Welcome to Laser Pointer Forums - discuss green laser pointers, blue laser pointers, and all types of lasers

Buy Site Supporter Role (remove some ads) | LPF Donations

Links below open in new window

FrozenGate by Avery

what is the importance of good heat sinking?

Joined
Jan 10, 2009
Messages
18
Points
0
There are several people here who have killed a lot more diodes than I'd care to, and I was wondering if I could get your input on this subject.

How important is it to heat sink these (mostly I'm thinking of the phr-803t's) diodes effectively?

It would seem that conventional wisdom is to use the largest heat-sink possible, but then I see builds like the leadlights, and the keychain builds, but don't hear anything about decreased diode lifespan.

Perhaps these diodes have become inexpensive enough to justify builds that don't have long life expectancy, or perhaps it's becoming clear that the diodes suffer more from being over driven than from heat saturation?

Any insight would be interesting to me.
 





well the diode can be damaged from heat, they "seem" to be tough but they really aren't. some die at 90mA some live comfortably at 180mA. some die instantly even with the best heat sinking at 110mA but some live with no heat sink at 150mA. I think someone on here drove one at 300mA and it lived a while :P each diode is unique.
 
this is exactly what I'm wondering; if the heat sinking doesn't seem to be a factor in these diodes living or dying, how important is it, really?

Obviously it might help, but if diodes with low current are dying in large heatsinks, and diodes with high currents are living with no (or little) heat sinking, is the heat sink making a real difference?
 
1st: Many laser enthusiasts like to push their LD's beyond what they were designed and built for. In the drive (correct me if I'm wrong) they operate in pulsed, we hobbyist like CW, or constant wave, meaning that it is constantly running, 100% of the time it's switched on. In the drive, there is heatsinking, although it isn’t “on” the entire time. So for our application, heatsinking will help keep the diode cool, because when we push a 120mW PULSE rated diode at 130+mW CW, it gets hot quick and heat is ONE way a diode can destroy itself. Now if the same diode is running at 40mW, no heatsink required, it simply won’t get hot enough, although having one is never a bad thing.

2nd. The (IgnorT’s term) “Duty Cycle Monster”. If you plan on leaving your laser on for 5 seconds, then heatsinking really isn’t necessary, because it won’t be on long enough to get hot. Time is the dimension we all have to consider in any formula. So if you plan on leaving it on for several minutes at a time, then good heatsinking is a must, but again, if you’re pushing the diode at or beyond it’s confirmed rating.

3rd. When comparing lead lights, pointers, and all the small high-powered green lasers that have little to no heatsink, remember, that the IR LD has been evolved for decades, over the blu-ray that has only been around for a few years now. So those LD’s are rated around 500mW or 1W, or what ever depending on the laser’s output power, and are pushing at 100mW (give or take) less then the diodes rating. So they don’t overheat because they aren’t being overworked. When we have 500mW blu-ray diodes, then you can EASILY have a 350mW blu-ray laser that will last literally FOREVER…
 
Generally you must keep the maximum die temperature substantially under 150C. I seem to recall that the theta-JC (thermal resistance Junction to Case) is around 60 to 80 K/W for the 5.6mm packages, and given a diode operating point of say, 130ma with a forward drop drop around 4.8V for a bluray, this is 0.624w dissipated in the die / package. This will cause the die temp to rise by 0.624W X 70K/W = 44C rise above case temp. if you want your die temp to rise no higher than say, 125C, the case temp cannot be allowed to rise above 125-44C = 81C. If max design ambient = 50C, then you need to heat sink rated for at least (81-50)/.624 = 33C/W. If you drive the device harder, you will need correspondingly more heat sink.

Another design consideration is the fact that operating life will be decreased by a factor of two for every 10C that the die temp / case temp rises. This degradation is driven by chemical diffusion which, above an activation energy, increases exponentially with temp. Basically, bigger heatsink = less temp rise above ambient = longer life. (google on arrhenius equation) In the example above, going from a 33 C/W heat sink to a 18 C/W heat sink will decrease die temp by 10C and double your laser diode operating life in a 50C ambient.

An added benefit of heat sinking is that you will also get better power stability. With increasing die temperature, the lasing threshold current increases and the slope of the output curve decreases. This is why LD's get dimmer as they warm up . If you can hold the case temp closer to ambient with a larger heat sink, you will achieve better power stability with time because the die will not heat up as much.

Finally, laser diodes have another failure mechanism, namely catastrophic optical damage. If you increase the drive current too high, the optical energy density within the resonator exceeds the damage threshold of the materials. When this happens, the die melts or cracks. The point at which catastrophic optical damage occurs can vary from device to device, depending on any imperfections, manufacturing variability from part to part, lasing threshold and slope, die temperature, etc. This is why one diode may operate for a long time at 180ma, while another diode will fail immediately at 100mA current. Reflecting laser light back into the resonator from poorly designed optics can also increase the optical flux within the laser cavity, causing the damage to occur at a lower drive level. Catastrophic optical damage will happen in a matter of microseconds due to electrical transients or optical overstress. Interestingly, cooling some laser diodes too much without decreasing drive current can trigger the catastrophic failure mechanism due to shift in threshold current and slope causing higher emission at low temp.
 
Achro said:
Generally you must keep the maximum die temperature substantially under 150C. I seem to recall that the theta-JC (thermal resistance Junction to Case) is around 60 to 80 K/W for the 5.6mm packages, and given a diode operating point of say, 130ma with a forward drop drop around 4.8V for a bluray, this is 0.624w dissipated in the die / package. This will cause the die temp to rise by 0.624W X 70K/W = 44C rise above case temp. if you want your die temp to rise no higher than say, 125C, the case temp cannot be allowed to rise above 125-44C = 81C. If max design ambient = 50C, then you need to heat sink rated for at least (81-50)/.624 = 33C/W. If you drive the device harder, you will need correspondingly more heat sink.

