super glue as thermal expoxy

well thats all i had.so i super glued a luz 1 water to a big heat sink from a stereo or something i foget but its was big well that heat sink got hot as hect . proveing to me it remved heat from lux i was overdriveing it very very high and it ran all night to batts died .was bright as hect so is it posable that super glue trasnfers heat?

I think Dat2zip did an experiment once, and used spit to transfer the heat, so i think superglue would be fine. But i'd bet that a superglued Lux would degrade faster than one that was epoxied with proper thermal stuff.

It sounds like you proved that super glue works. I didn't think it would hold up to high heat. But I know it does a good job of electrical insulation.

It works well. I used it to mount a heat sink on my 5w bike light. Dispite lots of use, it never overheated.

Al

Air:
Thermal conductivity 0.025-0.031 W/mK

Fiberglass:
Thermal conductivity 0.04 W/mK

Superglue Loctite 382:
Coefficient of thermal conductivity, ASTM C177, 0.11 W.m-1K-1
Glass transition temperature, ASTM E228, ºC 130
(*most epoxies transistion around 50-65 ºC)
(some high temperature epoxies transition 90 ºC and above-specialized)
(glass transition temperature is when material gets soft and
bonds get weak)

http://www.loctite.se/tds/382.pdf


Picked this as a typical epoxy Loctite Hysol 3450:
Coefficient of thermal conductivity, ASTM C177, 0.28 W.m-1K-1
Bond strength at temperature is shown on datasheet chart.
http://www.loctite.at/tds/3450.PDF

Nylon (Polymide):
Thermal conductivity 0.24 W/mK

Pure Water:
Thermal conductivity 0.58 W/mK

Wakefield 120 Thermal Grease:
Thermal Conductivity 0.735 W/mK
http://power.ece.uiuc.edu/Balog/images/How%20to%20keep%20your%20cool%20when%20working%20with%20power%20electronics.pdf

Fused Silicon Dioxide:
Thermal Conductivity 1.4 W/mK

Dow Corning SE4447 CV Thermal RTV:
Thermal Conductivity 2.5 W/mK
http://www.dowcorning.com/DataFiles/090007b58021adab.pdf

Artic Silver Thermal Epoxy:
Thermal conductivity. Greater than 7.5 W/mK
http://www.arcticsilver.com/arctic_silver_thermal_adhesive.htm

Stainless Steel:
Thermal conductivity 15-25 W/mK

Loctite Thermal Adhesive QMI 5030:
Thermal Conductivity, 25 W/mK
http://www.loctite.com/int_henkel/loctite/binarydata/pdf/lt3758a_ThermMgmt.pdf
http://www.loctite.at/tds/QMI5030.pdf

Alumina (Aluminum Oxide):
Thermal Conductivity 30 W/mK

Lead:
Thermal conductivity 34.7 W/mK

Carbon Steel:
Thermal conductivity 43-64 W/mK

Aluminum:
Thermal Conductivity 190 W/mK

Aluminum Nitride:
Thermal Conductivity 260 W/mk


Boron Nitride:
Thermal Conductivity 250-300 W/mk

Copper:
Thermal Conductivity 386 W/mK

Silver:
Thermal Conductivity 406 W/mK

Diamond:
Thermal Conductivity 500- 1000+ W/mK

mix in some Artic Silver with it /ubbthreads/images/graemlins/smile.gif

[ QUOTE ]
SureFireRocks said:
Air:
Thermal conductivity 0.025-0.031 W/mK

Fiberglass:
Thermal conductivity 0.04 W/mK

Superglue Loctite 382:
Coefficient of thermal conductivity, ASTM C177, 0.11 W.m-1K-1
Glass transition temperature, ASTM E228, ºC 130
(*most epoxies transistion around 50-65 ºC)
(some high temperature epoxies transition 90 ºC and above-specialized)
(glass transition temperature is when material gets soft and
bonds get weak)

http://www.loctite.se/tds/382.pdf


Picked this as a typical epoxy Loctite Hysol 3450:
Coefficient of thermal conductivity, ASTM C177, 0.28 W.m-1K-1
Bond strength at temperature is shown on datasheet chart.
http://www.loctite.at/tds/3450.PDF

Nylon (Polymide):
Thermal conductivity 0.24 W/mK

Pure Water:
Thermal conductivity 0.58 W/mK

Wakefield 120 Thermal Grease:
Thermal Conductivity 0.735 W/mK
http://power.ece.uiuc.edu/Balog/images/How%20to%20keep%20your%20cool%20when%20working%20with%20power%20electronics.pdf

Fused Silicon Dioxide:
Thermal Conductivity 1.4 W/mK

Dow Corning SE4447 CV Thermal RTV:
Thermal Conductivity 2.5 W/mK
http://www.dowcorning.com/DataFiles/090007b58021adab.pdf

Artic Silver Thermal Epoxy:
Thermal conductivity. Greater than 7.5 W/mK
http://www.arcticsilver.com/arctic_silver_thermal_adhesive.htm

