2020 Air Heatsink Modification Thread

[frou]

Everything is written down, somewhere deep in the guts of Apple. Is there a design document somewhere saying: We're going put an intentionally ineffective cooling in this thing, and here's why.

Seems a bizarre concept.

 

[Erehy Dobon]

No, there's a design document that says "let's see how many geeks who can't see the forest for the trees we can infuriate after we've determined that a case cooling fan is adequate for this model's intended target audience's usage."

I think Apple has done an admirable job with the MacBook Air 2020.

 

[gim]

No, there's a design document that says "let's see how many geeks who can't see the forest for the trees we can infuriate after we've determined that a case cooling fan is adequate for this model's intended target audience's usage."

I think Apple has done an admirable job with the MacBook Air 2020.

The cooling solution is not adequate because this machine throttles massively and the fan noise will be as obnoxious as in the previous version. You don’t have to be a geek to hear that fan noise and feel the high surface temperatures on your thighs. But I’m afraid Apple apologists like you won’t listen to reason. So why don’t you stop typing and leave this discussion to people, who are actually able to critically look at products before they buy them?

 

[PianoPro]

It's not Apple's goal to design the highest performance MBA they can. They don't want to convert higher paying MBP users to the MBA. They want differentiation between those products. Their margins and revenue are both hurt if MBP users can convert to an MBA and be satisfied with performance demanding apps. And more performance would do virtually nothing to increase MBA sales versus PC notebooks. MBA's are sold vs PC alternatives based on Mac OS being available at the right price premium. Not based on out-performing a similar priced PC notebook.

So the thermal throttling issue works to Apple's advantage. The MBA MacOS UI is generally as responsive as a similar clock rate MBP, which is a key for its users, and for PC competitive reasons. But throttling protects MBP sales. i.e. they ain't fixing it because it isn't broke, it's what they want.

 

[ilikewhey]

It's not Apple's goal to design the highest performance MBA they can. They don't want to convert higher paying MBP users to the MBA. They want differentiation between those products. Their margins and revenue are both hurt if MBP users can convert to an MBA and be satisfied with performance demanding apps. And more performance would do virtually nothing to increase MBA sales versus PC notebooks. MBA's are sold vs PC alternatives based on Mac OS being available at the right price premium. Not based on out-performing a similar priced PC notebook.

So the thermal throttling issue works to Apple's advantage. The MBA MacOS UI is generally as responsive as a similar clock rate MBP, which is a key for its users, and for PC competitive reasons. But throttling protects MBP sales. i.e. they ain't fixing it because it isn't broke, it's what they want.

fair enough, i think the one nuances the majority of us have here is the previous air model had a heatpipe, but yet apple decided to ditch it. if the previous air model started with passive cooling solution and the air never saw introduction of a heatpipe. this wouldn't be as big of a talking point. therefore in the eyes of many, this passive cooling seen like a step back from innovation.

 

[Mopar]

Have you guys actually looked at the way the airflow works in a MacBook Air? Have a look at these two photos.

This first image shows the air vents inside the case. There is an inlet at the top right-handside (black slot leading to five square inlet holes that suck air in over the motherboard). The air then moves over the board and is funnelled over the heatsink, where it ducts warm air away to the centrifugal fan on the left. The fan sucks this air in and blows it out of the exhaust slot at the top left-hand side. Pay particular attention to the heatsink and note the air dams between the heatsink and battery.

Now have a look at the removed backplate in this image. Note the shape of the air dams that run along the inside, creating a narrow funnel right where the heatsink is. This means any air that is sucked in through the back of the machine is drawn directly over the heatsink by the fan – in effect, the backplate and air dams form a heat duct that draws all the incoming air over the heatsink with nowhere else for it to go. Have a look at the square pad on the backplate, right in the middle near the leading edge: it is the same shape as the heatsink and sits over the fins, narrowing the air at that point so that it can only be drawn through the fins and over the sink (from left to right) by the fan on the right as it exhausts the warm air out the back of the machine.

I don't know where anyone got the idea that dirty air is sucked in through the keyboard (as I have read in other threads), or that the MacBook Air is a "passive" design – because it is not. It has a defacto heatpipe (OK, heat duct to avoid confusion) formed by the air dams and backplate of the machine. There is literally nowhere else for the incoming cool air to go but directly over the sink and through the fins.