Another design consideration is the fact that operating life will be decreased by a factor of two for every 10C that the die temp / case temp rises. This degradation is driven by chemical diffusion which, above an activation energy, increases exponentially with temp. Basically, bigger heatsink = less temp rise above ambient = longer life. (google on arrhenius equation) In the example above, going from a 33 C/W heat sink to a 18 C/W heat sink will decrease die temp by 10C and double your laser diode operating life in a 50C ambient.

An added benefit of heat sinking is that you will also get better power stability. With increasing die temperature, the lasing threshold current increases and the slope of the output curve decreases. This is why LD's get dimmer as they warm up . If you can hold the case temp closer to ambient with a larger heat sink, you will achieve better power stability with time because the die will not heat up as much.

Finally, laser diodes have another failure mechanism, namely catastrophic optical damage. If you increase the drive current too high, the optical energy density within the resonator exceeds the damage threshold of the materials. When this happens, the die melts or cracks. The point at which catastrophic optical damage occurs can vary from device to device, depending on any imperfections, manufacturing variability from part to part, lasing threshold and slope, die temperature, etc. This is why one diode may operate for a long time at 180ma, while another diode will fail immediately at 100mA current. Reflecting laser light back into the resonator from poorly designed optics can also increase the optical flux within the laser cavity, causing the damage to occur at a lower drive level. Catastrophic optical damage will happen in a matter of microseconds due to electrical transients or optical overstress. Interestingly, cooling some laser diodes too much without decreasing drive current can trigger the catastrophic failure mechanism due to shift in threshold current and slope causing higher emission at low temp.

Pretty good summary!

Don't forget when calculating the power dissipation in the Laser diode, to subtract the Optical Power Output from the total power, as it is dissipated outside the LD... ;)
 
Ace82 said:
3rd.  When comparing lead lights, pointers, and all the small high-powered green lasers that have little to no heatsink, remember, that the IR LD has been evolved for decades, over the blu-ray that has only been around for a few years now.  So those LD[ch8217]s are rated around 500mW or 1W, or what ever depending on the laser[ch8217]s output power, and are pushing at 100mW (give or take) less then the diodes rating.  So they don[ch8217]t overheat because they aren[ch8217]t being overworked.  When we have 500mW blu-ray diodes, then you can EASILY have a 350mW blu-ray laser that will last literally FOREVER[ch8230]


but he was talking about BR in leadlight or keychain hosts....
so question still remain and i am curious about this as well...

cheers
 
nozombies said:
There are several people here who have killed a lot more diodes than I'd care to, and I was wondering if I could get your input on this subject.

How important is it to heat sink these (mostly I'm thinking of the phr-803t's) diodes effectively?

It would seem that conventional wisdom is to use the largest heat-sink possible, but then I see builds like the leadlights, and the keychain builds, but don't hear anything about decreased diode lifespan.

Perhaps these diodes have become inexpensive enough to justify builds that don't have long life expectancy, or perhaps it's becoming clear that the diodes suffer more from being over driven than from heat saturation?

Any insight would be interesting to me.

Normally, heat sinking would become more important, the harder you intend to drive the diode. If you were driving a PHR real hard, at say, 175mA and outputting 140mW (just an example), you would want good heat sinking because it might only run for a few hours total, even then, and a key chain size host would get too hot. But if you cut it back to 100mA, a key chain size host would still definitely warm up, but at that stress level, it's not a problem.

But the main reason you won't be hearing too much about overheating being a problem with these keychain hosts is even simpler than all that. The batteries in these are so low in capacity (as little as 90mAH for 10180) that they will run out of energy from the battery before they have much of a chance to heat up! ::)
 
What a timely topic.  Just tonight I finished machining the ATM-20 (Aixiz Table Mount 2") Which is a clam shell heat sink for Aixiz laser diode module.  This is the prototype, but it looks and works perfectly.  The final sales version will be black anodized.  It is one inch square and has a 2" above the surface output.  The base is thru drilled on 1" centers to mount to optical tables.  One could use 1/4 x20 or 10/32" screws for mounting.

I saw what great things this community is doing at this forum and this is the first of many items that I would like to contribute.  My best guess is around $30.00 in quantities of one.  I am also able to laser engrave anything that you would like for an additional charge.  If you are interested in one or more..PLEASE!!!! PM ME!!!!  I am trying to attach a picture but it dosnt look like it is working.

Pat B.
 

Attachments

  • Ebay_craigslist.jpg
    Ebay_craigslist.jpg
    132.7 KB · Views: 118
To continue with this discussion..............do 9mm diodes tend to be more "forgiving" than 5.6mm diodes as far as overdriving?

Thanks as always,

Ray
 
The physical size of the diode package doesn't matter much as far as whether or not it needs heatsinking. Typically with low powered laser diodes, heatsinking is a source of peace-of-mind for the builder. It is necessary with high-powered diodes, since they naturally get quite hot, but as for the diodes we routinely see here on LPF, heatsinking is only really necessary if using long duty cycles. If TECs are used, then heatsinking is required for the TECs to efficiently dispose of the heat they absorb.

So to answer your question, no, IMHO, they are not crucial to the longevity of low-powered (<150mW) diodes. Once you get into higher powers than that, however, heatsinking is required.

Heatsinking is also more important for green lasers than many manufacturers think. Even a 5mW green laser is pumped by a 200-250mW diode. 200-250mW from a 5.6mm package means that heat will become an issue, Most small green pointers do not have anywhere near enough heatsinking, which usually leads to power drops/surges and mode-hopping.
 





Back
Top