You may have heard that I've contracted an aluminum case to be made. The prototype case panels came in recently and I've been putting them through their paces making sure that thermal performance doesn't negatively impact FPGA cores.
Bottom line: I've not found any issues so far. The cooling performance of the aluminum case is between an actively cooled 40mm fan + heatsink combo, and a passively cooled heatsink slapped on the Cyclone V.
Notice how the heat is distributed through the aluminum case?
Before I show you the data, let's talk about how to interpret results. The max temp of the Cyclone V variant used on the DE10-nano is 100 C. That's the internal / core temperature. I'm measuring the heatsink from the front of the case looking in. I searched around on the heatsink to find the hottest spot. Do note that the temperature looking straight down (bird's eye view) is about 5 C higher.
That I know of, there is no way to correlate core temperature and heatsink temperature. We can assume that the core temp isn't twice as high as the heatsink temperature is, but it's a baseless assumption.
Right, on to the data:
In all cases, the external temperature is less than 1/2 the max rated internal temperature. I've let Doom run overnight without any issues, as well as ran memtest overnight with 0 errors.
At this point, I'm prepared to place an order with the design house. This is not a guarantee that there will never be an issue with thermal performance, but I'm fairly happy with the results.
My plan is to wait until I have the cases in hand before listing them for sale. The world is a crazy place and I don't want to hold on to your money before I'm certain the cases will arrive.
Leave a comment below or on Twitter if you'd like to know more.