This is not a bad design. I'm not going to comment on the video circulating of the 2020 machine idling at 70C because I don't know the context (had it just come off a CPU/Graphics stress test? Was the OS still indexing? Etc). Nor am I going to comment on the thermal properties of the old 2C vs 4C CPU until I have one of these machines in my own hands. But it is likely going to be swings and roundabouts: double the processing power and heat/halve the amount of processing time and heat generated.

Guys, just because some people don't understand how something works – and seem to think it's a "deal-breaker" – doesn't mean it is inferior. The proof is in the pudding: does the machine get hot? Are lots of MacBook Air owners complaining about excessive heat or throttling or fan noise?

Well, are they?
[automerge]1584860885[/automerge]

It's a case fan, not a CPU fan. It's in between the discontinued fanless MacBook and the MacBook Pro.

It appears that the system management chip is programmed to keep the system below 100 °C. It will simultaneously run the fan at the maximum speed while monitoring the CPU frequency so it stays below this temperature ceiling, presumably the thermal design maximum for this particular model.

You guys do know that Apple doesn't let random schmucks walking by verify the thermal designs? And they aren't snot nosed 22-year-olds with a B.A. in Communications either.

The people who do this mostly have Ph.D.s in Mechanical Engineering specializing in Computational Fluid Dynamics. I bet the average age of the senior engineers who do this sort of work is 50.

Moreover, Apple tests many prototype designs before selecting on a particular one for production. Yes, they probably tested a few designs that were fanless, a few with heat pipes, maybe even a few with CPU fans. There are probably units with AMD CPUs, ones with ARM CPUs too.

Every design choice is a compromise: power, weight, bulk, cost, noise, convenience, battery life, etc.

The point of a MacBook Air isn't to harness the maximum capability of the CPU 100% 24x7. Maybe they could add a heat pipe and a CPU fan but that would force an increase in the case thickness. Maybe you need a bigger battery to handle the power needs. All of sudden you end up with a MacBook Pro.

It's not like Tim Cook said one day, "Hey, let's hold a design contest for the interns and ship the one that gets the most employee votes!"

You're the only poster here who seems to have noticed it is a case fan and understands how it is supposed to work.

FYI, the MacBook Pro is a completely different design inside. It has an open layout with dedicated CPU heat pipes because it has no air channels, ducts or dams funnelling air directly over a finned heatsink. They are chalk and cheese when it comes to thermal design.

 

[UBS28]

Yeah, I was actually going to point to that Appleinsider video. According to it, the new MacBook Air doesn’t experience a temperature drop even with the fans at maximum speed (8.000 rpm), so that’s all indicative of a lack of heatpipe.

I’d like to be wrong but, looking at the throttling at temperatures (Constant 100ºC), most probably it is passively cooled.

Seems I will not be getting a 2020 Macbook Air then. No heatpip, a basic component of a computer, no money.
[automerge]1584865292[/automerge]

It's not Apple's goal to design the highest performance MBA they can. They don't want to convert higher paying MBP users to the MBA. They want differentiation between those products. Their margins and revenue are both hurt if MBP users can convert to an MBA and be satisfied with performance demanding apps. And more performance would do virtually nothing to increase MBA sales versus PC notebooks. MBA's are sold vs PC alternatives based on Mac OS being available at the right price premium. Not based on out-performing a similar priced PC notebook.

So the thermal throttling issue works to Apple's advantage. The MBA MacOS UI is generally as responsive as a similar clock rate MBP, which is a key for its users, and for PC competitive reasons. But throttling protects MBP sales. i.e. they ain't fixing it because it isn't broke, it's what they want.

And I don’t want a computer who runs hot by doing easy things. It is bad for the life expectancy of the computer, comfort while using it and performance.

Seems it is a better idea to wait for the 2020 13” Macbook Pro for an ultra portable laptop In this case.

 

[JTToft]

Have you guys actually looked at the way the airflow works in a MacBook Air? Have a look at these two photos.

- Nice post and analysis. I'm with you about the fan blowing air out on the left, but I don't quite see the mechanics of how it also sucks air in on the right. The fan has an opening towards the rear of the machine that allows air to be blown out but it doesn't have a second opening that I can see on its right side that would allow it to suck in air. Can you explain?

 

[joptimus]

Has anyone heard of a heatpipe mod for the MBA? Sure, opening it would likely void warranty, but for some tech savvy people who have enough time on their hands...?

