Thanks for the comprehensive testing and the valuable data! I’ve planning to use such TIM pads for lights that are modded frequently to skip the annoying procedure of removing and reapplying the paste every time.
Could you maybe also test liquid metal thermal paste?
It might make sense to use an Osram White flat for testing.
Really have to keep Gallium away from Aluminum though, so in the right circumstance it’s useable but not in all cases. I’ve used it in special lights with good results, not sure really without all the fancy measuring equipment if it’s better or worse than just using a good paste. These days when I have a copper to copper interface I usually reflow the MCPCB to the pill…
One thing to note here is that Arctic MX4 is a very common thermal paste, but definitely not the best one on the market in terms of pure performance. Testing by pc reviewers has revealed that Thermal Grizzly Kryonaut is currently the highest performing one. When used with CPUs is typically achieves around 3°C less compared to cheap pastes. This might be important for someone trying to overdrive an Osram White Flat or similar.
Like I already said, a good paste might be interesting when using LEDs with the very highest power densities like the Osram White Flat. Otherwise, no. I've been using the same tube of Arctic Silver 5 for the past 10 years.
These Panasonic thermal pads here are really a much bigger change. No more cleaning!
The problem with these high performance pastes, from my experience,is they dry out after 6-12months and then become obviously worse than even the cheap pastes.
All those tests show short term performance with fresh paste, but I'd really like to find good paste with long term performance, the one that doesn't have to be replaced with fresh paste each xx months.
I noticed MX-2 and AS5 become dry several months after application in both flashlight and PC. Flashlight OTP triggered at lower than usual body temperature - one of advantages of NTC on LED MCPCB, driver correctly detected OTP threshold level (65C), but because paste was dry body temperature was lower than it should've been.
PC - CPU lags etc., after inspecting paste was also completely dry, this does not happen with generic lower thermal conductivity oil based pastes.
Been googling past few days and Gelid truely wins in almost every review plus it won’t dry out in few months due to viscosity. Looks like i found the winner…for now
But a better option is Phobya 2, thermal epoxy though
Does anyone have experience with MasterGel Maker Nano-High? It’s been favorably reviewed, e.g. Cooler Master MasterGel Maker Nano Review | eTeknix and apparently doesn’t tend to dry up as quickly as other pastes.
I just finished my 2nd tube and so far it’s been excellent. Opened my D4 after almost 18 months to do another emitter swap and the paste wasn’t dry. The problem is it’s sold out at every supplier in SA for the past 3 months. I should have bought more when I had the opportunity.
in our flashlight applications we try to ensure both surfaces are flat and smooth as is practical.
this reduces the dependency on the type of compound used.
then i use the same generic white compound i use when replacing igbt /ipm modules in motor drives.
its silicone oil and zinc oxide based.iirc its cooler master ice fusion.
cheap and does not dry out. some of these ipm’s are huge so low cost is important.
too many lights are built with sloppy surfaces that negate even the best mcpcb or compound.
speaking of mcpcb there is now no reason to use aluminum mcpcb with dielectric layers.
copper dtp serve as a very effective heat spreader with a low thermal resistance from the die to the copper.
makes life easier passing the heat to the lights body.