Pesquisar na Comunidade
Mostrando resultados para as tags ''throttling''.
Foram encontrados 1 registro
-
Hello everyone, I’ve been lurking on the forum for a while reading through the board repairs and schematic requests, and I finally have a situation on my workbench that I wanted to get some second opinions on. I am currently troubleshooting a high-end enthusiast motherboard (LGA 2066 socket) that is behaving very strangely under load. The board posts fine, boots into the OS, and handles general browsing or light tasks without any errors. However, as soon as I start a sustained rendering workload or a prime95 stress test, the VRM temperatures skyrocket instantly. We are talking about hitting 105°C+ within three minutes, followed immediately by severe CPU throttling. I have gone through the standard hardware diagnostics: Checked for shorts on the 12V rail (clean). Inspected the MOSFETs and driver ICs for physical damage or discoloration (nothing visible). Replaced the stock thermal pads on the VRM heatsinks with high-performance pads. This is incredibly frustrating for me because my background is mostly in enterprise environments. For example, at my day job, I manage a rack of ProLiant DL360 servers running 16 Core 2.0GHz Xeon Gold processors. We run those machines at 80-90% utilization for days at a time. Despite being packed into a tight 1U chassis, the power delivery systems on those boards never flinch. The thermal management is just on a completely different level compared to this "enthusiast" gear. On this specific repair, I am starting to suspect that maybe the capacitors in the filtering stage are degrading and failing to maintain capacitance once they heat up, causing the VRMs to work overtime to compensate. I hooked up an oscilloscope to the VCore phases, and the ripple does seem to get worse as the temperature climbs, but I am not sure if that is the cause or the symptom. Has anyone here experienced similar behavior where a board works perfectly cold but fails thermally under load despite fresh thermal interface material? Is it worth trying to recap the VRM area, or is this likely a design flaw in the PCB layers not dissipating heat effectively? I hate to scrap a board that posts and boots, but I can't trust it for any real work right now. Thanks for any insights!
SOBRE O ELETRÔNICABR
Técnico sem o EletrônicaBR não é um técnico completo! Leia Mais...