 

[Populus]

The only problem is the space. If there’s enough space it would be great, but for it to become a reality it would need to become a popular mod, something a popular YouTube channel like Snazzy Labs could spread among the community. Actually, I think I’m going to write Snazzy Labs to suggest this mod, I think it’s a great idea that could become popular... if there’s enough space inside.

 

[joptimus]

I would instantly buy it if it were like 30-50 bucks.

 

[Mopar]

- Nice post and analysis. I'm with you about the fan blowing air out on the left, but I don't quite see the mechanics of how it also sucks air in on the right. The fan has an opening towards the rear of the machine that allows air to be blown out but it doesn't have a second opening that I can see on its right side that would allow it to suck in air. Can you explain?

View attachment 900406

See below. That's the air intake bottom left, with the five open squares.

Here's a close-up. You can see the five intake squares bottom right, next to the hinge, which are fed by the intake slot that runs along the bottom of the hinged screen.

And here's what the exhaust (left) and intake (right) vents look like when you remove them as one piece.

The air is sucked in from the right and blown out through the left of the hinge slots along the rear bottom of the chassis. It gets drawn straight across the finned heatsink, making it an active air-cooled system.

 

[JTToft]

See below. That's the air intake bottom left, with the five open squares.

- Thank you. I get how it works with respect to the case and the air channel across from the right to the left.
What I don't get is how the fan sucks air in through that channel if it doesn't have an opening besides the one towards the rear that allows it to blow the air out.
Is the air supposed to be sucked in through the top of the fan where the fins are visible?

 

[Populus]

- Thank you. I get how it works with respect to the case and the air channel across from the right to the left.
What I don't get is how the fan sucks air in through that channel if it doesn't have an opening besides the one towards the rear that allows it to blow the air out.
Is the air supposed to be sucked in through the top of the fan where the fins are visible?

Theoretically this should be the air flow that this little fan enables:

However, and as seen on the Appleinsider stress test, this fan even at its max speed (8.000 Rpms) isn’t capable of lowering the temperature from those 100ºC, proving to be an extremely inefficient cooling solution, compared with a proper heatpip—driven refrigeration. At least from my point of view.

 

[JTToft]

Theoretically this should be the air flow that this little fan enables:

- What you're sketching can't happen as far as I can see. For suction to occur, air must pass through the fan. Your sketch shows it bypassing the fan (perhaps the sketch is just inaccurate on that point).

My question isn't what direction the air is supposed to flow. I understand that. My question is in what way the fan achieves sucking air in. Does the air go in through the top of the fan where the fins are visible and then out through the back?

 

[high heaven]

See below. That's the air intake bottom left, with the five open squares.

View attachment 900440

Here's a close-up. You can see the five intake squares bottom right, next to the hinge, which are fed by the intake slot that runs along the bottom of the hinged screen.
View attachment 900441

And here's what the exhaust (left) and intake (right) vents look like when you remove them as one piece.
View attachment 900442

The air is sucked in from the right and blown out through the left of the hinge slots along the rear bottom of the chassis. It gets drawn straight across the finned heatsink, making it an active air-cooled system.
View attachment 900444

What you are saying is already a stupid idea and it's not efficient enough to cool down MBA's CPU. It's not a sever and base on your theory, it has to be the opposite. Just imagine if your desktop has no CPU fans and case fans except for the rear fan. If there is a front fan, then it might work but MBA isn't even near that. Idk how does MBA even suck airs inside with a small fan and it's not even proven yet.

Also, the gap for the cooler is not even wide enough. It's very narrow and I don't think MBA's air can suck airs from right to left. Furthermore, a fan does not have a space to suck air from the right. How does it even bring airs from the right side if it doesn't have a gap?

Adding one heat pipe would be better like 13 in MBP.

MacBook Air's cooling system is... terrible...

This is an absolute statement, which is incorrect. As others have already suggested, this can be verified empirically. This is a comparative statement, which is probably correct. Given the same design (space, load, weight) requirements, I personally am not convinced that a heatpipe would...

forums.macrumors.com

When I searching the forum, I found this. It seems several people believe the current MBA's cooling solution is efficient enough... which is not.

