How does it compare to Arctic Silver Adhesive? Honestly - I wish I could say for sure. I canāt find (and I JUST looked again) any meaningful datasheet on the stuff. http://www.arcticsilver.com/arctic_silver_thermal_adhesive.htm
I trust wakefield - theyāve been in the Thermal management game for a LONG time, and theyāre well respected in applications way more important than overclocking a gaming computer. My opinion of Arctic silver has never been that high - Itās more of a marketing shill than anything. Additionally - Arctic silver fully admits that their material is quite electrically capacitive - thatās bad for some types of electronics (including buck/boost drivers). Kids overclocking their PC use Arctic Silver, Engineers and Scientists using Wakefieldā¦ Thereās probably a reason
Total respect for your work. Iām a little sad that you donāt want to commercialize these (youād be right in line, price-wise with Georgeās Taskled drivers, fwiw), but thereās a lot to be said for not taking on an outlandish number of projects.
That clarifies things a bit Happy to take the time - Iām glad you enjoy reading my ramblings.
P60 hosts donāt have much thermal mass - when running an XM-L at 3 Amps, thats (3 * battery voltage) Watts of heat (at 4.0V, 12 Watts) that gets generated between the emitter and the linear driver. A P60 canāt handle that for long without getting uncomfortably hot. 50C thermal-cut off will almost certainly trigger a step-down after some time of being on.
The 7135 has an on/off input that allows it to be turned on and off. The 7135 is perfectly comfortable being controlled even by high(ish)speed PWM so itās designed to tolerate switching on and off frequently and rapidly. Because of the way these chips operate, they donāt āSeeā the current of other chips in parallel Electrically, they have no idea that theyāre not all by themselves which means that adjacent chips have very little effect on each other.
One thing to be aware of with 7135 type chipsā¦ if you try and switch them off via the Vdd pin and an LED is connected, the circuit can still draw a couple of milliamps! The chip is specād with a standby current of 100 uA, but that is apparently with no load path connected. For the lowest possible standby current, you want to PWM/enable via an external FET.
Havenāt gotten it fired up yet (no time to write micro code), but if thatās true, that would suck. A lot. On a 17mm board, there just isnāt room for much more than I already have!