Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
That's if they were relying entirely on convection, I think FrankHahn's point is still valid as the fan only has to make up for what the natural heat rising can't do.

In building my storage enclosure I've tried to mimic a similar principle by keeping the PSU at the bottom...

But, since the PSU exhausts air - doesn't that mean it's pulling cool air out of the bottom of the case, instead of pulling warm air out if it were at the top?

I like that the 120mm fan in my PSU at the top of my case is pulling warm air out of the case.
 
But, since the PSU exhausts air - doesn't that mean it's pulling cool air out of the bottom of the case, instead of pulling warm air out if it were at the top?

I like that the 120mm fan in my PSU at the top of my case is pulling warm air out of the case.

just an fyi on the nmp psu location.. it's up top and pretty much the closest thing with the least obstructions to the fan itself.
 
there's some talk sprinkled throughout various different threads over the past few months..

one of the things which should definitely be brought into focus if the topic is thermodynamics of the nmp is the secondary piece of material in between the cpu/gpu processors and the heat sink.. there's more going on than "oh, it's just a normal heatsink with a different shape"..

Image

Okay, so they're using a heat spreader too.


...it appears to be a graphite heat spreader…

It appears to be a heat spreader, sure - - but not necessarily one that's made of TPG (Thermal Pyrolytic Graphite), as that material is good for up to 200W/cm^2…much more than they need, particularly if its thermal dimensionality behaviors align nicely to the design needs.

They might be using a silver foil TIM in the design too…but that's an even more off-the-shelf product these days than TPG is, and it has been used for awhile in high density electronics, especially for MILSPEC and/or NASA rated stuff. Given that TIM's are cheaper than TPG, its more likely too IMO, but the thermal pastes are cheaper still overall, so I'd not count them out quite yet.

-hh
 
just an fyi on the nmp psu location.. it's up top and pretty much the closest thing with the least obstructions to the fan itself.

Huh?

The PSU is on the side between the DIMMS and between the CPU and the IO port panel.

And totally irrelevant to a post talking about standard PSU with internal exhaust fans.
 
Huh?

The PSU is on the side between the DIMMS and between the CPU and the IO port panel.

And totally irrelevant to a post talking about standard PSU with internal exhaust fans.

i meant 'up top' as in higher than the other components.. it's still below the fan.. it's the only thing that really has a direct line of sight to the fan (not counting the actual heat sink of course)

i don't see why it's irrelevant but.. if you say so.

----------

but the thermal pastes are cheaper still overall, so I'd not count them out quite yet.

i think you first have to see the gpus/cpu as being (user) serviceable prior to any thermal paste talk.. or- maybe for the sake of this particular conversation, just pretend like it is.. :)

a user (and not just a user-- a service tech as well) would be required to remove/apply thermal paste in 3 different locations all of which are on vertical surfaces(unless those heat spreaders come off as well.. seems more hassle than it should be though)..
if the idea was to make it serviceable then i definitely think they'd try to eliminate paste from the equation.
 
Last edited:
...
i think you first have to see the gpus/cpu as being (user) serviceable prior to any thermal paste talk.. or- maybe for the sake of this particular conversation, just pretend like it is.. :)

a user (and not just a user-- a service tech as well) would be required to remove/apply thermal paste in 3 different locations all of which are on vertical surfaces(unless those heat spreaders come off as well.. seems more hassle than it should be though)..
if the idea was to make it serviceable than i definitely think they'd try to eliminate paste from the equation.

A fair point. I was looking at it mostly from the perspective of the touch labor for the initial fabrication (assembly line)…this is simply just the old "each additional manufacturing step costs money".

From this perspective, having no TIM (foil or paste) required by the design is the cheapest. Next up is paste, because it is generally has sloppier tolerances than foil. This leaves foil as both the best as well as the most expensive (typically).

And to acknowledge that there's going to be a serviceability requirement, these can all at least be dis/re-assembled (as opposed to the epoxies).


-hh
 
The world is too small and life is too short to hate. You could say I don't appreciate the new design as much as you do might, but calling me a hater is like calling Nelson Mandela or Siddhartha Gautama a hater. :)

You know that the term "hater" in general is not used in a derogatory way on these forums only to generalise oneside of the two camps in these "debates" and you've been on here long enough to know that. Though I wouldn't say my use of it wasn't exactly a term of endearment, apologies for that........;)

Interesting examples in your reply........NM was no saint in his early life whilst founding and active in the ANC......but have to agree that he definitely transcended that in later life. And Buddha, a real? transcendent! Thank you for not using Mother Theresa, I would have felt properly embarrassed!
 
