Air temperature inside totally enclosed plastic box with resistor dissipating 6W inside it?

[waveboy]

Hi,
Supposing you have a totally sealed hollow plastic rectanguloid box. Dimensions are 150mm by 60mm by 40mm
Wall thickness is 1mm
Plastic thermal conductivity is 0.2
There are extremely thin electrical wires going into the box. These wires supply 6W of power to a small resistor (Resistor dims = cylinder of size D=5mm by length = 30mm) which is in the box. There is no ventilation gaps where the wires go through the plastic box.
The box exists in an ambient air temperature of 30 Degrees C.

Please may i ask...How hot is the air temperature inside the box?

 

[IRstuff]

How accurately do you need the answer?

The cheap answer is to assume it's at least the temperature of the surface of the box with 6W leaving it, which is around 55C

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[LittleInch]

Any forced draft or just natural convection?
Box suspended in mid air on a flat surface?
Any radiant heat on the box?

conductivity 0.2 what? (units)


But 6 W - 55 seems quite high to me. that's almost too hot to touch / hold.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Latexman]

Isn't there a number of old threads solving and re-solving similar problems? This all seems very familiar. Maybe using Search (under the thread title between Forum and FAQs) would be fruitful.

Good Luck,
Latexman
Pats' Pub's Proprietor

 

[IRstuff]

Assuming the bottom doesn't contribute

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[LittleInch]

Fair enough.

Might even be a bit low according to this

https://www.electronics-cooling.com...on-heat-transfer-coefficient-on-a-flat-plate/

You could be looking at 5W/m2K.

Angle is everything - you assume it's standing on its end. Put it flat down on the big side and it all changes for the worse.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Compositepro]

I will depend on the elevation of the resistor in the box. Hot air rises from the resistor and therefore box surfaces below the resistor will see almost no heat (not just the bottom of the box).

 

[waveboy]

Any forced draft or just natural convection?
Box suspended in mid air on a flat surface?
Any radiant heat on the box?

conductivity 0.2 what? (units)

thermal conductivity (0.2) is W/(m.k)

Just natural convection
Box on a flat surface
No radiant heat on the box...just room light.

The internal temp of a similar box which had approx 30mmx4mm slits in each end was 71degC, so i'm sure it will be more than that.

If i could find the heat transfer coefficienct for the plastic then i could work it from that, because that has units of W (m^-2)(K^-1)...and i know the area and the wattage

 

[IRstuff]

So, using the same process I showed, you can find the temperature drop across the plastic.

You need to figure out what plastic it is and look up the thermal conductivity, which has units of W/m-K, not W/m^2-K

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[itsmoked]

It will cook like a bad 7-Eleven hotdog if this is a continuous application.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waveboy]

Thanks,
The thermal conductivity of most plastics is approx 0.2.

Though i need some way of getting the area into the calculation, as well as the wall thickness.. The internal temp will reduce as area increases (for the case described). In the case described, the area is 0.0348m^2

The following totally enclose PSU is 60W rated , and dissipates some 6W on full load. It almost certainly has 105degc rated electrolytic capacitors inside it. As such, at an external ambient of 45degC, it is likely to have an internal ambient of some 100degc.
So my guess would be that the internal ambient is approx 55 degc hotter than the external ambient for the case described in the original post. But i'm not sure how the 55degc is calculated?

(Also, it may be well hotter than 100degc inside..and maybe theyre just over-heating the electrolytic capacitors)

LPV-60-48 power supply

https://www.meanwell.com/Upload/PDF/LPV-60/LPV-60-SPEC.PDF

 

[IRstuff]

You already gave the wall thickness.

Let's assume everything goes through one wall

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[EdStainless]

Is there a can on the PSU inside the box?
If so then this approach works as the interior temperature will be fairly uniform.
If the inside of the box is 100C then individual components are likely 10-15C hotter than that.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[waveboy]

Is there a can on the PSU inside the box?

Thanks, No can (no metal inner enclosure)

 

[EdStainless]

I do presume that there is a fan on the PSU to eliminate hot spots.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[georgeverghese]

You'll have to pay someone to run the calcs for you, which will involve a few iterations - this isnt a request for a free tip.

 

[waveboy]

I do presume that there is a fan on the PSU to eliminate hot spots.

Thanks, but there is no fan on the PSU, or in it.

 

[EdStainless]

So the resistor will heat in still air only cooled by natural convection, and that space will then only cool by heat transmission through a plastic wall again only cooled by natural convection?
If that resistor doesn't run 200C I'll be surprised. And then there are the power transistors, APC components, and the caps to worry about.
A PSU in an enclosure is usually built so that all of the heat dissipating components have direct conduction paths to the metal case or to heat sink mounted in the case wall.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[waveboy]

Thanks, Yes thats right only natural convection, inside and out.
I tend to agree with what you say about temperatures…however, the following on page 36 shows component temperatures inside a “totally sealad plastic” case , 65W PSU in 40 degrees C ambient. (so the internal power dissipation would be near to 6W)
…..One of the components (at 230VAC input) is just 85.5degC.
..That must mean that the internal ambient is less than 85.5degC…which is something I find incredibly surprising since it’s a totally sealed plastic case and i thought it would just cook up to 100 degrees C plus. (I wonder if they failed to seal up the thermocouple wire entrance hole?)

Page 36 shows temperature of components in “totally sealed in plastic” 65w offline PSU…

https://ac-dc.power.com/sites/default/files/PDFFiles/der243.pdf

 

[IRstuff]

Seriously? You didn't think to mention ppg 25 through 29, which describes all the heat sink aluminum that's inside?

Not sure what you were referring to, but your link shows much higher temperatures


TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm