M1 Air with Thermal Pad Mod

[Queen6]

FWIW, 6 months with the thermal mod, system information indicates 98% battery health after 67 cycles.

The 12" Retina MacBook's were the same with the battery degrading faster in comparison to the actively cooled Air's & MacBook Pro's. Just the nature of being passively cooled which results in higher battery temps. My own 2015 12" Retina MacBook was used heavily in hot climates with the battery degrading to 90% in the first year and 85% by the second year.

CoconutBattery's online DB confirms the same general trend with battery capacity dropping off faster than the likes of the 2014 13" MBP which saw equal use fairing better holding just over 90% until 2020. For those who use portable Mac's predominately on the mains supply AlDente is an option to hold the Mac's battery charge at 50%-60% which is generally more optimal for lithium polymer batteries in a resting state i.e. no charge/discharge.

Q-6

 

[Tenkaykev]

FWIW, 6 months with the thermal mod, system information indicates 98% battery health after 67 cycles.

Ditto, MacBook Air base model purchased early December, Modded soon after purchase ( I'd already modded the 2020 i5 Air which my wife inherited when I got my M1 ) Battery health is reported at 100% after 56 cycles. ( My M1 Air is usually connected to the charger )

 

[Significant1]

FWIW, 6 months with the thermal mod, system information indicates 98% battery health after 67 cycles.

My non-modded 4month old is also 98% after 51 cycles. So in that limited comparison, the thermal does not seem to affect battery life.

 

[Queen6]

My non-modded 4month old is also 98% after 51 cycles. So in that limited comparison, the thermal does not seem to affect battery life.

It shouldn't unless your running the M1 Air 24/7 at full load, which for the most part is unlikely. If I had the M1 Air I'd likely go with the mod, look to pull as much heat away from the batteries. That said I still think if you want linear sustained performance the M1 MBP the better solution, nor is the active cooling system the sole reason to purchase the MBP.; battery, display brightness etc. design and for some the Touchbar.

Q-6

 

[Significant1]

It shouldn't unless your running the M1 Air 24/7 at full load, which for the most part is unlikely. If I had the M1 Air I'd likely go with the mod, look to pull as much heat away from the batteries. That said I still think if you want linear sustained performance the M1 MBP the better solution, nor is the active cooling system the sole reason to purchase the MBP.; battery, display brightness etc. design and for some the Touchbar.

Q-6

I am in the other boat. I don't want the fan to EVER kick in. I rather want it to take a bit longer. I have been running my intel without turbo-boost for the same reason. That was my primary reason to go Air. I do regret it a bit now, because there might be bigger difference between the displays beside brightness (contrast), become curious of touchpad due to all the debate and because people report the fan doesn't run during normal operation.

All the other differences are insignificant to me. But comming from mba 11" I think both are way too big and heavy. I therefore bought a cheap 2020 11" ipad pro with keyboard to supplement, but wish there had been a smaller M1 macbook.

 

[Queen6]

I am in the other boat. I don't want the fan to EVER kick in. I rather want it to take a bit longer. I have been running my intel without turbo-boost for the same reason. That was my primary reason to go Air. I do regret it a bit now, because there might be bigger difference between the displays beside brightness (contrast), become curious of touchpad due to all the debate and because people report the fan doesn't run during normal operation.

All the other differences are insignificant to me. But comming from mba 11" I think both are way too big and heavy. I therefore bought a cheap 2020 11" ipad pro with keyboard to supplement, but wish there had been a smaller M1 macbook.

M1 MBP fan will only spool up at 80C plus on sustained loads. The cooling solution to me is designed to remove the excess heat fast then shutdown the fan at the soonest opportunity. Even with moderate workloads you are hardily aware the MBP has an active fan as between the mass of the chassis and background noise the fan when running is near inaudible, mostly it rarely spools up.

After back to back 30 minute Cinebench R23 runs the fan only peaked at 4.5K, mostly was in the region of 3.5K with an ambient temperature of 26C. That's all 8 cores loaded 100% so unless the ambient temperature is a lot higher or a heavier load is applied; transcoding, rendering etc. for a significant period of time the fan isn't likely to get remotely close to the maximum of 7K.