He already explained why MBA's cooling system is such a failure.
[automerge]1584896981[/automerge]

Theoretically this should be the air flow that this little fan enables:

View attachment 900464

However, and as seen on the Appleinsider stress test, this fan even at its max speed (8.000 Rpms) isn’t capable of lowering the temperature from those 100ºC, proving to be an extremely inefficient cooling solution, compared with a proper heatpip—driven refrigeration. At least from my point of view.

Theoretically, that's a bad cooling solution.

1. The red area is blocked. It seems, it sucks air from the top side but it's already less efficient.
2. There has to be another fan for better performance but an MBA does not.

Apple MacBook Air 2019 has a heat sink and a fan that is not connected to each other resulting in a complete CPU failure – ElectroDealPro

electrodealpro.com

 

[Falhófnir]

I guess the thing I'm wondering is why they started offering the quad i5 and especially i7 again in this machine with it's quasi-passive cooling, if it's clearly not designed to be a performance computer. If they had this setup with the dual core i3 as the only option that seems fine, they want a low power computer that's meant to try and run on passive cooling as much as possible, with the fan to assist with airflow when really needed.

By offering the (seemingly TDP 12W) i5 and i7 though that seems to be a bit of a mixed message. You can have what is actually a reasonably powerful quad core chip, but it's going to struggle to run passively and the cooling for it isn't particularly effective.

I'm hoping some future testing or maybe even firmware updates will let the i5/i7 stretch their legs a bit more, as they seem largely wasted as things stand assuming the MaxTech test video is an accurate representation of what to expect.

On the whole though the base i3 model looks like really solid value for browsing use, typing notes or office productivity. I think that's the really important model/ market for this computer and I think it will do well.

 

[Mopar]

Theoretically this should be the air flow that this little fan enables:

View attachment 900464

However, and as seen on the Appleinsider stress test, this fan even at its max speed (8.000 Rpms) isn’t capable of lowering the temperature from those 100ºC, proving to be an extremely inefficient cooling solution, compared with a proper heatpip—driven refrigeration. At least from my point of view.

This is the direction of airflow. However the air gets sucked inside the centrifugal fan from the top (as seen in this photo), as there is a gap between the top of the fan and the floorplate.

Does the air go in through the top of the fan where the fins are visible and then out through the back?

Yes, that is where the exhaust air enters. Again, note the air dam between the fan and the battery. The fan is recessed in relation to the floorplate, so there is a gap to pull the air through.
[automerge]1584899294[/automerge]

What you are saying is already a stupid idea and it's not efficient enough to cool down MBA's CPU. It's not a sever and base on your theory, it has to be the opposite. Just imagine if your desktop has no CPU fans and case fans except for the rear fan. If there is a front fan, then it might work but MBA isn't even near that. Idk how does MBA even suck airs inside with a small fan and it's not even proven yet.

Also, the gap for the cooler is not even wide enough. It's very narrow and I don't think MBA's air can suck airs from right to left. Furthermore, a fan does not have a space to suck air from the right. How does it even bring airs from the right side if it doesn't have a gap?

Adding one heat pipe would be better like 13 in MBP.

View attachment 900468

MacBook Air's cooling system is... terrible...

This is an absolute statement, which is incorrect. As others have already suggested, this can be verified empirically. This is a comparative statement, which is probably correct. Given the same design (space, load, weight) requirements, I personally am not convinced that a heatpipe would...

forums.macrumors.com

When I searching the forum, I found this. It seems several people believe the current MBA's cooling solution is efficient enough... which is not.

He already explained why MBA's cooling system is such a failure.
[automerge]1584896981[/automerge]


View attachment 900470

Theoretically, that's a bad cooling solution.

1. The red area is blocked. It seems, it sucks air from the top side but it's already less efficient.
2. There has to be another fan for better performance but an MBA does not.

Apple MacBook Air 2019 has a heat sink and a fan that is not connected to each other resulting in a complete CPU failure – ElectroDealPro

electrodealpro.com

I think you need to study the tear-down before you reach a conclusion, as your understanding is incorrect. For example, you don't appear to know how a centrifugal fan works. Also, there are vents located at each end of the rear of the case for intake and exhaust air that are clearly visible. It is a very simple design.
[automerge]1584899643[/automerge]
Here is the MacBook Air 2018 teardown. It's important to note that this is not the 2020 model. But if it uses a similar design to the 2018 model, then the way its CPU is cooled is as described: https://www.ifixit.com/Teardown/MacBook+Air+13-Inch+Retina+2018+Teardown/115201

 

[high heaven]

I think you need to study the tear-down before you reach a conclusion, as your understanding is incorrect. For example, you don't appear to know how a centrifugal fan works. Also, there are vents located at each end of the rear of the case for intake and exhaust air that are clearly visible. It is a very simple design.

