Flashlight body, thermal resistance, measure...

17 posts / 0 new
Last post
arenat
Offline
Last seen: 2 years 2 months ago
Joined: 11/22/2010 - 01:00
Posts: 367
Location: Argentina
Flashlight body, thermal resistance, measure...

I manage to measure the Hugsby's P31 body thermal resistance, the method used is in the link below.

http://sound.westhost.com/heatsinks.htm#s4

 

To heat up the body used a BD438 , a little but powerful transistor, 36 W. The reflector is out  to access the pill with the wires.

 

The body was hold in a vise thermal isolated and heated at various currents...the pic below at 280 mA 3.2 V =.896W...the rise temperature was minimal , around 5 C,   ambient here 29C

 

Now what happen at 1,28 A 3,76 V=4,9W , the rise is 36C , too much, in an XR-E the junction temperature is at 104 C.

 

Below a pic heated at 2,7W perfectly stabilized at 56C even one hour later

 

 

Given the measures made the thermal resistance of this body is around 10 C/W ..

Can manage 3W easily and I guess a little more in winter time.

An XM-L in this body has a total thermal resistance of 13,5 C/W at 1A 3W could give 352 Lumen with a rise of 40C ...

Edited by: arenat on 01/08/2011 - 15:39
Don
Don's picture
Offline
Last seen: 10 months 3 weeks ago
Joined: 01/12/2010 - 16:32
Posts: 6617
Location: Scotland

Interesting!

 

Thanks for doing this, it's always useful to know. Do you have any stainless steel lights you could measure?

 

The numbers from my light tests are always to be found here.

https://spreadsheets.google.com/ccc?key=0ApkFM37n_QnRdDU5MDNzOURjYllmZHI...

arenat
Offline
Last seen: 2 years 2 months ago
Joined: 11/22/2010 - 01:00
Posts: 367
Location: Argentina

I have 2 P31 , one RC-G2 and other in the way,one  MTE F15 and 2 C78 ....all aluminium, forgot to say, the pill of this P31 is filled with melt solder , I know that  alloys aren't good  thermal transmiters but is better than air

Match
Match's picture
Offline
Last seen: 5 years 5 months ago
Joined: 11/14/2010 - 11:57
Posts: 1488
Location: South Carolina, USA

Very nice test, arenat.  I'm assuming this represents worse case scenario in which the torch is left on high while set down.  I wonder how much impact the torch being held would change the outcome.  The little P31 might be able to sustain 5w biologically cooled.

arenat
Offline
Last seen: 2 years 2 months ago
Joined: 11/22/2010 - 01:00
Posts: 367
Location: Argentina

What you are assuming is correct Match. Here I'm follow the most conservative path. In fact this torch days ago has had a XP-G R5 at 1,4A and meanwhile hot never pass I guess 50 C after running 5-6 minutes.That was 5W. The thermal contact was bad due the reflector/star height issue and a lot of light were missing but anyway the output was impresive in a dark enviroment, it lighting up all in front 12 meters wide.

I'm planing an XM-L at around 4-5W for this one, the big die maybe match better than the XP-G one which look more like a flame all unfocus...plan B an XR-E R2 or XR-E Q4 warm at 1A or a little more because I'm using the 3 modes drivers...

Budgeteer
Budgeteer's picture
Offline
Last seen: 7 years 11 months ago
Joined: 08/24/2010 - 02:19
Posts: 3087
Location: Slovenia (EU)

This thread put a spark on my wish to buy a non contact thermometer to measure heat buildup on flashlights (with emitters installed).

kragmutt wrote:

They're gonna send you a green redcat with a black LED.

arenat
Offline
Last seen: 2 years 2 months ago
Joined: 11/22/2010 - 01:00
Posts: 367
Location: Argentina

Budgeteer wrote:

This thread put a spark on my wish to buy a non contact thermometer to measure heat buildup on flashlights (with emitters installed).

 

I own a decent one and don't work for small things close range, ,  you never know what is measuring also the type of material change the reading. Those are aimed at enough area stuff, engines parts ,freezer, oven and the like...

The proper way to measure the emitter temperature is attaching a K thermocouple to the ceramic substrate  near as possible the solder pad... could be glued, soldered, press with a bolt and a plastic piece to isolate it, even hold by hand .

Budgeteer
Budgeteer's picture
Offline
Last seen: 7 years 11 months ago
Joined: 08/24/2010 - 02:19
Posts: 3087
Location: Slovenia (EU)

I was wondering if i could go away with:

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=390225147065&rvr_id=1...

Would not care if the temp registered was in the +- 2C ballpark

 

Guess not...

kragmutt wrote:

They're gonna send you a green redcat with a black LED.

Budgeteer
Budgeteer's picture
Offline
Last seen: 7 years 11 months ago
Joined: 08/24/2010 - 02:19
Posts: 3087
Location: Slovenia (EU)

On a side note, is it really that simple? Comparing heat from a resisor to heat from a led at about same wattage? I would guess the resistor is prone to emitt more heat than a led... or not?