But, since the PSU exhausts air - doesn't that mean it's pulling cool air out of the bottom of the case, instead of pulling warm air out if it were at the top?
Ah, no, I opted for a fanless PSU, again mimicking the new Mac Pro somewhat. It's oriented so that some of the air pulled into the case comes through the PSU's "top" (I actually have it aimed vertically, meaning the case needed feet so I can fit the mains cable in).
The exhaust fan I'm using is great, a Prolimatech Ultra Sleek Vortex 14, which is a 140mm fan with only 15mm depth (fits better into my case). When I saw it online I thought it was too good to be true but it really is silent at its lower speeds, and still shifts a lot of air.

It's interesting though as I'll need to see how it compares to the Mac Pro fan, which looks like it may be closer to 150mm, but a lot taller, plus it has a more vertical fin arrangement. Though that may be because the fan isn't fully exposed, but it will need to push a lot of air to keep the machine cool.
 
for heat transfer, you want turbulent flow. this talk of laminar flow across the core is mind-boggling...

also, according to the tech specs: 12 dBA idle.
 
for heat transfer, you want turbulent flow. this talk of laminar flow across the core is mind-boggling...

also, according to the tech specs: 12 dBA idle.

That's funny. The first person to quote correct facts seems to have ended the discussion. :D

Or as Dr. Feynman said: "For a successful technology, reality must take precedence over public relations, for Nature cannot be fooled."
 
That's funny. The first person to quote correct facts seems to have ended the discussion. :D

Or as Dr. Feynman said: "For a successful technology, reality must take precedence over public relations, for Nature cannot be fooled."

There's also the peak dba and power consumption, and this still being speculation is only really getting going to get started after the tear down and measurements - the facts.

The master of explaining complicated stuff simple such as his diagrams would appreciate a design that does the same in reality :D
 
I hope they can keep the entire heatsink under 70c since that is the TJMax of xeon CPU's. I believe it was a bad idea mixing two different compenents with different TJMax temps all on the same heatsink. The Firepros are 100c and the Xeon is 70c. If the firepros start to bring the temps up to 70c then the CPU is going to start to throttle even if it's not under load. I am very curious to see how well Apple did with keeping the beast cool and quiet. (I am assuming it the FirePros had a 100c limit as most GPU's do..)
 
I hope they can keep the entire heatsink under 70c since that is the TJMax of xeon CPU's. I believe it was a bad idea mixing two different compenents with different TJMax temps all on the same heatsink. The Firepros are 100c and the Xeon is 70c. If the firepros start to bring the temps up to 70c then the CPU is going to start to throttle even if it's not under load. I am very curious to see how well Apple did with keeping the beast cool and quiet. (I am assuming it the FirePros had a 100c limit as most GPU's do..)

The CPU has twice the number of fins compared to each GPU. That should help.
 
for heat transfer, you want turbulent flow. this talk of laminar flow across the core is mind-boggling...

Thanks. Makes sense, especially seeing pictures of the nMBP core. For some reason, I had pictured it hollow (which would be a terrible waste of air mass).

I feel like I have a good grounding in the basics of the involved physics, but surely didn't think this one through well. Like I said, not mechanical eng (s/w and electronics though, but never had to design an individual heatsink, let alone a complete system).
 
Thanks. Makes sense, especially seeing pictures of the nMBP core. For some reason, I had pictured it hollow (which would be a terrible waste of air mass).

I feel like I have a good grounding in the basics of the involved physics, but surely didn't think this one through well. Like I said, not mechanical eng (s/w and electronics though, but never had to design an individual heatsink, let alone a complete system).

My physics grounding is pretty far away from this kind of thing (nuclear) but thermodynamics is universal. I always expected heat pipes and thermal paste as other technologies aren't mature enough yet for an Apple 'Genius' to apply properly.
 
The CPU has twice the number of fins compared to each GPU. That should help.

You're right but think of it like this. The heatsink it one solid chunk of metal, there is still going to be heat build up that may eventually rise up to 70c and beyond even with enough air flow.
 
You're right but think of it like this. The heatsink it one solid chunk of metal, there is still going to be heat build up that may eventually rise up to 70c and beyond even with enough air flow.

i think the idea is to spread out the heat prior to it even hitting the heat sink..

so instead of having 70º hot spots hitting the heat sink directly, you'll have a larger area of 35º temps hitting it more evenly.
 
i think the idea is to spread out the heat prior to it even hitting the heat sink..

so instead of having 70º hot spots hitting the heat sink directly, you'll have a larger area of 35º temps hitting it more evenly.

That seems reasonable. althought 35 does seems low I get the point ;)
 
Something interesting from the reviews:

Both Macworld and FCP.co mentioned a gentle but notable current of warm air rising from the machine, but that the Mac Pro's fan was barely noticeable. If something is placed on top of the machine's vent hole, like a book, the Mac Pro will shut itself off before any damage can occur.
 
Something interesting from the reviews:

Probably not an airflow sensor but the temps will rise pretty rapidly if you cover the top. Would be interesting to blow some dry ice through the bottom to detect the airflow pattern. I would use a big fat Havana but I guess these sites couldn't do that without upsetting somebody reading it!
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.