M1 Air with Thermal Pad Mod

[downshiftdre]

Hi all just "upgraded" from an early 2020 MBA i3 to the M1 MBA 8/256. 2020 lurker here.

On my prior i3 MBA, I had performed the thermal pad mod with great results as shown with various users in the extensive thermal mod thread. lol at that laptop performance compared to the M1.

I ordered the 200mm x 200mm x 3mm thermal pad and it will be arriving tomorrow. The 3mm is because the heat sink appears to sit lower and therefore farther from the back case when it is closed. The 3mm will ensure contact to the aluminum.

Since the heat sink on the M1 MBA is an elongated piece from the processor all the way to the edge, I plan to cover the whole piece edge to edge. Since this processor barely breaks a sweat I will also remove the sticky tape covering on the case to ensure best adhesion. I have no issue with additional heat that will be on the backside since I know this processor barely heats up doing mainstream things.

Geekbench doesn't seem to have any impact on M1 thermal since the processor barely heats up to 50C without breaking a sweat through multiple repeated runs of up to 5 times with no change to score.

The test will be performed with Cinebench R23 through the 10 minute thermal test run for multicore to see if the score drops into the low 6000s or not. On the single loop, multicore runs from YouTube it seems the M1 MBA and MBP are aligned in scores and only on the 2nd or 3rd run does the MBA begin to throttle.

Any input, please provide.

Stay tuned.

EDIT: 10 minute and 30 minute cinebench runs performed with incredible results shown in page 1 and page 2 of this thread.

 

[kepler20b]

also please let us know if the process of removing the back cover is the same,

and if you actually do get heat xfer onto the back shell when you load the cpu


I also bought some pads + new paste in anticipation of buying an intel mbp this fall. I went with the m1 air and I dont know if there is really any reason to do this mod now. but I look forward to your results.

 

[downshiftdre]

Results are in

The M1 Macbook Air no longer thermal throttles under a 10 minute Cinebench R23 test. And now operates at MBP levels for such a test.

Cinebench R23 10 minute multi core tests

Before thermal pad mod:
6412 with a warm backshell case near where backshell meets hinge

After thermal pad mod:
7713 with a hot backshell case (i can touch it and leave my hand on it but is like a hot cup of tea)

The heatsink has a lot of top surface area so it is able to quickly transfer heat to the backshell. While watching the thermals, it appeared the system stabilized into a heat equilibrium at around 3-4 mins in. After the test was complete the system quickly went back down to ambient temps within a few minutes and the backshell did not retain heat.

As for the heat on backshell, For my use case, if I am using it on my lap, I am almost always doing very basic things: movies, music, internet browsing etc. Regular use does not heat up the backshell case.

If I am going to be doing something cpu intensive and being serious, then the laptop is on a desk and the heat will not bother me nor will I know about it. At least all that heat is not staying inside the laptop baking things. The M1 MBA has no vents whatsoever.

I will continue to monitor how this works for me.

Overall, if you dont want to spend $300 for the M1 MBP, this $15 reversible thermal mod gets you near equal to it since it seems the M1 MBP barely even runs its own fan based on all youtube reviewers.

 

[jay-m]

That's great news. I have 2 leftover thermal pads from old project and I'll use them for the same purpose as you once my MBA arrives next week.

What ambient temperature were you testing at (might be approximate, like "I was wearing a t-shirt")?
Was 3mm right thickness? Isn't it too loose or too tight? Ifixit's teardown shows that there is a 'step' in the radiator, I wonder if lower part needs thicker pad.
Did you remove the black masking covering backshell?

 

[Conutz]

Results are in

The M1 Macbook Air no longer thermal throttles under a 10 minute Cinebench R23 test. And now operates at MBP levels for such a test.

Cinebench R23 10 minute multi core tests

Before thermal pad mod:
6510 with a warm backshell case near where backshell meets hinge

After thermal pad mod:
7559 with a hot backshell case (i can touch it and leave my hand on it but is like a hot cup of tea)

The heatsink has a lot of top surface area so it is able to quickly transfer heat to the backshell. While watching the thermals, it appeared the system stabilized into a heat equilibrium at around 3-4 mins in. After the test was complete the system quickly went back down to ambient temps within a few minutes and the backshell did not retain heat.

As for the heat on backshell, For my use case, if I am using it on my lap, I am almost always doing very basic things: movies, music, internet browsing etc.

If I am going to be doing something cpu intensive and being serious, then the laptop is on a desk and the heat will not bother me nor will I know about it. At least all that heat is not staying inside the laptop baking things. The M1 MBA has no vents whatsoever.

I will continue to monitor how this works for me.

Overall, if you dont want to spend $300 for the M1 MBP, this $15 reversible thermal mod gets you pretty darn close.

Have you got some pics of before and after?

 

[downshiftdre]

Ambient temperature is around 65-70F. Jeans and long sleeve.

I used 3mm for all of it. There is a step down in the heat sink so I cut 2 seperate pieces. If I were to do over again, I would use 2mm for the higher portion of heat sink and 3mm for the lower portion.

