Lights and high ambient temperatures

BundyRoy

Dedicated elf
Joined
Apr 9, 2014
Messages
1,026
I'm thinking of getting some more smart pixels but this is for lights in general. Last year we had a few days where the ambient temperature didn't drop below 40C until between 9-10pm (so lights are on in that temp). It was hot. 43-45 in the day with one day at 48C (although that was just after new years so lights off by then).

So how does 40C temp effect life operation and lifespan. I had some leds (Big W fairy lights) on the roof last year and they still work but the silicon casing (well I hope it's just the casing) has gone a nice brown/grey colour from the heat.

Just wanted to double check before the spend.
 

Kaden

Pixels! I need more pixels!
Joined
Jan 10, 2012
Messages
724
Location
Gold Coast
UV light will cause the waterproofing to go brown, but heat won't affect the led strip/string.

Also the brown color won't be noticeable at night time.

I would pay attention to keeping the power supplies cool, their efficiency is affected by heat.
 

BundyRoy

Dedicated elf
Joined
Apr 9, 2014
Messages
1,026
Thanks Kaden. I was hoping it was just the covering. I had noticed the brown colouring didn't seem to effect the lights at night.

From what I had seen in some of the posts about enclosures and ventilation holes I figured I might have to be careful with the power supplies.
 

David_AVD

Grandpa Elf
Community project designer
Generous elf
Joined
Jun 12, 2010
Messages
4,681
Location
Victoria Point (Brisbane)
+1 about PSU ventilation. Some like to cram as much in their PSU/controller boxes as possible, but it pays in the longer run to allow more space and ventilation.
 

BundyRoy

Dedicated elf
Joined
Apr 9, 2014
Messages
1,026
Whilst I think of it, if a power supply shuts down from excessive heat is there a indicator light or something that lets you know the cause. I guess that if it restarts once cool there can be only one cause. It's just whether the overheating is caused by ambient temp or overloading that becomes the issue.
 

Kaden

Pixels! I need more pixels!
Joined
Jan 10, 2012
Messages
724
Location
Gold Coast
BundyRoy said:
Whilst I think of it, if a power supply shuts down from excessive heat is there a indicator light or something that lets you know the cause. I guess that if it restarts once cool there can be only one cause. It's just whether the overheating is caused by ambient temp or overloading that becomes the issue.


As with most things it is more complicated than that.


A PSU can shutdown due to overload.
A PSU can shutdown due to overheat.
A PSU can shutdown much sooner due to both, allow me to explain.


The hotter the PSU gets, the lesser it's efficiency (The figures below are examples to demonstrate my point as I can't find the effiency graph at the moment)
- a 350 watt PSU running with a 300 watt load will not shutdown if the ambient temp is 25 degrees celcius
- a 350 watt PSU running with a 50 watt load will not shutdown if the ambient temp is 25 degrees celcius
- a 350 watt PSU running with a 300 watt load *may* shutdown if the ambient temp is 50 degrees celcius
- a 350 watt PSU running with a 50 watt load *may not* shutdown if the a ambient temp is 50 degrees celcius.


If I can find the graph I will update this post.
 

David_AVD

Grandpa Elf
Community project designer
Generous elf
Joined
Jun 12, 2010
Messages
4,681
Location
Victoria Point (Brisbane)
The efficiency of the power supply can change with load, but doesn't change with temperature. Maybe what you're getting at is the "Derating Curve" that needs to be applied?

In a nutshell, a power supply may run at 100% load for ambient temperatures up to say 50°C. At 60°C the max permissible loading may drop to 50% and fall to under 10% once the ambient temperatures hit 60°C !

If the power supply is cooped up inside a sealed (or jam packed) enclosure, the "ambient" inside the box can be much higher than outside. This can mean it will shutdown well before the outside air reaches what you'd call a high temperature.

EDIT: Added derating curve chart from a Mean Well SP-320 series PSU.
 

Attachments

  • derating_curve.png
    derating_curve.png
    15.5 KB · Views: 14

Kaden

Pixels! I need more pixels!
Joined
Jan 10, 2012
Messages
724
Location
Gold Coast
David_AVD said:
Maybe what you're getting at is the "Derating Curve" that needs to be applied?


That's the one :) You explained it better (and more accurately).
 

BundyRoy

Dedicated elf
Joined
Apr 9, 2014
Messages
1,026
I think I read somewhere today (can't find it now) that if you are operating in hot temps you are better off with 12V lights. Is this right. If so why, does heat cause the resistance of the wiring to go up and hence more voltage drop or is it just to with heat effecting a 5v psu more than a 12V.
 

multicast

Senior elf
Joined
Jul 13, 2013
Messages
715
David_AVD said:
Off the top of my head I can't think of any real reason 12V would be better than 5V when it comes to ambient heat.


The only thing i could come up with, is that generally lower output voltage PSU's are less efficient, than Higher voltage ones.


If you're hitting 45 ambients, you need to seriously think about the problem. Active cooling ( thats not just forced air with fans ) becomes a possibility.
 

BundyRoy

Dedicated elf
Joined
Apr 9, 2014
Messages
1,026
Thanks for the feedback Qiang Fu but it would basically never be 45 when lights are on. Max of 40 and even then, it is not that often. If it becomes a problem would just turn lights off. Would be too hot to leave the airconditioning inside anyway.
 

fasteddy

I have C.L.A.P
Global moderator
Joined
Apr 26, 2010
Messages
6,648
Location
Albion Park NSW
Looking at the de-rating graph it shows there is a slight advantage using 12vdc as far as the power supply is concerned with high ambient temperatures.
Also at higher ambient temps the cable resistance does go up slightly so using 12vdc could mean a slightly less percentage of voltage drop

As far as 12vdc pixels it only becomes a disadvantage using 12vdc in hotter temps if the LED is single LED control and is run at the full rated current because this means a lot of heat is dissipated through the resistor to bring the voltage down to drive the LED. This is why most 12vdc single LED pixels are usually driven at around 9.5mA instead of the normal full rating of 18.5mA to 20mA
 

multicast

Senior elf
Joined
Jul 13, 2013
Messages
715
However theres more to the picture..


If you are using a 12V psu at 88%, vs a 5V psu at 84%, Remember you're running considerably more power out of the 12V and that might just mean the amount of heat you have to get rid of from your PSU is worse for 12V.


simple comparative example;


10 strings of 50 pixels at 10mA per channel = 500 x .01 x 3= 15A


15A @ 12V * (1-.88) = 21W of energy to get rid of
15A @ 5V * *(1-.84 ) = 12W of energy to get rid of out of your box.


So, while your 12V is more efficient, you quite likely have more energy to get rid of anyway, and its the total thermal load that is important here.
 
Top