Stainless Steel:
Thermal conductivity 15-25 W/mK

Loctite Thermal Adhesive QMI 5030:
Thermal Conductivity, 25 W/mK
http://www.loctite.com/int_henkel/loctite/binarydata/pdf/lt3758a_ThermMgmt.pdf
http://www.loctite.at/tds/QMI5030.pdf

Alumina (Aluminum Oxide):
Thermal Conductivity 30 W/mK

Lead:
Thermal conductivity 34.7 W/mK

Carbon Steel:
Thermal conductivity 43-64 W/mK

Aluminum:
Thermal Conductivity 190 W/mK

Aluminum Nitride:
Thermal Conductivity 260 W/mk


Boron Nitride:
Thermal Conductivity 250-300 W/mk

Copper:
Thermal Conductivity 386 W/mK

Silver:
Thermal Conductivity 406 W/mK

Diamond:
Thermal Conductivity 500- 1000+ W/mK

[/ QUOTE ]
i dont understand does that mean super glue is real bad or good? i need the homer simpson version please /ubbthreads/images/graemlins/smile.gif

o btw i had itoverdrived at 4xaa nimh around 4.8 volts fully charged of charger it was so bright but i used a reflector that was wrong it didnt throw far but lighted up a big area

How about Silicon Carbide (Sandpaper)?

Super Glue is real bad, 'Homer' /ubbthreads/images/graemlins/smile.gif . But that doesn't mean you can't use it - what really counts is the percentage of metal to metal contact. The Super Glue is only occupying the air gaps that exist. Whether those gaps comprise 10% or 90% of the contact area is anyone's guess. But the fact that excellent heat transfer exists with your setup leads me to guess you have a large percentage of metal to metal contact, and that is where the majority of the transfer occurs.

I've used Super Glue on both of my Hotlips builds. The heat transfer is excellent not because of the Super Glue, but because both mating surfaces are flat.

I don't understand the chart, that or I don't trust it...It has been shown time and time again that those lead thermal pads compleatly suck at thermal transfer and that artic silcer is at least 10x better....Acodring to that lsit it appears this is not the case.

Also wth? locktight works 4 times betteer than Artic Silver? For some reason I highly doubt this. It makes no sense what so ever. Why would people not use locktight on their CPU's etc instead of buying expensive thermal past?

Something is wrong or im not understanding the chart, there is no way locktight or lead is better at transfering heat than artic silver!

intoxicated presents a good question here, I have loctite at home but I would have to spend $10 plus shipping to get the artic alumina or artic silver

Gentlemen,

The Loctite Corporation produces many products other than their infamous thread locking compounds. Feel free to peruse their web site .

Dan

This is just another example of where theory collides with actual practice. The Luxeon is super-glued to a heat-sink, and the heatsink gets hot as heck. That tells me all I need to know - superglue works really well.

If this chart claims that diamonds work better than super-glue, who cares? Diamonds are not adhesive.

I don't care wtf locktight makes..There have been numerous studies in the past that show only one uncommon thermal compound to work really well, and thats shin-etsu...How many times were these compounds applied in this chart? just one sample for each reading? A good test would use a minimum of 10 PROPER applications. Superglue as said has little to no thermal properties, I have even read that peanut butter would work better. If only it was adhesive! Well im sure superglue will work, it is just not very good, at least some AS should be mixed in to help a little...or rather some superglue should be mixed in with AS..AA..or other non epoxy to really transfer heat well. if the emitter heats up enough, it will transfer heat to almost anything.

Goodness

Good point. If it heats up enough, it will transfer heat to anything. May not transfer it quick enough with certain things, but still warm up and fool you like it's working.

Lightscene,

Careful there. The loctite under the Luxeon is ??? of which we don't know the properties. I'm sure Lumileds chose it for many reasons. One for sure is the automated machines that dispense them. The other I'm sure is for the thermal properties. That is not ordinary loctite.

One thing to note and has been pointed out above is not necessarily the thermal compound that is used, but, more sigficantly is the actual thickness of the joint. If both surfaces are very flat then you will have a more ideal metal to metal contact which if you don't care about isolation is very good.

loctite works for some of you only for the fact that your joint or surface to surface is probably very thin.

If you blob'd some loctite and sat the Luxeon on top of the blob without squishing it down, I think you would see a completely different result and in this case the thermal efficiency of the goo in this case loctite, would perform very badly.

I've always said that spit could perform better than many adhesives. My only caveot is that the thickness would be less than 0.001" thick or a very thin layer. In that case it would work better than a 0.10" layer of AA or AS probably.

Keep it thin and you could get by with many substitutes. In some cases the slug must be isolated so you could be playing with fire and smoke your precious LED if you do end up with a metal to metal contact. Be aware of what you are doing.

dat2zip, I was referring to the original post in this thread. He used super-glue and it worked. The proof is in the pudding.

Wouldn't it be fair to say that the Arctic Alumina will typically do better than super glue simply because it will fill any small gaps, scrtaches, or imperfections, thus increasing the area of thermal transfer?

On the other hand, it should be OK if the contact area is large enough and the heatsink can absorb or radiate enough heat to keep the temperature below superglue's "Glass transition Temperature" of 130 C.

Daniel