I preferred the M1 MBP over the Air for the larger battery capacity, brighter display and active cooling as in my experience where you don't have full control of the work environment these factors can and do matter. Touchbar I can take or leave. My only real negative with the M1 portables is the weak port solution, especially the MBP with previous models offering six ports versus just the two.

Q-6

 

[Significant1]

M1 MBP fan will only spool up at 80C plus on sustained loads. The cooling solution to me is designed to remove the excess heat fast then shutdown the fan at the soonest opportunity. Even with moderate workloads you are hardily aware the MBP has an active fan as between the mass of the chassis and background noise the fan when running is near inaudible, mostly it rarely spools up.

After back to back 30 minute Cinebench R23 runs the fan only peaked at 4.5K, mostly was in the region of 3.5K with an ambient temperature of 26C. That's all 8 cores loaded 100% so unless the ambient temperature is a lot higher or a heavier load is applied; transcoding, rendering etc. for a significant period of time the fan isn't likely to get remotely close to the maximum of 7K.

I preferred the M1 MBP over the Air for the larger battery capacity, brighter display and active cooling as in my experience where you don't have full control of the work environment these factors can and do matter. Touchbar I can take or leave. My only real negative with the M1 portables is the weak port solution, especially the MBP with previous models offering six ports versus just the two.

Q-6

Battery life is plenty good with air and should not really be a deciding factor. I do compilations, but not long rendering etc., so my air never really heats up and thermal throttle, which is why I also have not felt the need to implement this mod, though I have bought the materials. But my intel machines easily starts to sounds like a jet-engine with same workloads.

 

[Queen6]

Battery life is plenty good with air and should not really be a deciding factor. I do compilations, but not long rendering etc., so my air never really heats up and thermal throttle, which is why I also have not felt the need to implement this mod, though I have bought the materials. But my intel machines easily starts to sounds like a jet-engine with same workloads.

I've been in the situation where professionally that additional battery life has very much saved the day (10 hours on a 2014 13" MBP, paid for itself in a single click 👍), so I will always defer to a longer runtime.

Intel versus Apple Silicon totally different animals as the power envelopes (TDP) are completely different. This fast hex core Intel Windows notebook might be able to compare with the M1 MBP, equally all restrictions will be off and the CPU alone pulling close to 100W with three 12V fans. A restricted i9 will have the same issue and accompanying "jet" noise.

Computationally the M1 MBP is about the same, however remains quiet. Is like having 17" notebook performance in a 13" form factor, as of now that never ceases to impress...

Q-6

 

[consciousstream]

I recently bought a macbook air for my wife and modified it using the cheapest pads I could find per Timmilea's instructions. It worked great!

I also took it a step further, and I wonder what you all will think. I ordered two cheap 4x4 24 gauge copper plates, and wrapped them neatly in foil so that the copper was completely covered. Then I placed them on the bottom panel of the Hard shell case we have. This causes the case to bow out slightly, but does not keep it from snapping in place. Beneath the copper plates, I also lined the back with aluminum foil [aesthetically, this makes it so that one cannot even tell that there are plates on the back of the case]. Since doing this, I have received sustained scores of between 7400 and 7800 on multi-core benchmarks [6 tests of minimum 10 minutes each over the course of an hour). My logic was that the extra conductive material would provide more area for which the head to dissipate into, thus enabling the computer to achieve equilibrium faster before throttling.

It seems to be working!

 

[timmillea]

Welcome consciousstream! And well done for continuing the experimentation.

In fact, Apple could easily have added extra thermal conductivity to the MBA's bottom case but chose to retro-fit the M1 to the previous Intel design instead. For comparable results to mine, you can set a custom (minimum) duration in CineBench so that it is a continuous test, i.e. no 'breathers' between runs. I did one of 3 hours but found that the results were already 'steady state' within the first hour. For 'split' performances, remember to do the de-averaging calculations in a spreadsheet.

Cheap is good, passive is good so I applaud you!