I did remove the black masking covering on the backshell.

I did have to push the backshell while tightening up the screws in that area to get that initial squeeze with the pad. Also, Make sure to push the 2 tabs on the backshell in first before tightening screws.

 

[downshiftdre]

Have you got some pics of before and after?

 

[Conutz]

Thanks. Nice and easy it seems, although I think you’re brave!

 

[circatee]

More like very brave! Nicely done, and thanks for sharing...

 

[jerryk]

Is that something like this? https://www.amazon.com/100x100x1-5m...YMYXFC4G84V6&refRID=56G0P4C6YMYXFC4G84V6&th=1

 

[downshiftdre]

Is that something like this? https://www.amazon.com/100x100x1-5m...YMYXFC4G84V6&refRID=56G0P4C6YMYXFC4G84V6&th=1

Yes exactly that one with the 6W/m thermal conductivity. I would recommend to get 2mm for the higher portion of heat sink and 3mm for the lower part of heat sink. Less of a squeeze when buttoning it all back up. I went with 3mm all around for this first try at it.

 

[kepler20b]

did you also reapply thermal paste?

 

[downshiftdre]

did you also reapply thermal paste?

No I did not touch the thermal paste. Just cover off, apply thermal pads to heatsink, close cover. This mod is only for the thermal pad and is already beyond expectations.

 

[kepler20b]

thanks for the showing your progress.


I think for my use case i'll

reapply thermal paste

and only add thermal pads to the higher portion of the heatsink. I dont want the back shell to get too hot and dont mind a little throttling under sustained load. it appears that the "lower" part of the heatsink sits directly on top of the actual cpu/sodimms so you're gonna get the most heat xfer from the lower plate. adding pads to the higher plate may not much impact if you already have pads on the lower portion of the heatsink

 

[downshiftdre]

thanks for the showing your progress.


I think for my use case i'll

reapply thermal paste

and only add thermal pads to the higher portion of the heatsink. I dont want the back shell to get too hot and dont mind a little throttling under sustained load. it appears that the "lower" part of the heatsink sits directly on top of the actual cpu/sodimms so you're gonna get the most heat xfer from the lower plate. adding pads to the higher plate may not much impact if you already have pads on the lower portion of the heatsink

You have it backwards on the heatsink heights. The raised portion of heatsink is right over the cpu. The drop then happens where the heatsink narrows as you travel the length of the heatsink from cpu to corner edge of the laptop.

I do see what you are saying and thought of that before I did it this way to cover it all. Heat transfers from hot to cold. You are thinking to put heat sink at the corner portion (dropped portion), so the heat will first transfer the length of the heat sink from cpu to corner and once that gets saturated will start to transfer from corner heatsink to the backshell aluminum case.

My thought process was under major thermal load, its all going to be nearly same high temperature so I wanted to maximize any surface area with thermal pad on the heatsink to case to end up at a better heat capacity state for best heat extraction out of the laptop internals.

 

[kepler20b]

You have it backwards. The raised portion of heat sink is right over the cpu. The drop then happens at the corner edge of the laptop.

I do see what you are saying and thought of that before I did it this way. Heat transfers from hot to cold. You are thinking to put heat sink at the corner portion (dropped portion), so the heat will first transfer the length of the heat sink from cpu to corner and once that gets saturated will start to transfer from corner heatsink to the backshell aluminum case.

thx,

yeah pretty much. but the difference between our methods probably won't have a big impact as it is so difficult to saturate the heatsink under 99% of workloads.

 

[downshiftdre]

thx,

yeah pretty much. but the difference between our methods probably won't have a big impact as it is so difficult to saturate the heatsink under 99% of workloads.

When you do the thermal paste please post your results and also how and you used to cover the processor and the unified memory chips. The unified memory chips have this weird thick glob of black goop on it and is unlike the thin paste on the processor. Unknown to me if that black goop is thermal paste or something else.

 

[rulymammoth]

Is there any risk of damage or other unintended consequences from doing this?

 

[Tenkaykev]

Is there any risk of damage or other unintended consequences from doing this?

I did both the shim mod and pad mod on my i5 air. The shim required removing the heatsink where there’s a risk of stripping the threads on the tiny screws. The pad is risk free as it’s just applying a thermal pad to the top of the heatsink to bridge the gap to the bottom of the air and help thermal transfer.

 

[adrianlondon]

I'm not an expert in these things, but logic tells me that if the thermal pad is enough to make the bottom of the casing hot, then the thermal paste must already be doing a good job. So I'm not sure it's worth removing the heatsink to check and replace the paste.

I have a 2020 i5 Air, and I read the main thread on this topic but so far have done nothing. It's winter here at the moment, so the only time I notice the fans come on is when upgrading Big Sur betas. However, a few months ago in the main heat of summer, watching videos would also do it, so I'll probably install the thermal pad before it gets hot here again. I can't be bothered to do the shim/paste. Lazy.

 

[Nitrogen]

Is there any risk of damage or other unintended consequences from doing this?

