And as you said you are not confident enough to go into the box, it might be as simple as clean radiator itself.
This bridge usually running at higher temp that others, I would start be really worried above 100C, but to be fair 80ish is sort of normal.
Grab yourself a dust cleaning air pressured can, and some heat-sink paste from maplin.
That is very low speed of fan. And this is how it will be at 83C.
If you have any manual fan settings try boost it’s speed. As well I believe you may as well use some Greenpeace bullshit power conservation settings
Also have you tried re-setting the power management settings?
I got called out a couple weeks back to a client who thought he had a faulty-out-of-the-box Mac Pro, because the fan in the PSU was spinning up and humming like mad as soon as he logged in… After talking him through loads of troubleshooting over the phone I eventually had to go down there…
When I got there, I opened System Preferences, saw a third party application had been installed called “Fan Speed Control”… I’m sure you can guess the rest :crazy:
I’m sure you do, but just in case you don’t then I’m going to say it: backup backup backup. Please. Backup your data. ASAP. Regularly. If you don’t know how to do it, I can help you set it up, just let me know. This is something that is much better done pro-actively.
In the last couple of years, I have had folk come to me with a “dead computer” which has the only copy of their: “entire masters’ thesis”, “all accounts and invoices for the last year”, and “all photos since I started shooting digital in 2000”. The first two I was able to rescue all of the data, the last one I had to send off to a professional data rescue company who were able to retrieve 90% of the files. In all three of these cases, I had previously told these people to back up all of their data regularly, and they didn’t, so I ended up spending a lot of time making up for it.
I’m not saying necessarily that the chipset overheating is going to damage your data, but it’s not outside the realms of possibility. Suppose the overheating affects the disk controller which starts writing data to random sectors of the disk instead of where it’s meant to? Or it starts spinning the disk constantly at high speed, leading to it dramatically exceeding its expected usage cycle and wearing out early? Or…well, anything. I work with large enterprise-scale computing systems, and I spend a lot of time focussing on the non-functional aspects (security, availability, reliability, disaster recovery, etc). Some of the stories I could tell of “wow we never thought that could ever possibly happen but it did” would absolutely astound you…