My guess from your experiment is that the copper plates add thermal inertia, and hence slow down the throttling a little, as well as helping to spread heat. The hard shell case is likely to act as an insulator however.

As you know, when I run my MBA under sustained load I do it with the bottom case removed and on a stand to allow far better air circulation. What I have not yet tried is putting it on a block of steel or similar with high thermal inertia. I imagine silicone grease would then be involved and it all sounds very messy.

Keep up the experiements!

 

[Queen6]

@consciousstream I'd just be mindful of the battery temp under heavy loads as you do not want to be increasing it overly or the longevity will suffer.

Q-6

 

[timmillea]

There have been several mentions of the battery temperature of late which I have let pass. Now is the time to comment.

Needless to say, Apple put temperature sensors everywhere, including the battery, so that everything operates, at all times, according to Apple's thermal designs. If the battery gets near its temperature curve, CPU power is reduced in proportion. We don't have to think about these things, they are already taken care of.

The change brought about by a simple thermal pad mod is simply an exchange of a warmer bottom case for better sustained performance. No thermal sensors are disabled and everything runs according to design. As has been well-documented, the higher-end MBAs already have a thermal sponge to give a lift to performance on their specs sheet.

Has there been one report of a failed M1 MBA due to the mod yet? Usually someone makes up such a report and others jump on the bandwagon but there has been nothing, zilch.

I hear that there will be an updated MBA in 2022. I would be astonished if there were not a better thermal design in that. Myself, I will keep my existing MBA and perhaps put the money towards a 6k HDR display which this little, disguised beast can drive.

 

[Queen6]

There have been several mentions of the battery temperature of late which I have let pass. Now is the time to comment.

Needless to say, Apple put temperature sensors everywhere, including the battery, so that everything operates, at all times, according to Apple's thermal designs. If the battery gets near its temperature curve, CPU power is reduced in proportion. We don't have to think about these things, they are already taken care of.

The change brought about by a simple thermal pad mod is simply an exchange of a warmer bottom case for better sustained performance. No thermal sensors are disabled and everything runs according to design. As has been well-documented, the higher-end MBAs already have a thermal sponge to give a lift to performance on their specs sheet.

Has there been one report of a failed M1 MBA due to the mod yet? Usually someone makes up such a report and others jump on the bandwagon but there has been nothing, zilch.

I hear that there will be an updated MBA in 2022. I would be astonished if there were not a better thermal design in that. Myself, I will keep my existing MBA and perhaps put the money towards a 6k HDR display which this little, disguised beast can drive.

Lithium batteries exposed to elevated operating temperatures do not fail in the short term. What factually happens is the batteries capacity depletes at a faster rate in comparison to being used at the recommended parameters. People should be aware that such trade off exists and they may be impacted dependent on the relative ambient temperature.

If you don't understand the battery technology Apple employs, it's easily looked up. Likely one of the major reasons Apple sets recommend operating ranges for temperature. So there's no need for drama about batteries suddenly failing they wont, equally people should be informed of the facts.

In general you don't want lithium polymer batteries to exceed 40C as performance/longevity can be negatively impacted, nor is the slope linear being more logarithmical as the temperature increases the negative impact on the batteries accelerates. There's likely more to Apple's design than just surface temperature of the baseplate as they are very deliberate in such matters.

FWIW I also have a 12" Retina MacBook so can see first hand the impact of utilising the baseplate as a heatsink with the trend of the battery loosing capacity faster than it's counterparts, as it can and does push the battery temp past 40C with less than 10W. My rational for posting was simple as the Air in question is in a hard case which will likely insulate, driving up internal temperatures.

Thermal sponge on the 16GB model is likely to dissipate the additional heat from the SOC as the additional RAM will generate slightly higher temperatures so 16GB doesn't perform worse than the 8GB model on CPU bound tasks. That in isolation illustrates the very fine balance Apple needs to strike.

I use my Mac's professionally frequently in the subtropics/ tropics in an engineering role. As much as I appreciate the simplicity of passively cooled systems both macOS & Windows the reality is if you need sustained heavy performance an actively cooled notebook will serve better. More so if time and money is a factor.