Can imagine battery coils heating up to 110+ F territory easily.
Well

it already was 113 F in the test below

this article shows how keeping the battery over 40 Celsius for longer periods of time might end up killing battery capacity. https://batteryuniversity.com/learn/article/how_to_prolong_lithium_based_batteries

here for example, m1 mbp battery temperature rounded to 36 Celsius max during blender, while MBA had it around 43 Celsius.

imagine lower case temps with thermal pad mod.
Afaik there is a plenty of 65-70% battery capacity MB 2015-2017 just because of cramped hardware and entirely fanless design, so temperatures as a factor. Most of MBP 13’ retinas and old MBAs are 80+ % battery health due to at least 1 fan and slimmer heatsink connected to the fan (and the heat spreading not as much to interiors)

Bottom line - while it helps performance and makes mba perform like mbp in sustained performance, that mod may lead to even more premature deterioration of one thing about new M1 macs we like the most - it’s battery life.

 

[downshiftdre]

Can imagine battery coils heating up to 110+ F territory easily.
Well

it already was 113 F in the test below

this article shows how keeping the battery over 40 Celsius for longer periods of time might end up killing battery capacity. https://batteryuniversity.com/learn/article/how_to_prolong_lithium_based_batteries

here for example, m1 mbp battery temperature rounded to 36 Celsius max during blender, while MBA had it around 43 Celsius.

imagine lower case temps with thermal pad mod.
Afaik there is a plenty of 65-70% battery capacity MB 2015-2017 just because of cramped hardware and entirely fanless design, so temperatures as a factor. Most of MBP 13’ retinas and old MBAs are 80+ % battery health due to at least 1 fan and slimmer heatsink connected to the fan (and the heat spreading not as much to interiors)

Bottom line - while it helps performance and makes mba perform like mbp in sustained performance, that mod may lead to even more premature deterioration of one thing about new M1 macs we like the most - it’s battery life.

My thought process is that I think the thermal pad mod helps to extract the heat from the inside to the outside of the laptop. Currently heat sits in the heat sink and as it reaches max heat capacity it will heat soak the interior of the laptop. Baking everything in it like an oven. We don't feel it from the outside because the thermal tape and the back shell prevent it from being exposed.

The hot back shell with the thermal pad mod is not an indicator that battery is also hot, it is only showing that heat is escaping from the inside to the outside.

One more thing in response to the battery heat theory. The batteries get hot already when charging and you can feel that near the middle of the laptop. This thermal pad mod does not heat up that part of the back shell meaning the batteries are unlikely to be impacted by this mod.

 

[Nitrogen]

My thought process is that I think the thermal pad mod helps to extract the heat from the inside to the outside of the laptop. Currently heat sits in the heat sink and as it reaches max heat capacity it will heat soak the interior of the laptop. Baking everything in it like an oven. We don't feel it from the outside because the thermal tape and the back shell prevent it from being exposed.

The hot back shell with the thermal pad mod is not an indicator that battery is also hot, it is only showing that heat is escaping from the inside to the outside.

One more thing in response to the battery heat theory. The batteries get hot already when charging and you can feel that near the middle of the laptop. This thermal pad mod does not heat up that part of the back shell meaning the batteries are unlikely to be impacted by this mod.

I did similar mod to mbp 13 early 2015.

The battery toasted like hell.
Benching or testing it yourself before theoretical assumptions might yield better results.

 

[ScooterComputer]

reapply thermal paste

and only add thermal pads to the higher portion of the heatsink. I dont want the back shell to get too hot and dont mind a little throttling under sustained load. it appears that the "lower" part of the heatsink sits directly on top of the actual cpu/sodimms so you're gonna get the most heat xfer from the lower plate. adding pads to the higher plate may not much impact if you already have pads on the lower portion of the heatsink

As @adrianlondon mentioned, I don't think I'd bother with the thermal paste UNLESS doing what you're looking to do doesn't seem to produce the expected thermal spread to the cover. (eg If the cover gets hotter, the thermal pads are working and hence so is the heatsink, and therefore the current thermal paste.) That would save you the hassle of peeling the heatsink.
I'm interested in your results, vs @downshiftdre. If you are able to accomplish similar improvements with a slower heat spread and/or (hopefully) a lower overall cover temp (by avoiding the heatsink directly above the CPU from piping heat directly into the cover, using air insulation as a damper), it would be even better IMHO (better "solution" for laps, especially!).

 

[downshiftdre]

I did similar mod to mbp 13 early 2015.

The battery toasted like hell.
Benching or testing it yourself before theoretical assumptions might yield better results.

Interesting experience. Can you describe in numbers what you mean by “battery toasted like hell”?

It failed in 2 days? It failed over a period of 3 years? Please describe.

Do you by any chance put your laptop on your bed and run it? I avoid that unless its minor low cpu intensive tasks (safari, music etc.). I can see how a laptop being in a position like that may start to spread its heat to other components as the bed or blanket starts to act like an insulator and spreader.

I am running the 30 minute cinebench test now and will have before, during and after temperatures of M1 SoC cores and the battery and battery management temperatures.