Arctic MX-5 vs Indium Foil DIY thermal pad

J

JSHamlet234

Gawd
Joined
Apr 9, 2021
Messages
943
So I mentioned in the thermal paste thread that I ordered some Indium foil to attempt to use like a thermal pad. Well it's here, and I had to chance to run it against my new favorite thermal paste.

The sheet I received was 100mm x 100mm x 0.2mm. It cost me $20, and it's enough to do 9 CPUs, so at ~$2 per application, it's not the most expensive TIM I've ever used, but it's definitely not the cheapest either.

First things first - 0.2mm is a little thicker than I thought it would be, but it ensures that it will be thick enough to provide consistent contact.

Test Bed:

Intel 5960X @ 4.2GHz / 1.175V manual voltage, all EIST and C-States disabled
ASUS X99-A USB 3.1 Motherboard, BIOS 3701
EVGA CLC 280mm mounted as front exhaust
4X Arctic P14 PWM rad fans in push/pull @1470RPM (90% PWM)
Coolermaster HAF-XB EVO with Noctua A20 Chromax top intake fan @ 810RPM (90% PWM)
Prime 95 26.6 Small FFTs
15 minute run time to ensure steady state
Ambient temperature measured 3 inches from top intake fan.
Temperatures reported are an average over the final 60 seconds as reported by HWINFO64.

Arctic MX-5

Ambient Temp: 23.6C
Water Temp: 31.6C
Package Temp: 61.7C
Cores: 54.4, 53.6, 61.7, 56.9, 58.0, 57.3, 58.6, 54.6
Delta from ambient temp: 38.1C
Delta from water Temp: 30.1C
Average Power: 198.0 watts
Thermal Resistance (system): 0.19242 C/watt

Indium Foil [0.2mm thickness]

Ambient Temp: 23.8C
Water Temp: 31.6C
Package Temp: 60.5C
Cores: 53.3, 54.1, 60.5, 58.0, 57.7, 57.8, 57.5, 55.1
Delta from ambient temp: 36.7C
Delta from water temp: 28.9C
Average Power: 197.5 watts
Thermal Resistance (system): 0.18582 C/watt

The improvement over good paste was 1.4C - this is about the same improvement I would have expected from liquid metal based on my previous testing.

I'm gonna chalk this one up in the WIN column.
 
Last edited:
JSHamlet234 said:
So I mentioned in the thermal paste thread that I ordered some Indium foil to attempt to use like a thermal pad. Well it's here, and I had to chance to run it against my new favorite thermal paste.

The sheet I received was 100mm x 100mm x 0.2mm. It cost me $20, and it's enough to do 9 CPUs, so at ~$2 per application, it's not the most expensive TIM I've ever used, but it's definitely not the cheapest either.

First things first - 0.2mm is a little thicker than I thought it would be, but it ensures that it will be thick enough to provide consistent contact.

Test Bed:

Intel 5960X @ 4.2GHz / 1.175V manual voltage, all EIST and C-States disabled
ASUS X99-A USB 3.1 Motherboard, BIOS 3701
EVGA CLC 280mm mounted as front exhaust
4X Arctic P14 PWM rad fans in push/pull @1470RPM (90% PWM)
Coolermaster HAF-XB EVO with Noctua A20 Chromax top intake fan @ 810RPM (90% PWM)
Prime 95 26.6 Small FFTs
15 minute run time to ensure steady state
Ambient temperature measured 3 inches from top intake fan.
Temperatures reported are an average over the final 60 seconds as reported by HWINFO64.

Arctic MX-5

Ambient Temp: 23.6C
Water Temp: 31.6C
Package Temp: 61.7C
Cores: 54.4, 53.6, 61.7, 56.9, 58.0, 57.3, 58.6, 54.6
Delta from ambient temp: 38.1C
Delta from water Temp: 30.1C
Average Power: 198.0 watts
Thermal Resistance (system): 0.19242 C/watt

Indium Foil [0.2mm thickness]

Ambient Temp: 23.8C
Water Temp: 31.6C
Package Temp: 60.5C
Cores: 53.3, 54.1, 60.5, 58.0, 57.7, 57.8, 57.5, 55.1
Delta from ambient temp: 36.7C
Delta from water temp: 28.9C
Average Power: 197.5 watts
Thermal Resistance (system): 0.18582 C/watt

The improvement over good paste was 1.4C - this is about the same improvement I would have expected from liquid metal based on my previous testing.