My thoughts are people want to do the mod; have at, equally be aware of all the facts; ambient & battery temperature being significant in this case...

Q-6

 

[timmillea]

"What factually happens is the batteries capacity depletes at a faster rate in comparison to being used at the recommended parameters"

Nothing can operate outside Apple's PROGRAMMED parameters.

"equally people should be informed of the facts"

They should also be informed about mis-information.

"My thoughts are people want to do the mod; have at, equally be aware of all the facts"

I don't believe this post added any data or "facts". I struggle to see a factlet.

Nothing should be claimed without evidence. A rant is fine but please don't confuse opinion with 'facts'. Back everything up with hard data, as this thread has mostly done.

 

[Queen6]

"What factually happens is the batteries capacity depletes at a faster rate in comparison to being used at the recommended parameters"

Nothing can operate outside Apple's PROGRAMMED parameters.

"equally people should be informed of the facts"

They should also be informed about mis-information.

"My thoughts are people want to do the mod; have at, equally be aware of all the facts"

I don't believe this post added any data or "facts". I struggle to see a factlet.

Nothing should be claimed without evidence. A rant is fine but please don't confuse opinion with 'facts'. Back everything up with hard data, as this thread has mostly done.

Simple facts and only reinforces that you don't get the post, the intention or remotely understand the battery technology. Running the Air with the mod in 20C on a stand with the baseplate removed is magnitudes different to it being on the desk in 30C, hence why those in the latter category see vastly differing results, resultant confusion and potentially pushing the battery pack past 40C.

If the battery runs hot the charge rate will be reduced, however the effects are not reversible as battery capacity will diminish. This takes several years, so it just depends if people value 10%-15% additional performance over long term battery longevity.

There's a no shortage of technical data related to lithium polymer batteries and the effects of temperature if one cares to actually look, as in solid engineering & scientific data versus your "opinion". Again I'm not saying don't do the mod, just be aware of all the positives and negatives with ambient temperature being a major factor.

Q-6

 

[timmillea]

Q-6, You misunderstand - the bottom of an external protective plastic case removed, not the aluminium bottom case of the MBA.

Apple takes care of the thermal envelopes with sensors everywhere and their algorithms, not mine or yours.

Please don't patronise me re. battery technology. Lithium polymer is relatively old hat, the research papers about it are two decades old now. It is a cheap and convenient battery tech. The charge/discharge algorithms are the only advance and Apple are ahead of the market, but not the science, with this.

I provide data with actual numbers and charts from my own experiments backed up with a summary and hypothesise as a suggestion for further work. There is nothing but bluster an opinion in your posts. Provide your own evidence to support 'paintbrush' claims or desist, as a benefit to the wider community.

1) Aim of experiment
2) Conditions of experiment
3) Results
4) Conclusion and suggestions for further work

Simples!

 

[consciousstream]

My impression was that the back panel does not actually even touch the battery. As a result, the ambient temperature inside the chassis is unaffected by the modification, which means that the temperature of the battery would be likewise unaffected by the modification.

If this is the case, then the above poster's concerns are unfounded.

Edit: Thanks for the replies and support! I am aware that the back panel of the shell case acts as insulation. However, I wanted to see if the tradeoff of Thermal inertia with the copper plates would offset the use of the insulating case. I wanted my wife to have the benefits of the mod without the downside of not being able to use a case. It seems to be working thus far!

I will run your test later today to see if it works over longer stretches.

Thanks again for the support!

 

[timmillea]

My impression was that the back panel does not actually even touch the battery. As a result, the ambient temperature inside the chassis is unaffected by the modification, which means that the temperature of the battery would be likewise unaffected by the modification.

If this is the case, then the above poster's concerns are unfounded.

Edit: Thanks for the replies and support! I am aware that the back panel of the shell case acts as insulation. However, I wanted to see if the tradeoff of Thermal inertia with the copper plates would offset the use of the insulating case. I wanted my wife to have the benefits of the mod without the downside of not being able to use a case. It seems to be working thus far!

