How can you tell your LS is too hot?

Orion · Aug 1, 2002

If someone is running an LS at 4.5 volts, they need a good heatsink. If you didn't have a good heatsink, how can you tell your LS is too hot? Where would this manifest itself? Would you feel it, would the light exhibit abnormal behavior, . . . would men in dark suits break into your house, and wrestle you to the ground?

Thanks!

hotfoot · Aug 1, 2002

Hmmmm, still struggling a little with this one:

Orion, some of us here put a temperature probe directly to the core PCB (if you're using a LS, not just the bare emitter), as close as possible to the emitter. If the temp of the core PCB stays under 105C you should be pretty safe (if its the 1W that is - the 5W's max allowable core PCB temp is 75C, AFAIK). Since the probe itself acts like a heatsink, it would be wise to factor that in as well.

Its not the most scientific or reliable method, but unless you've got access to some really good equipment, I truly dunno how to do it better. Anyone else care to share good ways to measure this?

**DONOTDELETE** · Aug 1, 2002

I don't know how to tell (Wayne? Or Lambda? oh wait, Lambda's gone camping), but I do know this: the ready-to-install versions of the LS come mounted on a big aluminum plate for a reason. And these are meant to be driven at only the 350Mah factory spec.

For overdriven applications (like DirectDrive using 3x1.5V cells) I would imagine an even bigger heatsink is required, or at least a well-conducting attachment of the stock plate to the (metal) body of the light.

dat2zip · Aug 1, 2002

OK, I just erased my first thought.

2nd thought...

I'd suggest hooking the LS using Arctic Silver3 to a good size heat sink or test flashlight of similar size and mass. Now with the batteries in a battery holder wire up the LED with or without a switch and touch the heatsink over a 1-10 minute duration. If it gets soo hot so that you can no longer touch it ~150F (65C) that is probably too hot for reliable operations.

You can get a good non contact IR thermometers at Radio Shack or (I haven't tried one) an IR ear temperature probe. Since I haven't heard of anyone trying this I can't really recommend the latter.

This will allow you to read the temperature of a fairly small spot near the LED.

Contact thermometers would work too. Use a dab of Arctic silver3 to ensure the temperature probe tip has good thermal conductivity to the spot you are trying to measure.

Absolute in no way apply 4.5V to an LS without a large heat sink. The emitter will fry in about a minute or less. The LS/O will probably exceed the maximum temperature.

Orion · Aug 2, 2002

Well, I know that Wayne has used a 4.5v direct drive setup in his copper pipe mod because of the excellent heat sinking properties of the copper.

If I DO have a decent heat sink, would I still want to play it safe and resistor it by 2 or 3 ohms? I'll tell you right now that I'm using a heatsink that was used to cool a PC processor.

bucken · Aug 2, 2002

dat2zip triggered an interesting thought...

A lot of us probably have one of those thermometers in the medicine cabinet. You know, the ones that you stick in the ear, push a button, and it gives you a temperature in about a second, or so. Has anyone ever tried using one of these to measure anything other than body temperature? Enough temperature range?

If it would work, this might be a cheap and otherwise practical source of a useful monitoring device.

If you still only have one of the "stick it under your tongue" thermometers, then this would be a great way to rationalize an upgrade for the workbench... I mean... medicine cabinet.

Wim Hertog · Aug 2, 2002

No, the IR 'ear' thermometer will not work!
Temp. range: 35-45° C.

Doug S · Aug 3, 2002

No need for any instrumentation. If you are running an LS with 4.5V at it's terminals it will be too hot regardless of any heatsink short of liquid N2 cooling.

Orion · Aug 3, 2002

Any idea what mA the LS would be running at 4.5V, or is that too uneasy to just answer for every setup?

Also, as I stated above, Wayne has directly driven an LS at 4.5V and it seemed to work for him!

*Thinking maybe I should get a 3ohm resistor.

**DONOTDELETE** · Aug 3, 2002

Actually I ran one of Wayne's overdriven 4.5V Blasters for 64 hours and it only got barely perceptibly warm to the touch after a few minutes, and stayed that way to the end...of course that big aluminum MagLite body actually became a gigantic extension of the internal aluminum disc heatsink he installed.

Doug S · Aug 3, 2002

Originally posted by Orion:
Any idea what mA the LS would be running at 4.5V, or is that too uneasy to just answer for every setup?

Also, as I stated above, Wayne has directly driven an LS at 4.5V and it seemed to work for him!

*Thinking maybe I should get a 3ohm resistor.

