EVGA VRM Follow-ups
Added 2016-11-10 15:17:49 +0000 UTCWe have two follow-up articles planned for the EVGA VRM fracas regarding the 10-series cards. The first is this one, the second will be post-testing.
It was difficult trying to figure out the best way to collect direct VRM data without interfering with the components' abilities to transfer heat to the thermal pad, and then to the heatsink. Consulting thermal engineers has given us the right path: We're using a new set of flat thermocouples with self-adhesive 'tape' of sorts, and which can sustain high temperatures output by the VRM. This resolves both concerns of testing that we ran into with our existing thermocouples, primarily electrical conductivity and thermal conductivity.
Pretty cool stuff. It'll take about a week to really dig in, but we should have a final answer once we're done.
As for the inside look: I'm hoping to apply these methods to a few of the top heatsink designs in the future. I don't know that we'll get more than one use out of each of these probes, but one per device wouldn't be too terribly expensive (they're a bit more than $10 each). It's more of a time concern. I'm thinking we'll run these tests on the Twin Frozr, ACX, and maybe Windforce coolers for this generation. It's just a matter of getting more cooler types in.
http://www.gamersnexus.net/guides/2676-evga-vrm-test-planning-new-thermocouples