I will run your test later today to see if it works over longer stretches.

Thanks again for the support!

I did do tests, posted on this thread eons ago, showing that the mod does raise the battery temperature. I attach the chart. However, Apple's algorithms kick-in no matter what is over-heating to keep the whole computer safe.

Good luck to those who wish to reverse-engineer Apple's algorithms and think they know better.

Chart attached: unmodified (blue), with mod (green), with mod & laptop stand (grey). These are the Apple sensor recorded battery temperatures over a 60 min (minimum) Cinebench test, plus (the remainder of) the 80 minutes after cool-off, i.e. idle.

Conclusion
The mod does increase battery temperature but only within Apple's designed safe limits. Under a sustained load of around 15 minutes duration or less, the mod keeps the battery cooler. A laptop stand extends this to around 20 minutes. At peak, steady-state under maximum sustained load after an hour, the battery is around 4 degrees warmer with the mod, and 2 degrees warmer with the mod on a stand, compared to no mod.

 

[Queen6]

Q-6, You misunderstand - the bottom of an external protective plastic case removed, not the aluminium bottom case of the MBA.

Apple takes care of the thermal envelopes with sensors everywhere and their algorithms, not mine or yours.

Please don't patronise me re. battery technology. Lithium polymer is relatively old hat, the research papers about it are two decades old now. It is a cheap and convenient battery tech. The charge/discharge algorithms are the only advance and Apple are ahead of the market, but not the science, with this.

I provide data with actual numbers and charts from my own experiments backed up with a summary and hypothesise as a suggestion for further work. There is nothing but bluster an opinion in your posts. Provide your own evidence to support 'paintbrush' claims or desist, as a benefit to the wider community.

1) Aim of experiment
2) Conditions of experiment
3) Results
4) Conclusion and suggestions for further work

Simples!

Ok thx for the clarification of the hard case, just that I've seen stranger things here with some running the old 15" MBP's with the baseplate removed 😱 equally they are real burners.

The limits of the battery technology stands, you may not be affected as your clearly in a colder climate, others in hotter climates may. As your familiar you'll know that 40C - 45C you'll observe a drop off in performance and increased wear levels that's just how it is, anyone can see for themselves with a simple search.

If I had the M1 Air I'd look at the mod differently as the performance increase will be dependant on ambient temperature. What I'd consider is the battery temp pre & post mod and just see what the delta was. I can see on the 12" rMB (fully charged) that running CB R23 the battery temperature inside 30 minutes increases by 12C/13C with an ambient of 26C on a aluminium stand, off the stand 40C+. Think this was one of the factors of Apple dropping the rMB as they couldn't increase performance without raising both external & internal temperature as this Mac relies on the baseplate for cooling. Seems something your not willing to consider despite being a potential issue for others...

Q-6

 

[Queen6]

My impression was that the back panel does not actually even touch the battery. As a result, the ambient temperature inside the chassis is unaffected by the modification, which means that the temperature of the battery would be likewise unaffected by the modification.

If this is the case, then the above poster's concerns are unfounded.

Edit: Thanks for the replies and support! I am aware that the back panel of the shell case acts as insulation. However, I wanted to see if the tradeoff of Thermal inertia with the copper plates would offset the use of the insulating case. I wanted my wife to have the benefits of the mod without the downside of not being able to use a case. It seems to be working thus far!

I will run your test later today to see if it works over longer stretches.

Thanks again for the support!

It doesn't, however the baseplate radiates heats back the battery. My 12" Retina MacBook is the same as it utilises the chassis for cooling resulting in higher battery temps in comparison to other portable Mac's.

Q-6

 

[consciousstream]

It doesn't, however the baseplate radiates heats back the battery. My 12" Retina MacBook is the same as it utilises the chassis for cooling resulting in higher battery temps in comparison to other portable Mac's.