A conventional lightbulb has the "nice" feature to emit 3% energy as light the rest is heat...

kragmutt wrote:

They're gonna send you a green redcat with a black LED.

sixty545
Offline
Last seen: 1 year 7 months ago
Joined: 10/25/2010 - 14:15
Posts: 702
Location: Denmark (GMT + 1)

Budgeteer wrote:

I was wondering if i could go away with:

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=390225147065&rvr_id=1...

Would not care if the temp registered was in the +- 2C ballpark

 

Guess not...

I have one that looks a lot like that, but other colors and a laser switch button and C/F display switch (did cost me $35 in a local store). It has a distance Spot Ratio: 8:1  "Yours" has a better (narrower) ratio 12:1

I just tested on a WF-502B with XP-G at 1.4Amp - Ouch - 49 celsius, very plausible.

 

Budgeteer
Budgeteer's picture
Offline
Last seen: 7 years 11 months ago
Joined: 08/24/2010 - 02:19
Posts: 3087
Location: Slovenia (EU)

How was the 49 degrees by touch? Guess very hot but holdable.

kragmutt wrote:

They're gonna send you a green redcat with a black LED.

sixty545
Offline
Last seen: 1 year 7 months ago
Joined: 10/25/2010 - 14:15
Posts: 702
Location: Denmark (GMT + 1)

Budgeteer wrote:

How was the 49 degrees by touch? Guess very hot but holdable.

Exactly, If I got paid I think I could hold on to it. Wink

arenat
Offline
Last seen: 2 years 2 months ago
Joined: 11/22/2010 - 01:00
Posts: 367
Location: Argentina

I was so upset about the broken lens that forgot completely the Ultrafire P10...Laughing

This thing is so lightweight that seem weaker than the P31 in this aspect. Well, the greater area do his job and the numbers are better .Around 8,8 C/W , ambient today 30,5 C.

Sadly for the record this one has weight added to the pill, a few grs but given the lightness of this thing is around 20% of the total

 

No, is not that simple but the hard part is in the datasheet, that parameter is called thermal resistance. If I'm not wrong,  since I'm not much into the matter just read a bit about , knowing the thermal resistance of the emitter in question,XR-E 8 C/W, XP-G 6 C/W , XM-L 2,5 C/W  at more efficient the lower the number. As say, knowing the thermal resistance  of both, emitter and body is possible predict with reasonable certainty how hot  an arrange could go at a given current.

That is good to keep the led in the cold area where the thermal loss is minimal and if not at least to know what can expect really.

sixty545
Offline
Last seen: 1 year 7 months ago
Joined: 10/25/2010 - 14:15
Posts: 702
Location: Denmark (GMT + 1)

arenat wrote:

No, is not that simple but the hard part is in the datasheet, that parameter is called thermal resistance. If I'm not wrong,  since I'm not much into the matter just read a bit about , knowing the thermal resistance of the emitter in question,XR-E 8 C/W, XP-G 6 C/W , XM-L 2,5 C/W  at more efficient the lower the number. As say, knowing the thermal resistance  of both, emitter and body is possible predict with reasonable certainty how hot  an arrange could go at a given current.

That is good to keep the led in the cold area where the thermal loss is minimal and if not at least to know what can expect really.

You're quite right. The thermal resistance is equivalent to electrical resistance R. If you call the temperature difference between two sides for V and the heat that flows from the hot to the "cold" side for I you can use Ohms law V = R * I to calculate.

V is degrees (celcius and Kelvin), I is Watt and the unit of R is degr. / Watt.

The easiest is to think of a wall with different temperatures on the two sides. It has a thermal resistance R and there is a heat flow through the wall (I  Watt). Hope you can use this (for understanding)

arenat
Offline
Last seen: 2 years 2 months ago
Joined: 11/22/2010 - 01:00
Posts: 367
Location: Argentina

Many thanks sixty545,but  I knew that, look at the first post.

The Ultrafire R5 at 1A as came from factory could give 283 Lumen.

That's the result of working close at 80C where you get 90% of 250% minus 7% blocked by the lens...that at an ambient T of 25 which is average  in many locations.

At 1,4 A the junction could reach 105 C easily and that mean 80% , 20% loss....

sixty545
Offline
Last seen: 1 year 7 months ago
Joined: 10/25/2010 - 14:15
Posts: 702
Location: Denmark (GMT + 1)

arenat wrote:

Many thanks sixty545,but  I knew that, look at the first post.

The Ultrafire R5 at 1A as came from factory could give 283 Lumen.

That's the result of working close at 80C where you get 90% of 250% minus 7% blocked by the lens...that at an ambient T of 25 which is average  in many locations.

At 1,4 A the junction could reach 105 C easily and that mean 80% , 20% loss....

Sorry, it was not my intention to patronize you, I must read the thread more carefully. Perhaps my explanation can be of help to others. I very much appreciate your work and are going to read about the method right away. I just did'nt had the time before.

arenat
Offline
Last seen: 2 years 2 months ago
Joined: 11/22/2010 - 01:00
Posts: 367
Location: Argentina

Do not worried , reading my post I  realize it  was my fault, anyway great explanation and very useful of course.