I'm gonna chalk this one up in the WIN column.
Click to expand...

Can you link to the foil that you ordered, please?
 
Intrigued to know the temps without either of them. You never know...
 
Any risk of pieces flaking off like with aluminum foil (don't ask)?
 
michalrz said:
Any risk of pieces flaking off like with aluminum foil (don't ask)?
Click to expand...

I don't think it's a big risk, but use common sense because it's highly conductive. Indium is a very soft, very ductile metal that wants to bend, not break or tear if that makes any sense. I had a chance to inspect it yesterday because I wanted to try mounting my water block in a different orientation to see how it would affect temps. The block came off without any resistance and the foil remained the processor. It didn't squeeze out beyond the edges of the CPU or anything. I did not try to remove the foil because it seemed to have bedded in nicely, and I didn't want to disturb it. I've mounted, unmounted, and remounted the block several times with no decrease in performance.

I'm surprised that using indium sheets isn't a more common practice, especially in the extreme overclocking arena. It's not that expensive, it performs slightly better than paste, and it's reusable. Then again, there probably isn't much profit to be had, and marketing is what tends to drive this industry. Unlike paste or liquid metal where a manufacturer can mix up a "secret sauce" of fairly cheap ingredients and make a large profit, indium is just indium. You can't patent it, nor can you differentiate yours from anyone else's. It costs what it costs, and it does what it does.
 
JSHamlet234 said:
I'm surprised that using indium sheets isn't a more common practice, especially in the extreme overclocking arena.
Click to expand...
I'm no doctor, but the only thing I can think of is some potential detrimental result of thermal expansion/constriction.
That said, I'll surely try it, thanks.
 
michalrz said:
I'm no doctor, but the only thing I can think of is some potential detrimental result of thermal expansion/constriction.
That said, I'll surely try it, thanks.
Click to expand...

Could be. I know it's the same stuff Intel and AMD use to solder their CPUs. I had a chance today to take some pics of the test bed. Also attached a screen shot of HWINFO64 with the best thermal results I've gotten so far after reorienting the block 90 degrees CCW and getting the most even mounting pressure I could. Ambient temp was 26.9C for these readings, fans maxed out. Prime95 26.6 small FFT. Not bad for a 5960X during a summer heatwave.
 

Attachments

  • IMG_20210627_210957683.jpg
    IMG_20210627_210957683.jpg
    276 KB · Views: 0
  • IMG_20210627_211033469.jpg
    IMG_20210627_211033469.jpg
    377.5 KB · Views: 0
  • IMG_20210627_211125665.jpg
    IMG_20210627_211125665.jpg
    435.8 KB · Views: 0
  • IMG_20210627_211143812.jpg
    IMG_20210627_211143812.jpg
    321.5 KB · Views: 0
  • thermals 6-27-21 1648.png
    thermals 6-27-21 1648.png
    1.3 MB · Views: 0
Last edited:
JSHamlet234 said:
Could be. I know it's the same stuff Intel and AMD use to solder their CPUs. I had a chance today to take some pics of the test bed. Also attached a screen shot of HWINFO64 with the best thermal results I've gotten so far after reorienting the block 90 degrees CCW and getting the most even mounting pressure I could. Ambient temp was 26.9C for these readings, fans maxed out. Not bad for a 5960X during a summer heatwave.
Click to expand...
now you make me want try it against my graphite pad......what did this hole idea come from anyways?
 
I was watching a video by Der8auer where he talks about soldering a non-soldered CPU using a sheet of indium and all the stuff involved with that, and I said to myself, hey, why couldn't I use that same stuff as a thermal pad? So I researched it, and apparently it actually is used as TIM for industrial laser diodes and other applications.
 
JSHamlet234 said:
I was watching a video by Der8auer where he talks about soldering a non-soldered CPU using a sheet of indium and all the stuff involved with that, and I said to myself, hey, why couldn't I use that same stuff as a thermal pad? So I researched it, and apparently it actually is used as TIM for industrial laser diodes and other applications.
Click to expand...
Any possible long term drawbacks to using it?
 
JSHamlet234 said:
So far so good. It's possible that it might slowly diffuse into the copper and form an alloy, maybe over a period of years.
Click to expand...

I ordered one but I'm not expecting good long term results.
The only post I can see where someone actually used thin indium sheets on a CPU started off good (pretty much the same as paste), but after a month, idle temps increased by 5C and load temps increased by 15C.
So please remember to update your long term results.
 