Q-6

With the placement of the copper plates on the underside of the case, I don't think the area of the back case that is close to the battery heats up to a greater degree than without the modification anyway. Everything is positioned to draw heat away from that portion of the back panel. I believe earlier posters also already did tests on the differences in temperature between modded and unmodded M1s. The difference in heat on the battery was negligible if I remember correctly.

 

[Queen6]

I did do tests, posted on this thread eons ago, showing that the mod does raise the battery temperature. I attach the chart. However, Apple's algorithms kick-in no matter what is over-heating to keep the whole computer safe.

Good luck to those who wish to reverse-engineer Apple's algorithms and think they know better.

Chart attached: unmodified (blue), with mod (green), with mod & laptop stand (grey). These are the Apple sensor recorded battery temperatures over a 60 min (minimum) Cinebench test, plus (the remainder of) 20 minutes after cool-off, i.e. idle.

View attachment 1819083

Maybe so, however here's your Air & your words with a battery temp close to 45C with an ambient in the low 20C range, so the battery temperature is clearly impacted by high workloads for extended periods of time. Logically the heat imparted to the baseplate by the thermal pad will radiate both externally & internally.

To reiterate I'm not against the mod, it's great way to extend the thermal envelope of the M1 Air, equally battery temperature is a factor and should be open for discussion. You have no right or permission to silence another members opinion so knock it off...

Many were not overly pleased with Apple and the 12" rMB as battery performance versus typical portable Macs as the 12" mostly dropped to 80%-85% in a couple of years. Likely culprit was the elevated battery temps due to the baseplate being the heatsink and radiating heat back to the battery, Apple learned the lesson and moved on.

Ultimately all I'm saying is for those in hotter climates this is a consideration worth monitoring as the battery will run hotter...

Q-6

 

[Queen6]

With the placement of the copper plates on the underside of the case, I don't think the area of the back case that is close to the battery heats up to a greater degree than without the modification anyway. Everything is positioned to draw heat away from that portion of the back panel. I believe earlier posters also already did tests on the differences in temperature between modded and unmodded M1s. The difference in heat on the battery was negligible if I remember correctly.

It's a slow process as the elevated temperature wicks away at the capacity. For some there will be no issue and others potentially problematic, nor should it be arbitrarily dismissed. Ideally you want to keep the battery pack in the 20C range, equally that's very far from practicality by any means. You should pay attention if the battery is exceeding 40C for longer periods of time and look to avoid, simply raising the rear of the notebook helps to dissipate excess heat.

Apple's own spec states a max of 45C for storage, which is very likely related to the battery. As with all in life if you need to push people or equipment hard, you do so and deal with the consequences as and when

Q-6

 

[houttbe]

Q-6, You misunderstand - the bottom of an external protective plastic case removed, not the aluminium bottom case of the MBA.

Apple takes care of the thermal envelopes with sensors everywhere and their algorithms, not mine or yours.

Please don't patronise me re. battery technology. Lithium polymer is relatively old hat, the research papers about it are two decades old now. It is a cheap and convenient battery tech. The charge/discharge algorithms are the only advance and Apple are ahead of the market, but not the science, with this.

I provide data with actual numbers and charts from my own experiments backed up with a summary and hypothesise as a suggestion for further work. There is nothing but bluster an opinion in your posts. Provide your own evidence to support 'paintbrush' claims or desist, as a benefit to the wider community.

1) Aim of experiment
2) Conditions of experiment
3) Results
4) Conclusion and suggestions for further work

Simples!

This may have been addressed in the discussion but if it was, I overlooked it.

Does it make sense to carry out the mod if the MBA has an external protective plastic case covering the bottom plate? Mine has some vents but I expect that the hard case reduces or perhaps even prevents the conduction and radiation of heat accumulating in the bottom aluminum plate.

 

[steve217]

@timmillea, thanks for the hard quantitative data you've provided on this mod. My 2mm thermal pad is on the way.

Would you mind providing some non-quantitative data?

To wit:

• Would you open the back of your Mac and post a photo? I would like to compare the machine after 8 months of this mod with the pictures in this post.
• Any sign of heat wear?
• What’s your Handbrake experience been like?
• Any particular incidents where the heat was noticeable?