get Northbridge temp down to 120 F ( 49c ) direct the airflow over the Northbridge Heatsink
All internal fans, minimum to no noise 32deg C ambient (Florida):
View attachment 826866
View attachment 826867
View attachment 826868
That is true but physically the fans can only blow in the same direction. So they are generally cooling that area. So whatever lies in that area (designed by Apple) will be cooled as well. What am I doing is: I take the highest temperature and base some sensors on that temperature. If you think about, it it works.Thanks for the sharing, however, I personally don't think any fan base on the heatsink temperature is a good idea.
If the heatsink is detached. The heatsink temperature will be low, so the fan won't spin up. But the diode temperature will be very high, and your setting will give it less fan to cool down.
That is true but physically the fans can only blow in the same direction. So they are generally cooling that area. So whatever lies in that area (designed by Apple) will be cooled as well. What am I doing is: I take the highest temperature and base some sensors on that temperature. If you think about, it it works.
Ok you say base it on the diode not the heatsink. I will change that.NO, I mean when the rivet breaks. That's the most critical situation. Your fan profile won't help, but make it worst.
MacsFanControl using "target fan speed mode" to control the fans, which will completely override ALL system protection. No matter how hot the NB chip, if the NB heatsink is cool, your fan will NOT spin up.
I give you an example.
1) under normal situation. NB 55C, NB heatsink 48C
According to your setting.
Boost A ~3000RPM, Intake and Exhaust ~1260RPM
2) NB rivet broken. NB 130C, NB heatsink 35C
According to your setting.
Boosta A 5200RPM, Intake and Exhaust 600RPM (no extra cooling provide by Intake and Exhaust fans even though the NB is clearing overheating)
3) NB rivet broken, NB 130C, NB heatsink 35C
All fan base on NB temperature, min 50, max 60
Boosta A 5200RPM, Intake and Exhaust 2800RPM (all fans work at max speed and hopefully can safe the NB)
IMO, it's better to base on the actual point that you want to protect. You want to protect the NB, not the NB heatsink. Make the fans base on NB heatsink is NOT fail safe.
How much is the price of the fan in comparison to the price of the baseboard? And the inconvenience of replacing one?@startergo,
Is it recommended to run the fan at such high speed?
Your IOH and CPU temps are definitely impressive, however the Fans (Intake, Exhaust and BOOSTA) are running at much higher speed (2X or more than designed idle, if I'm not mistaken) which will definitely shorten the lifespan of the fan itself.
To my knowledge, for many people (including me) the temp @ 75C idle is expected.
So if 75C is considered 'safe', aiming for below 60C would be kind of overkill (un-required, though the NB chip will be very happy ).
I experiment using iStat to cool the NB chip a little bit by increasing both intake and exhaust to ~630 and it did reduce by 2-3C (become: 73-72C) while increasing to 700rpm will lower the temp further by 3-4C to 70-71C. I settle to use the 630rpm as with 700rpm the boosters (A & B) will increase to 1200rpm, while with 630, the boosters stays at 1170.
I'm aiming for a balance approach to get longer life span of the NB without compromising too much on other components of the system unnecessarily.
Thanks
@startergo,
Is it recommended to run the fan at such high speed?
Your IOH and CPU temps are definitely impressive, however the Fans (Intake, Exhaust and BOOSTA) are running at much higher speed (2X or more than designed idle, if I'm not mistaken) which will definitely shorten the lifespan of the fan itself.
To my knowledge, for many people (including me) the temp @ 75C idle is expected.
So if 75C is considered 'safe', aiming for below 60C would be kind of overkill (un-required, though the NB chip will be very happy ).
I experiment using iStat to cool the NB chip a little bit by increasing both intake and exhaust to ~630 and it did reduce by 2-3C (become: 73-72C) while increasing to 700rpm will lower the temp further by 3-4C to 70-71C. I settle to use the 630rpm as with 700rpm the boosters (A & B) will increase to 1200rpm, while with 630, the boosters stays at 1170.
I'm aiming for a balance approach to get longer life span of the NB without compromising too much on other components of the system unnecessarily.
Thanks
Don't set the Exhaust, but only intake. Then the Booster won't have higher idle RPM.
There is a specific minimum ratio between Exhaust fan and Booster fan. iStat use "minimum fan speed" to control the fans, which will not manually override this setting (for info, MacsFanControl use "target fan speed", which will override this setting). When you increase the Exhaust RPM, the Booster RPM will also be increased.
And IMO, it's better to keep positive pressure inside the case anyway (high intake than exhaust), otherwise, the dust will try to leak into the case via all the gaps. When intake > Exhaust, positive pressure will push the air out of the case via all the gaps, which also better for graphic card cooling (especially if the graphic card is open fan type).
I believe this is the reason why Apple make the Exhaust fan is linked to all Intake and Booster fans. So, that when Exhaust spin up, Intake + Booster will also spin up, which make sure always has positive pressure inside the case.
Found some from aliexpress:My Dual Quad 2,93 Ghz cMP 4,1>5,1 has been ON 24 hrs since yesterday and with a slight adjustment to the USB fan position the Northbridge has maintained an average of 54~55°C = 129°F.
I can't stress enough the importance of having space behind the Mac Pro to allow expelled case air to dissipate.
I'm more than happy with this for the present. Summer of course will be a different kettle of fish.
Does anyone have any suggestions for a slim ( 8 ~ 10mm thickness) x 80mm USB or 'any' fan with at least 20 CFM & preferably brushless with similar dimensions ? I'm pressed for Googling time this week.