Interesting link re. dead fan on AMD

[Guest]

<A HREF="http://www.asus.com.tw/Products/Techref/Cpu/amdbdas.html" target="_new">http://www.asus.com.tw/Products/Techref/Cpu/amdbdas.html</A>

Seems that there is some time for monitoring software to react to the death of a heat sink fan if it is installed properly. Not much margin for installation error though :-(


In theory, there is no difference between theory and practice.
In practice, there is.<P ID="edit"><FONT SIZE=-1><EM>Edited by ergeorge on 04/13/01 03:29 PM.</EM></FONT></P>

 

[FUGGER]

I said that weeks ago =)

Thats why I dared any AMD lemming to test their thermal protection. I was hoping that some of you lemmings would try something that stupid.

Anyway. GG on finding that design flaw.

 

[Guest]

You said nothing of the sort.

Did you not bother to look at the link?

I'll summarize for you:

Improperly installed heatsink = gone in 60 seconds = likely to fast for monitoring to catch. So be careful out there.

Properly installed heatsink with dead fan = gone in 1 hour = plenty of time for monitoring to catch and shut down.

BTW: Have you found any references for your claim that a HS fan is only good for about a year?

And finally, yes AMD does need thermal protection. But its hardly the issue you make it out to be.

In theory, there is no difference between theory and practice.
In practice, there is.

BTW: I've removed my speculation on Fuggers motives & literacy in the interest of maintaining a rational & fact oriented discussion. My apologies for my moment of unprofesionalism.<P ID="edit"><FONT SIZE=-1><EM>Edited by ergeorge on 04/12/01 09:57 PM.</EM></FONT></P>

 

[kurokaze]

whats your challenge?

 

[Guest]

Just for reference, here is fuggers claim that HSF failure brings instant death to AMD CPUs:

<A HREF="http://forumz.tomshardware.com/modules.php?name=Forums&file=faq&notfound=1&code=1" target="_new">http://forumz.tomshardware.com/modules.php?name=Forums&file=faq&notfound=1&code=1</A>

This is also the thread where he claims that a heat sink fan has a life expectancy of about 1 year. We haven't heard back on that claim yet either.

And here is the link to some research done by Asus showing that it takes 1 hour to kill an Athlon CPU by stopping the HS fan:

<A HREF="http://www.asus.com.tw/Products/Techref/Cpu/amdbdas.html" target="_new">http://www.asus.com.tw/Products/Techref/Cpu/amdbdas.html</A>

In theory, there is no difference between theory and practice.
In practice, there is.<P ID="edit"><FONT SIZE=-1><EM>Edited by ergeorge on 04/13/01 03:49 PM.</EM></FONT></P>

 

[FUGGER]

george, I have seen claims from 2 seconds ~ 2 minutes. posted on this forum many times.

Grizely pulled a 10 second toaster.

Reguarding the HSF life expectancy. Its a good rule to replace it once a year due to debris clogging the fans reducing the RPM. you may want to clean the fins once in awhile.


Then again, just ignore it and hopefully it goes away.
Go ahead and pull the plug on your HSF and test the mobo for shutdown. I dare you =)

 

[AmdMELTDOWN]

so if your fan dies and you're on a beer run, when you get back and your hahathlon is dead. what do you? what do you?



"Amd cpu...Gone in 2 secs flat, it truly is a fast chip!"

 

[Guest]

The usual...
Someone posts evidence from a reputable source. You post hearsay & conjecture.

>george, I have seen claims from 2 seconds ~ 2 minutes.
>posted on this forum many times.

For an improperly installed heatsink, yes, very likely. Especially if overclocked.

For a dead fan on a properly installed heatsink? Sorry, I don't buy it. Neither does Asus it seems.

Anybody else replace their heat sink fan every year? Cleaning it is a good idea of course.

Actually, you're increasing your risk of failure if you replace an properly operating fan every year. Ever see the bathtub curve for component failure? Your risk of failure is highest the moment you plug it it. That risk decreases rapidly the longer it operates succesfully until it levels out at some non-zero value. Then as the component starts to reach the end of its life expectancy, the failure risk gradually increases. When you plot it over time, it looks like the cross sction of a bathtub.

My guess is that the initial drop in Pf takes a few days, a week or so tops. Given the usual 50,000 MTBF, you're well within the long flat section of the curve @ 1 year. So replacing it at that time actually increases your risk.

And I don't need to pull the plug, Asus did it for me.


In theory, there is no difference between theory and practice.
In practice, there is.

 

[Guest]

>so if your fan dies and you're on a beer run, when you get
>back and your hahathlon is dead. what do you? what do you?

Nope, either your bios detects the fan failure immediately and shuts things down 59 minutes, 59 seconds before your CPU fries, or it detects that your temperature has crossed the predefined limitand shuts things down. If you've set that limit sanely, you have plenty of that hour left.



In theory, there is no difference between theory and practice.
In practice, there is.