What are the maximum temperatures for Intel and AMD proces..

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

Hello

I'm going to build microprocessor driven speed regulator for CPU fan. And i
don't know the maximum temperature that CPU can tolerate. Does anyone know?
greetings

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

szlovak wrote:

> Hello
>
> I'm going to build microprocessor driven speed regulator for CPU fan. And i
> don't know the maximum temperature that CPU can tolerate. Does anyone know?
> greetings

You might get some good answers also, from one of the hardware groups, like
comp.sys.ibm.pc.hardware.chips.

This is probably minimal help here:

http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/25325.pdf

It's an AMD whitepaper -- Athlon MP thermal testing results. The methodology
section has a couple pictures, and a little description of how they made
measurements.

Bryan

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

Bryan Hoover napisa³(a):

> szlovak wrote:
>
> > Hello
> >
> > I'm going to build microprocessor driven speed regulator for CPU fan. And i
> > don't know the maximum temperature that CPU can tolerate. Does anyone know?
> > greetings
>
> You might get some good answers also, from one of the hardware groups, like
> comp.sys.ibm.pc.hardware.chips.
>
> This is probably minimal help here:
>
>
> http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/25325.
> pdf
>
> It's an AMD whitepaper -- Athlon MP thermal testing results. The methodology
> section has a couple pictures, and a little description of how they made
> measurements.
>
> Bryan

thanks

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

szlovak wrote:
> I'm going to build microprocessor driven speed regulator for CPU fan.
> And i don't know the maximum temperature that CPU can tolerate. Does
> anyone know?

How are you going to be measuring the temperature? For AMD CPUs (socket-A
and K8), the maximum CORE temperature varies between 100 deg C and 70 deg C
depending on what the CPU is. Since it's very unlikely you're going to be
measuring the core temperature, you have to aim significantly lower than
this. If you give a few more details on how you're planning to measure it I
can probably give a more accurate answer.

--
Michael Brown
www.emboss.co.nz : OOS/RSI software and more
Add michael@ to emboss.co.nz ---+--- My inbox is always open

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

Michael Brown napisa³(a):

> szlovak wrote:
> > I'm going to build microprocessor driven speed regulator for CPU fan.
> > And i don't know the maximum temperature that CPU can tolerate. Does
> > anyone know?
>
> How are you going to be measuring the temperature? For AMD CPUs (socket-A
> and K8), the maximum CORE temperature varies between 100 deg C and 70 deg C
> depending on what the CPU is. Since it's very unlikely you're going to be
> measuring the core temperature, you have to aim significantly lower than
> this. If you give a few more details on how you're planning to measure it I
> can probably give a more accurate answer.

I'm going to do this with Microchip microcontroller (PIC12F or C) and with
termoresistor or diode (as the sensor) with other parts etc. Temperature sensor
will bo placed on the CPU radiator

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

szlovak wrote:
>> szlovak wrote:
>>> I'm going to build microprocessor driven speed regulator for CPU
>>> fan.
[...]
> Temperature sensor will bo placed on the CPU radiator.

CPU radiator = heatsink I presume? Whereabouts on the heatsink?
Thermal-taped onto the underside of it close to the die, or attached to one
of the outer edges? Assuming you're cooling a K7, then if you're on the
underside close to the die, and using a copper-based heatsink, you could be
looking at getting up to 50 deg C under full-load conditions (depending on
your heatsink). If you're on the outer edge of the heatsink then you're
quite possibly looking at 35 to 40 deg C, again heatsink (and airflow)
depending.

The best way to do it is to hook everything up and see what the
microcontroller reads at full load (ie: BurnK7 left running for half an hour
before measurement) with the sides of the case on, etc. Now make the
microcontroller regulate the fan to keep the temperature below that. One
thing you'll have to be a bit careful of is that the thermistor is not
affected by airflow through the case. For example, having the thermistor
poorly attached on the side of the heatsink will give "bad" readings as it
will be cooled by the airflow off the heatsink, and the heatsink will be
much hotter than what is being read.

--
Michael Brown
www.emboss.co.nz : OOS/RSI software and more
Add michael@ to emboss.co.nz ---+--- My inbox is always open

 

[Guest]

Archived from groups: alt.comp.hardware.overclocking (More info?)

Michael Brown napisa³(a):

> szlovak wrote:
> >> szlovak wrote:
> >>> I'm going to build microprocessor driven speed regulator for CPU
> >>> fan.
> [...]
> > Temperature sensor will bo placed on the CPU radiator.
>
> CPU radiator = heatsink I presume? Whereabouts on the heatsink?
> Thermal-taped onto the underside of it close to the die, or attached to one
> of the outer edges? Assuming you're cooling a K7, then if you're on the
> underside close to the die, and using a copper-based heatsink, you could be
> looking at getting up to 50 deg C under full-load conditions (depending on
> your heatsink). If you're on the outer edge of the heatsink then you're
> quite possibly looking at 35 to 40 deg C, again heatsink (and airflow)
> depending.
>
> The best way to do it is to hook everything up and see what the
> microcontroller reads at full load (ie: BurnK7 left running for half an hour
> before measurement) with the sides of the case on, etc. Now make the
> microcontroller regulate the fan to keep the temperature below that. One
> thing you'll have to be a bit careful of is that the thermistor is not
> affected by airflow through the case. For example, having the thermistor
> poorly attached on the side of the heatsink will give "bad" readings as it
> will be cooled by the airflow off the heatsink, and the heatsink will be
> much hotter than what is being read.

Talking about temperature I meant that I will monitore it on th heatsink , yes
, but I would rather measure it by the mainboard and then see differences on
heatsink . For test i will use overclocked celeron 566@850 MHz. This speed
regulator will also work well under normal conditions, without overclocking, to
silent the CPU fan. So now, knowing that there is other temperatures on
heatsink and mainboard I will only take under attention this from mainboard. I
don't know for now how to mount Resistor or other sensor proffesionaly, but
surely on the not ventilated side of heatsink. Is the temperature showed by
mainboard the core temperature with good approximation? If so , I will not
overcross 70 degrees, because my regulator will be universal, highest
temperature will cause fan to run full speed