Falkentyne said:
I ordered one but I'm not expecting good long term results.
The only post I can see where someone actually used thin indium sheets on a CPU started off good (pretty much the same as paste), but after a month, idle temps increased by 5C and load temps increased by 15C.
So please remember to update your long term results.
Click to expand...
I actually found the thread I believe you're referring to at https://www.overclock.net/threads/indium-sheets-instead-of-thermal-paste.1054484/page-3

"Update after 1 week: Due to unknown reasons, my temps have recently increased by about 5C when at idle and 15C at load compared to immediately after applying the indium. For this reason, i have decided to go back to AS5. When i decide that i want to lap my CPU, i will again try the indium."

He used a sheet that was 0.005", which is 0.127mm. I also found a Youtube video where Hardware Canucks tested both a 0.15mm sheet and a 0.2mm sheet on a 6700K. The .2mm sheet was slightly better than paste, and even though they didn't mention the results from the 0.15mm sheet in the video, they responded in the comments that the 0.15mm sheet was about 3-4C worse than paste. It could be that with tolerances being what they are, and the fact that this stuff might cold-flow a bit, that a sheet thinner than 0.2mm could be problematic, sort of like what happens when you use too little liquid metal on uneven surfaces.

4 days so far. We'll see what happens.
 
With enough thickness I'll wager it continuing to perform well for at least a decade, possibly forever.
Its melting point is 156C, its not going anywhere soon at PC temps.
https://www.rsc.org/periodic-table/element/49/Indium

This pdf has a diagram showing its expansion with temperature
https://nvlpubs.nist.gov/nistpubs/jres/30/jresv30n6p427_A1b.pdf
From around 25C to 100C it expands only just over 0.2%.
Not being liquid it will only move where there is an excess of pressure, it will otherwise remain static.
There could be a minor loss of contact at lower temps causing a slight rise in temp.
At high temp full contact will be made, when it matters.

Edit
There could be a slight downside if the PC is ever allowed to overheat, it may expand+move enough to leave a slight loss of contact at normal operating temps.
But I think it would be such a low cooling loss that it wont matter unless going for an absolute max overclock.
It would be interesting to see the temp at the CPUs normal max speed with normal cooling, then what happens when used at the CPUs throttling temp (cooling disabled) for an hour and then back at the CPUs max speed with normal cooling again.
The variance in temp before and after is critical to how useful this will be long term.

ps
Nice one OP for giving it a go.
 
Last edited:
Yeah, the 0.2mm sheet is pretty meaty. We'll see what happens. Over time, the mount for the waterblock will also experience some fatigue, but I think there's enough bulk material to maintain contact.

Re-ran the test tonight. Ambient temp was 27.1C. Package temp and hottest core were the same as last night. There were some slight changes in a couple of cores, with a slightly lower average temp due to Core 6 being down more than 2 degrees from last night.
 

Attachments

  • thermals 6-28-21 2355.png
    thermals 6-28-21 2355.png
    1.4 MB · Views: 0
Reading the Wiki gives useful information, ie its Thermal conductivity = 81.8 W/(m⋅K)
https://en.wikipedia.org/wiki/Indium
But there are potentially serious issues.

Indium oxidises and the oxide is conductive.
It seems improbable there would be enough oxide developed from the edges of a thin film in use on a processor (unless there is a large overhang) but when handling the material enough oxide might get on fingers and could be transferred to electronics.
Invisible conductive Indium films are used on products like LCDs so it could be difficult to locate.

Potentially worse though, both Indium and its oxide are toxic. It is listed as harmful with an attached warning.
It has a biological 1/2 life of 2 weeks in the body.
"People can be exposed to indium in the workplace by inhalation, ingestion, skin contact, and eye contact. Indium lung is a lung disease characterized by pulmonary alveolar proteinosis and pulmonary fibrosis, first described by Japanese researchers in 2003. As of 2010, 10 cases had been described, though more than 100 indium workers had documented respiratory abnormalities.[80] The National Institute for Occupational Safety and Health has set a recommended exposure limit (REL) of 0.1 mg/m3 over an eight-hour workday.[81]"


Guys, take serious precautions when handling, storing and disposing of.
Locate every little off cut and store them away safely.
 
Nenu said:
Reading the Wiki gives useful information, ie its Thermal conductivity = 81.8 W/(m⋅K)
https://en.wikipedia.org/wiki/Indium
But there are potentially serious issues.