<font size="2" face="Verdana, Arial">Orion: The is no one answer to this due to the very wide manufacturing variation of Vf @ 350mA as stated by the datasheet [ranges from 2.79 to 3.99V]. If one assumes that the sample has the typical Vf of 3.42v@350mA@25C Tj, and the datasheet value of 1 Ohm for the dynamic resistance then the LED would draw about 1.5A at 4.5V if one could hold the junction temp at 25C which of course is not practical. Due to the -2mV/C tempco of Vf, the current will likely be even higher than the 1.5A given above. If you really need to know the answer to the question you posed, you need to test your particular sample. You will need a good bench power supply capable of supplying at least 2 amps at 4.5V. Unless your test set up is very "stiff" [voltage doesn't droop with increasing current] you will need to adjust the power supply voltage repeatedly as the LED heats up an draws increasing current.
Regarding your statement about Wayne's set up, I haven't seen the post that you are referring to so I am guessing here. I suspect that you [or Wayne] are confusing "running at LED at 4.5V" with running on 3 alkaline cells that when fresh have an *open circuit voltage* of 4.5V. This later set up does not deliever 4.5V to the LED.

**DONOTDELETE** · Aug 3, 2002

It's too hot when you can heat coffee with it.

**DONOTDELETE** · Aug 3, 2002

That's it. I'm breaking down and buying that Radio Shack IR temp probe. Actually it's a steal at $59.95...

Of course I could always wait for one of their famous $2.97-priced Clearance Sales, maybe in the 23rd century when they blow out old stock...

Doug S · Aug 3, 2002

Originally posted by MR Bulk™:
That's it. I'm breaking down and buying that Radio Shack IR temp probe. Actually it's a steal at $59.95...

Of course I could always wait for one of their famous $2.97-priced Clearance Sales, maybe in the 23rd century when they blow out old stock...

<font size="2" face="Verdana, Arial">It might be worth checking on Ebay.
Don't forget that you need to take into account the junction to board thermal resistance to get actual junction temperature. For example, if your IR thermometer is looking at the back of a "5W" Luxeon Star which is running at 3W, the actual junction temperature will be 3W X 11C/W= 33C higher than the reading. 11C/W is the junction to board thermal resistance of the "5W" Luxeon Stars. It is 17C/W for the "1W" devices.

Orion · Aug 3, 2002

I suspect that you [or Wayne] are confusing "running at LED at 4.5V" with running on 3 alkaline cells that when fresh have an *open circuit voltage* of 4.5V. This later set up does not deliever 4.5V to the LED.

<font size="2" face="Verdana, Arial">I was talking about 3 alkalines running the LED. So what do you mean that the latter set up does not deliver 4.5 volts to the LED?

Doug S · Aug 3, 2002

Originally posted by Orion:

So what do you mean that the latter set up does not deliver 4.5 volts to the LED?

Click to expand...

<font size="2" face="Verdana, Arial">What part of my statement isn't clear?? Sorry, I couldn't resist

. Because alkaline cells have internal resistance, when they are supplying current their terminal voltage drops below what it is when they are not. The higher the current, and the longer they provide it, the more the voltage drop. For example, 3 new alkaline "C" cells driving a Luxeon Star with "typical" characteristics will have a terminal voltage of under 4 volts after the first few seconds. The voltage will continue to drop since the internal resistance of the cells will continue to rise as the are used.

How can you tell your LS is too hot?

Orion

Flashlight Enthusiast

hotfoot

Flashlight Enthusiast

DONOTDELETE

Guest

dat2zip

Flashlight Enthusiast

Orion

Flashlight Enthusiast

bucken

Enlightened

Wim Hertog

Enlightened

Doug S

Flashlight Enthusiast

Orion

Flashlight Enthusiast

DONOTDELETE

Guest

Doug S

Flashlight Enthusiast

DONOTDELETE

Guest

DONOTDELETE

Guest

Doug S

Flashlight Enthusiast

Orion

Flashlight Enthusiast

Doug S

Flashlight Enthusiast

Similar threads

Latest posts

How can you tell your LS is too hot?

Flashlight Enthusiast

Flashlight Enthusiast

**DONOTDELETE**

Guest

Flashlight Enthusiast

Flashlight Enthusiast

Enlightened

Enlightened

Flashlight Enthusiast

Flashlight Enthusiast

**DONOTDELETE**

Guest

Flashlight Enthusiast

**DONOTDELETE**

Guest

**DONOTDELETE**

Guest

Flashlight Enthusiast

Flashlight Enthusiast

Flashlight Enthusiast

Similar threads

DONOTDELETE

DONOTDELETE

DONOTDELETE

DONOTDELETE