[Honeywell PTM7950 Phase Change Thermal Pads/Sheets] Application, Tips, and Results

My current Geekbench score:

That puts the Framework laptop in first equal place with the MSI Summit E15 for single core score. …and 3rd place in the multicore score. All that from a single test score entry (whereas other top multicore score entries have lower single core scores). I’m happy with the performance as it stands at the moment.

Think it’ll be a couple of months before the scores are indexed.

Going to wait for that Canadian winter to come…then maybe I can get into first place.

4 Likes

Just an FYI, looks like LTT saw your post and made a video testing PTM7950 out of it: Reddit told me to buy this – PTM7950 Thermal Pad - YouTube

5 Likes

Oh cool, thanks @lhl, super interesting video!!

edit: (and yay I made it onto an LTT video, been watching for like a decade :grin:)

4 Likes

Just wanted to send many thanks for creating this thread. I’m working on applying PTM7950 soon for the first time and really appreciate the tips and experiences you’ve shared. Also noticed the amount of time it must have took to make this thread as well as formatting it.

Sincerely, an internet stranger.

-RB

4 Likes

Thanks for the kind words @RB0617. I really appreciate it – it’s nice to know that my efforts are helping others and can play a part in keeping the DIY spirit alive. I’m hoping your PTM7950 application ends up being worth it!

5 Likes

Hi Michael,

I ordered PTM 7950 some from ebuy7, but they don’t look right. For whatever reason they have these dark solid specks in the pad. They are not a uniform grey.

Not sure if I’m allowed to post a imgur link but here goes: Imgur: The magic of the Internet

Do these pads look legit to you?

Appreciate the help.

Thanks,

@John_Chu I’ve never seen that before but it looks like they might just be a bunch of air bubbles. If you peel back the adhesive and they disappear, I think it’s fine. Otherwise, the material looks pretty much the same to me.

That being said, it’s hard to tell if they’re legitimate without applying it and testing the results or some chemical composition test.

@Michael_Wu Hi Michael, thanks for replying back. I didn’t think of air bubbles, interesting thought. I tried putting the pads in the fridge to see if the specks go away if it was due to temperature but no change. Then I tried to smooth out the specks and see if they disappear similar to smoothing out the air bubbles on a screen protector but also no change. I guess I’ll have to just bite the bullet and see if it works. ebuy7 hasn’t been super helpful unfortunately. Thanks again for the help!

@John_Chu Sure thing! :confused: I recommend having backup thermal paste/material to revert to if things go poorly. If you’re not returning the material back to ebuy7 (which I’m not sure they will even accept), might as well peel back the film and have a look at what the specks are. If they’re actual debris, I’d caution against using it as that shouldn’t be present in thermal paste/material. If you still want to test it out, I think as long as it’s not some abrasive/hard material then it’s probably fine (as it can scratch the die/heatsink). But this is just IMO/IME. Good luck and keep us posted if you can!

1 Like

@Michael_Wu Yeah absolutely. I’m interested to see myself, if its genuine. I was able to get a reply for ebuy7 and they said the specks in the pad were due to melting in transit. Not sure I believe that explanation to be honest. The pad would have to get over 45c to even start the phase change. Well, either way we are going to find out if its real or fake.

2 Likes

Does anyone with an i5-1240P have the dimensions of the CPU & GPU die heatsink areas on hand?

I’m keen to give this a shot, but want to make sure I have enough pad material on hand before pulling it apart!

I just applied PTM7950 on my i5-1240P and the results are absolutely amazing! :astonished:
I got about 10 degrees difference (from 100 and throttling to 90 and no throttling)


With the stock cooling paste I had a lot of thermal throttling (look at the red line)


After my (not very clean) application of PTM7950, the CPU only throttled a little at the beginning (the material was probably not fully liquified by then) and it was about 10 degrees cooler.


@Angus_Gratton
I forgot to take a picture/measures of the 1240p die, but I used about 1,2 cm * 3 cm of the pad. So even the smallest sizes available (4*4 cm) are big enough for multiple applications.

9 Likes

Just bought some 0.2mm and tried it on two laptops.

On one I got a 10 degree average drop and the other a 14 degree average drop. Both were using Kryonaut paste before.

8 Likes

I don’t think I’ve seen anyone mention this, but it looks like Amazon has these now. Sold by a third party seller called JOYJOM but reviews seem to say it’s legit… Shipped by Amazon, so I’m confident enough to give it a whirl on my 11th gen. I’ll see if my results line up with what’s described here!

US - https://www.amazon.com/dp/B0BRJB8JNX
CA - https://www.amazon.ca/dp/B0BRJB8JNX
UK - https://www.amazon.co.uk/dp/B0BRJB8JNX
(Probably more, but that’s all I’ve bothered to check :slight_smile: )

7 Likes

Thanks! I appreciate the post, I’ll pick up some at some point I suppose :slight_smile:

Hi everyone! Back again to report that the Amazon stuff arrived today and works pretty well!

Here’s the test environment details:
Intel Core i5 1135G7
16GB DDR4 3200
Windows 11 Pro
1TB Crucial P3
100% battery with 65W AC adapter.

Tests performed on a marble surface with 3 passes of the same benchmark (3DMark Night Raid) each. Previous TIM was Arctic MX-6.

With the MX-6, the CPU often dropped from full boost as shown in this particular run:


In fact, there are some cases where the CPU hit almost 95 degrees celcius and disasterously throttled to 500 MHz!! Unfortunately, I misplaced the temp graph for this but the frequency graph really speaks for itself.

With the PTM7950 the CPU frequencies smooth out and I haven’t seen any huge throttling, however it is still there in the end of the CPU specific test.



Temperatures peak at around 85 degrees rather than the low 90s like with the MX-6.

(EDIT: Hmm, looking at this again it looks like this doesn’t even boost past 4 gigahertz like before! This is odd, but it might just be a one off thing. I’ll run a few more tests!)

All in all I’d say this stuff is pretty legit! I have to say though with my regular workflow I would never really get the original TIM to thermal throttle, but it is nice to know I have some of the best stuff out there in my computer. It is quite expensive, so while it’s nice to do I don’t really think it’s a must have. It’s not like I’m always running benchmarks on my laptop or anything :stuck_out_tongue:

For those who have been waiting for a more reliable source before taking the plunge, I can vouch for the product I received on Amazon, would be great if you’re constantly doing CPU intensive tasks on higher end laptops where thermal throttling and noise is a real issue!

EDIT:

After a couple of heat cycles, I decided to go for an AIDA 64 stability test as others have done above. Looks like the results I’m getting line up pretty well here, with just a little bit of throttling at the beginning.

It’s unfortunate that I didn’t have results from the MX-6 thermal paste I had in there before, but I seem to recall people mentioning that repasting over the stock paste didn’t make much of a difference, so I assume there would have been a fair amount of throttling there in that case. (If another 11th gen i5 user can comment on this one, that would be great!!)

Overall my conclusion doesn’t change, this is some pretty potent stuff! I’m gonna be putting it in everything now, heh.

5 Likes

This is a 10 minute soak test I did the day I fitted mine -

I would point out the next day I did it again and the CPU Throttling was green throughout.

1 Like

Another 15 minute run I did just now.


Not bad on GPU either…

Hey you are the only person who Ive seen thoroughly test these. I was wondering if you can test this one on your system so we have comparable results

It works pretty well on mine, but this could just be because the temps across all cores is more even, and its still not as good as it could be. (Using hardwareinfo there was a core to core delta of 13 C before and 4 C after)