K ceramics.

Most poly caps are likely to remain within +/-1 % within the range of 0 -

60 C.

** Not necessary at all in the vast majority of cases - like coupling, decoupling and PSU filtering. Electros actually work better when hot as their ESR goes down by several times.

Polystyrene caps have the best tempco and polypropylene types have a small, negative tempco.

..... Phil

It is a sealed unit pressure tested for 1 metre immersion.

Hot, cold and humidity can be tested in a chamber.

On one project I tested abrasion in a gravel filled cement mixer.

No aluminium elctrolytics. There are 2 tants. The MCU is specified to

-40C with internal oscillator and I can test that.

Hot and cold I can test in a chamber. Solar irradiation is not as easy.

The battery is my main concern.

Competitors seem to be using lithium polymer but they are only specified up to about 60C.

Yellow is probably close enough to black as far as longwave thermal radiation is concerned so I wouldn't worry about the visible colour too much. White would be better still.

Do you have any control of the shape of the box or its physical design?

A second skin designed to take the suns direct heat and lose some of it by convection might keep things from getting too hot. As Phil said it isn't likely to get much above 70C in direct sun but that is bad for batteries.

ISTR that after ~55C thermal radiation losses begin to dominate over passive convective cooling so that rate of the temperature climb slows after that.

An outer shell with ribs or fins designed to be self shading (think cactus body) and/or with an insulator behind it covered in foil will significantly slow down ingress of solar heating. A Stephenson screen used in meteorology for air temperatures gives you a rough idea.

If the internal power consumption is small you might be able to buy enough time by lining the box interior with the stuff sold for putting behind radiators (ie foil backed polystyrene foil facing inwards).

You might be able to limit the enclosure temperature rise over the typical daytime to something like 10C above ambient. Do you have the option of having a passive heat shunt inside? Some of the most aggressive challenging hot sunny climates also have cold desert nights.

Regards, Martin Brown

Raveninghorde schrieb:

Hello,

But it should be filled with dry nitrogen before sealing to prevent condensation.

Bye

I'm getting ready to test one with a pressure washer, like you would use to clean your roof or driveway. More for sales hype than actual electronics evaluation, but I'm still looking forward to it!

This could be it's own thread here on SED - the hideous exposures and treatment of one's project. (Intentional or otherwise.)

I'm getting ready to test one with a pressure washer, like you would use to clean your roof or driveway. More for sales hype than actual electronics evaluation, but I'm still looking forward to it!

This could be it's own thread here on SED - the hideous exposures and treatment of one's project. (Intentional or otherwise.)

===============================================

I still cringe thinking of the nasty things Dave.L.Jones did to his multimeters - which were far superior to my daily use meters - grrrrr! :)

hest

ate of

by

ing the

.

that

Not my area of expertise (except as it relates to gels on flash photography), but Rosco has a pretty good online resource that may help answer the transmission and thermal properties of the color "yellow". See this link:

Maybe if the OP can give us the closest Rosco color #, he can get a more accurate estimate here. (?) For example, Roscolux 4590 (CalColor 90) has a transmission of 66%. Link:

CalColor 90 Yellow&imageName=3D../../images/filters/roscolux/4590.jpg

For photography buffs, you can get a whole swatchbook sample pack of Rosco Gels (hundreds of them, each measuring about 1-inch by 3 inches) at bhphoto or adorama. I think I paid under $3 for mine.

_Swatchbook.html

ROFL!

--
SCNR, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.

In very cold conditions, the blowing snow can create some nasty static electricity build up.

Or cover the went with some suitable material letting the air pressure stabilize, put preventing water from penetrating.

That is only 1.1 bar.

How would that device survive a transport in a non pressurized cargo bay at 10000 m ? Would it explode due to the internal 1 bar pressure ?

Understood.

I have a pressure release valve. I never fit a battery in a sealed unit without a pressure release valve.