My Nexus L1 VRM temperature has been rising. I under clocked it to get a little relief, but it is concerning. What do you all suggest?
0
Ju/jon-heliumdeployHELIUM DEPLOY· answered 2h after the question
Reseat the heatsink with fresh thermal paste on the ASIC and check that the VRM's heatsink is tight—most VRM heat comes from the regulator working hard when the chip draws current, and if the ASIC sink isn't making good contact the chip runs hotter and pulls even more. Make sure the Nexus has 5 cm of clear space around it and isn't on cloth or in a closed shelf. Underclocking helps short-term but treats the symptom; the real fix is better contact and airflow.
If the VRM still climbs after reseating, post the actual VRM temp from the web UI and the ASIC temp.
Common mistake: assuming the VRM is faulty when it's just working harder because the main chip is poorly cooled and drawing extra current to compensate.
Install some small aluminum heat sinks on the backside of the board around the board window in the back behind the regulators. This board looks like it was built like the ones that get copied. Which the via holes are filled. On the little miner boards it was decided by side by side tests that boards that didn't have the vias filled ran 20% cooler. But the VRM temp can be high as 150c. Full loading is usually around 130c and a lot of designers try to design the miner so it never goes above 100c.
The heatsink on the VRM is tight, I added a fan on the front that is direct on the VRM heatsink, the temps are much cooler. I need to keep an eye on it.