Indium oxidises and the oxide is conductive.
It seems improbable there would be enough oxide developed from the edges of a thin film in use on a processor (unless there is a large overhang) but when handling the material enough oxide might get on fingers and could be transferred to electronics.
Invisible conductive Indium films are used on products like LCDs so it could be difficult to locate.

Potentially worse though, both Indium and its oxide are toxic. It is listed as harmful with an attached warning.
It has a biological 1/2 life of 2 weeks in the body.
"People can be exposed to indium in the workplace by inhalation, ingestion, skin contact, and eye contact. Indium lung is a lung disease characterized by pulmonary alveolar proteinosis and pulmonary fibrosis, first described by Japanese researchers in 2003. As of 2010, 10 cases had been described, though more than 100 indium workers had documented respiratory abnormalities.[80] The National Institute for Occupational Safety and Health has set a recommended exposure limit (REL) of 0.1 mg/m3 over an eight-hour workday.[81]"


Guys, take serious precautions when handling, storing and disposing of.
Locate every little off cut and store them away safely.
Click to expand...
The fumes would be a real concern with the heating and cooling cycle of the CPU, and home PCs are not air tight.
 
Actually, I found this on Indium lung:

"The production of indium-tin oxide has increased in the past decades due to the increased manufacture of liquid crystal displays (LCD). Taiwan is one of the highest indium-consuming countries worldwide. After repeated inhalation, indium oxide (In2O3) particles would accumulate in the lungs, resulting in severe lung effects. We report two workers of an LCD producing facility with elevated serum indium level up to 149 and 73.8 μg/L (normal value <3.5 μg/L), which was much higher than that observed in previous case reports in Taiwan. We collected their detailed working history, symptoms, pulmonary function, radiologic findings, and followed up for more than one year. We also performed workplace evaluation of the facility. We observed that sandblasters who clean components of ITO thin-film production machinery by sandblasting with aluminum oxide tend to have higher indium exposure with worse pulmonary functions and HRCT findings."

https://www.sciencedirect.com/science/article/pii/S0929664620303831

Please don't sandblast your indium.
 
lol
Some people like to smell new things, a no nose warning sign might be appropriate.
Hey maybe a not fully decayed version of Indium is behind Supermans laser vision?
 
Nenu said:
Reading the Wiki gives useful information, ie its Thermal conductivity = 81.8 W/(m⋅K)
https://en.wikipedia.org/wiki/Indium
But there are potentially serious issues.

Indium oxidises and the oxide is conductive.
It seems improbable there would be enough oxide developed from the edges of a thin film in use on a processor (unless there is a large overhang) but when handling the material enough oxide might get on fingers and could be transferred to electronics.
Invisible conductive Indium films are used on products like LCDs so it could be difficult to locate.

Potentially worse though, both Indium and its oxide are toxic. It is listed as harmful with an attached warning.
It has a biological 1/2 life of 2 weeks in the body.
"People can be exposed to indium in the workplace by inhalation, ingestion, skin contact, and eye contact. Indium lung is a lung disease characterized by pulmonary alveolar proteinosis and pulmonary fibrosis, first described by Japanese researchers in 2003. As of 2010, 10 cases had been described, though more than 100 indium workers had documented respiratory abnormalities.[80] The National Institute for Occupational Safety and Health has set a recommended exposure limit (REL) of 0.1 mg/m3 over an eight-hour workday.[81]"


Guys, take serious precautions when handling, storing and disposing of.
Locate every little off cut and store them away safely.
Click to expand...
Ok, so I read the same thing thing and got the impression this isn't too bad. I don't know how many people work with Indium, but 10 cases in sevel years (and 100 documented abnormalities) doesn't seem that high, and that's for people who work with it every day. If you take reasonable precautions (don't lick the foil, wash your hands promptly after handling and maybe wear gloves, which you might want to do anyway to avoid finger oils on your processor, and you handle the stuff maybe a couple times a year, it doesn't seem worth worrying about. A materials safety data sheet doesn't look too bad either.
 