"1. The red area is blocked. It seems, it sucks air from the top side but it's already less efficient.
2. There has to be another fan for better performance but an MBA does not."

Oh, I do know how it works and I already explained it. Clearly, you are the one who didn't understand.

But still, what you are explaining is WAY less efficient than what previous MacBook Air has. Most CPU coolers are directly connected to CPU itself, unlike MBA. It just moves hot air out, not the heat from the CPU itself. Do you get what I'm saying?

As you can see, the previous-gen MBA has a cooler directly connected to the CPU with a heat pipe that can transfer heats from CPU to fan efficiently.

Clearly, MBA after 2018 is a mess in terms of the cooling design.

 

[Mopar]

"1. The red area is blocked. It seems, it sucks air from the top side but it's already less efficient.
2. There has to be another fan for better performance but an MBA does not."

Oh, I do know how it works and I already explained it. Clearly, you are the one who didn't understand.

But still, what you are explaining is WAY less efficient than what previous MacBook Air has. Most CPU coolers are directly connected to CPU itself, unlike MBA. It just moves hot air out, not the heat from the CPU itself. Do you get what I'm saying?

View attachment 900482

As you can see, the previous-gen MBA has a cooler directly connected to the CPU with a heat pipe that can transfer heats from CPU to fan efficiently.

Clearly, MBA after 2018 is a mess in terms of the cooling design.

The red area is not blocked.

Hot air is only hot because it is transferring heat from the CPU.

In any case, it is pointless to argue with someone who doesn't appear to understand the basic concepts of thermodynamics.

 

[high heaven]

The red area is not blocked.

It is and even it is not, that's really a narrow space to transfer heat.

Hot air is only hot because it is transferring heat from the CPU.

Like I said, It's not even efficient enough to transfer heat. Do you even know how the cooling system works? I already doubt it can suck airs from the right side cause there is no additional fan to pull cool airs from outside.

He already explained this issue.

In any case, it is pointless to argue with someone who doesn't appear to understand the basic concepts of thermodynamics.

I think it is pointless to talk with you since you know nothing about this and yet praising Apple's mistake.

 

[JTToft]

Yes, that is where the exhaust air enters. Again, note the air dam between the fan and the battery. The fan is recessed in relation to the floorplate, so there is a gap to pull the air through.

- Very good. That’s what I wanted to know.

 

[Mopar]

Like I said, It's not even efficient enough to transfer heat. Do you even know how the cooling system works? I already doubt it can suck airs from the right side cause there is no additional fan to pull cool airs from outside.

He already explained this issue.

You are both wrong. There is a clear heat-exchange path from one side of the case to the other – over the motherboard and CPU heatsink – once the floorplate is in place. If you can't see it, what can I say?

 

[high heaven]

You are both wrong. There is a clear heat-exchange path from one side of the case to the other – over the motherboard and CPU heatsink – once the floorplate is in place. If you can't see it, what can I say?

First of all, this is a bad design in terms of cooling cause it has to be the opposite.

This is how the most server works. There are fans from behind to blow airs efficiently. But MBA's cooling design is reversed. Instead of placing a fan behind, the fan is located on the front side. Do you really think hot air can move outside smoothly? NO.

Also, the red area is so small, unlike the server that I showed above. That's why what you are explaining is clearly a stupid idea. There is a clear heat exchange? Gosh, that does NOT mean it works fine. Dont forget the main topic of this thread: POOR COOLING PERFORMANCE DUE TO POOR COOLING DESIGN. It's not about flowing hot air out, it's about cooling CPU's heat itself. Why dont you build a computer with only a heatsink and a case with a rear fan?

The red area I marked is blocked. Instead, it sucks hot air from the top side. However, that's such a small space.

I have no idea what you wanna say but that does not change the fact that MBA's cooling system is terrible in terms of design.

 

[high heaven]

By the way, previous-gen MBA used Intel U series with TDP 15W which is now being used for 13 inch MacBook Pro. It is very ironic since the old MBA used TDP 15W U series CPU which was fine but the new MBA with TDP 9W Y series suffers from heat and thermal throttling quite a lot.