Armenius said:
The fumes would be a real concern with the heating and cooling cycle of the CPU, and home PCs are not air tight.
Click to expand...
Because it will be used well below the melting point and only a tiny tiny edge will peek out from between the cooler/cpu, I'm not sure this will be a problem.
 
toast0 said:
Ok, so I read the same thing thing and got the impression this isn't too bad. I don't know how many people work with Indium, but 10 cases in sevel years (and 100 documented abnormalities) doesn't seem that high, and that's for people who work with it every day. If you take reasonable precautions (don't lick the foil, wash your hands promptly after handling and maybe wear gloves, which you might want to do anyway to avoid finger oils on your processor, and you handle the stuff maybe a couple times a year, it doesn't seem worth worrying about. A materials safety data sheet doesn't look too bad either.
Click to expand...
Yeah, my take on it is to wear gloves and wash them straight after.
Employ good storage practices.
Oh and dont breath ;)
 
jesus, that warning stuff is for people working with and handling it every day, day in, day out. relax....

1624990638118.png
 
pendragon1 said:
jesus, that warning stuff is for people working with and handling it every day, day in, day out. relax....

View attachment 370385
Click to expand...
We are relaxed, this is a competent discussion.
The information needed publishing so it will be handled appropriately.
If a small amount falls on the floor and somehow is ingested by a pet or child, it could be unpleasant.
Also its probably sensible to wear gloves and to keep them clean.
Some industrial risks apply to those handling and cutting, it would be foolhardy not to take care.
 
Nenu said:
We are relaxed, this is a competent discussion.
The information needed publishing so it will be handled appropriately.
If a small amount falls on the floor and somehow is ingested by a pet or child, it could be unpleasant.
Also its probably sensible to wear gloves and to keep them clean.
Some industrial risks apply to those handling and cutting, it would be foolhardy not to take care.
Click to expand...
Exactly. I'd rather be on the careful side handling material that even just has a minute possibility of toxicity from handling it.
 
  • Like
Reactions: Nenu
like this
Results from today. 27.1C ambient. Looks to be running just a bit warmer today, but that could also be within the margin of error of my ambient temp thermometer. Water temp and VRM temp are also up a notch, so maybe that's it. First test is the usual Prime 26.6, second test is Prime 30.4 with AXV2+FMA3 enabled.
 

Attachments

  • thermals 6-29-21 1642.png
    thermals 6-29-21 1642.png
    1.3 MB · Views: 0
  • thermals 6-29-21 1650 AVX.png
    thermals 6-29-21 1650 AVX.png
    1.4 MB · Views: 0
If water temp is up too, I wouldn't worry yet. If temp rises and ambient/water temp drops, something's wrong (or your pushing your PC harder than normal after a low-demand period).
 
Yeah water temp is up 0.3, average core temp is up 0.5. The hottest core is up 0.9 which is something I'll have to keep an eye on. So far this is just margin of error stuff. Fingers crossed.
 
Aside from the conductivity issues, i'm curious how this would do on a smaller scale - like a chipset or NVME controller.
 
TheSlySyl said:
Aside from the conductivity issues, i'm curious how this would do on a smaller scale - like a chipset or NVME controller.
Click to expand...

The only caveat besides the conductivity issues is that it needs strong mounting pressure to work. I believe 100 lbs is considered ideal for this material, but I would imagine that I'm getting by with a bit less that that right now. I think 30-60 lbs is typical on a modern CPU, and though there are ways of measuring that fairly easily, I haven't gone that far. I do know that if I mount the block too tightly, I start losing DIMM slots, so I have to stay within certain limits. Hand tight on the screws plus a quarter turn with a philips head seems to work.

I think this stuff could be interesting for the people worried about GDDR6X temps, but someone would have to figure out the right thickness. This stuff has a lot more give than a copper shim, but not nearly as much as a silicone-based thermal pad or thermal putty.
 
Last edited:
OK, ran P95 26.6 again, and the hit I'm taking on Core 2 seems repeatable. Hopefully that trend doesn't continue. Here's a table of core delta vs water temperature, which should relate to the effectiveness of the TIM most directly

June 27June 28June 29
Core 020.721.021.5
Core 121.720.420.8
Core 226.726.827.6
Core 324.023.923.8
Core 423.823.623.3
Core 523.423.323.4
Core 624.722.623.5
Core 721.421.821.7
Core Mean23.322.923.3
 

Attachments

  • thermals 6-29-21 1750.png
    thermals 6-29-21 1750.png
    1.3 MB · Views: 0
Its great you have gone this far but you must record ambient temp each time for the figures to mean something.
Also repeat conditions that can affect heat transfer in the room. Like whether the room door or windows are open / closed.
Water loops are slow to change, it can take longer to get a viable result otherwise which might cause a result to be taken too early.
 
You must log in or register to reply here.
Back
Top