In order for this test to be comparative to other you need to find a way to hold the gap between 2 copper slugs constant (spacer or holder for upper slug). Every material has diffrent heat conductivity and test results depend on layer thickness. This setup is close o professional thermal interface test. only difference s that professional test measures temperature of both slugs near test sample and can automatically plot temp-conductance curve. Edit: I saw other video where you explain ball bearings spacers. This is the best thermal interface test I have seen on youtube apart from professional tester.
Heh, just saw your edit. Was going to mention the ball bearings. It the best test I could think of on a limited budget. The weaknesses of my test setup, in my opinion, are: -Not measuring ambient temperature and humidity -amazon temperature sensor....and only having the single sensor. -not monitoring watts/kwh that the hotplate is using.
Forgive me if this is not related but: When the temperature goes UP, the thing seems to slightly be pushed outwards. Is there a chance you applied too much of it? I am asking because the temperature seems to be going up much faster once the puddy is squezzed out.
Good observation on it pushing out. I apply the same amount for every hotplate test, which is to fill the gap and I wipe up any excess before the test starts. This gives each putty a perfect disc measuring 1.90mm thick. As for why this Putty squeezed out more than others, it is gars for me to know. I just know that it did. One that pushed out even more was the Silicone caulking test, which was kind of boiling or offgassing. I don't recommend using silicone caulking.
@@snarksdomain So far I am using Gelid GP ultimates and they seem to do the work just fine (laptop, 145W gpu, 94W CPU at max turbo, vapor chamber cooling, Asus advantage edition). I was considering changing them for K5 due to the fact that I had to change Liquid Metal for Kryosheets due to bad contact pressure between cooler and cpu/gpu after changeing the thermal puddy to pads. With 1.0mm pads the laptop would shut down, with 0.5mm pads my temps were fine but gpu performance dropped by 30%.
Thanks for your vote of confidence and also for your subscription. I should note that I'll be testing out K5 Pro on my test card anyway since I've received a lot of flack from CSGR regarding the hotplate test. I should also note that I'm making improvements to my test methodology and equipment. This hotplate test should be viewed as a reference, but not my final verdict. Feel free to check out my stream later tonight, where I will be applying Lipoly SH-Putty3 to the test card. Stream usually runs for ~1 hour.
So i was thinking of buying some jeyi thermal putty should I go with it I’m not really to sure I did watch your video and I’m thinking about buying 100g of the thermal putty since it’s the closest to TG-pp10
*Update: I was wrong about the density of TG-PP10. It is somewhere between 3 - 3.2 g/cc. I kept looking at the 1kg container and realizing something was 'off'. There was just no way theybhad fit 833ml (cc) into it as the tub comes around 1/3 to 1/2 full. I emailed T-Global and they responded back that it was a clerical error and 5hat it was around 3, and very likely 3.2 g/cc. This makes the comparative cost of Jeyi a better than I thought previously. Old info: So this stuff has a higher density than TG-PP10. Jeyi is 3.2g/cc, TG-PP10 is 1.2g/cc. On a 3090 where I installed copper shims I used 51 grams of TG-PP10. With the higher density of the Jeyi I would have needed 136 grams to do the job (14x10g bottles). With shipping that's like $120 of Jeyi to do the job of $15 of TG-PP10. If we can find a place to buy this stuff in bulk then it might be a viable option. I just weighed the amount in a Jeyi container and it was at least 10 grams. I should pack the same container full of Jeyi and then with TG-PP10 and weigh them both to confirm the difference in density.
Hey Nathenial, if you're looking for Jeyi putty also check this link out. I can't be 100% certain but I believe this is the same product and almost half price: a.aliexpress.com/_mOVUSCO
@nator1654 It's hard to say. I'll need to get some TP3 pads. Competition will be stiff at the upper end. The real trick is to have high thermal performance, and still be soft e ought to account for different pads thicknesses required. Thermal Putties have the clear advantage for varying component heights and compressibility. Putties also lend themselves well to shim applications. Either way, I hope we end up with a selection of high performing options. That way each person can decide for themselves what they'd like to use.
I'll likely need to buy some. I've been told I can expect very similar results as the Jeyi 8100 putty I've tested. It's also possible a sample has been sent already, I just haven't received it yet.
For sure. Lately I've been hunting for a good alternative to TG-PP10 as it is now obsolete (but still limited quantities of inventory at Digikey). After testing on the hot plate and a 3070 Ti I really like Penchem TH949-1, followed by Jeyi 8100, Penchem TH855-5, and TH930 (same performance as TG-PP10). ua-cam.com/video/1vMv20rpte8/v-deo.html ua-cam.com/video/zSX_7P0HZqc/v-deo.html
The Upsiren 3D Graphite beats all the other pads I tested. I'd still go with Upsiren UX Pro Putty as it seems to even match or outperform the 3D Graphite pads. Updated chart at the end of this video: ua-cam.com/video/raHKGnJ4LYM/v-deo.html
@Silva Teixeira yep. So if you watch some of my testing videos you'll see I'm using an RTX 3070 Ti with 8 GB of GDDR6X. I've also been using putty on 3080's, 3090's and also on cards with GDDR6 (3070, 3060ti, 6800XT)
Yep. I've got 4 tubes of it and will be testing it when I get a chance to do so. I had considered testing some I pulled off EVGA card but decided it would be best to test new stuff to get most accurate results. Work is very busy currently so it's hard to find time for testing but I'll make it happen at some point.
@@hurricane5952 Okay. Just tried it out on the hotplate and the results are looking good. Going to run 2 more tests for confirmation. I'll work on the video shortly here. I'll also test on the 3070 ti in the same manner as my other tests and see how it shakes out. DO NOT throw out your EVGA putty if you open you card. From what I can see so far it tests in the high end of putties, not the low end.
K5 Pro is absolute garbage, I did try this out on an RTX 3090 once and results are worse than generic 3W/mK pads. K5 Pro is also very hard to remove, I had to use lots of white spirit and later IPA to remove it.
Yeah, it's definitely less viscous than TG-PP10. I haven't pulled the copper blocks apart yet as I want to do a confirmation run. Will see how easily it cleans up after 2 cycles.
I learned more about K5 Pro in 2 minutes and 46 seconds from you than I have from any other source. Thanks for this!
Glad it was helpful!
In order for this test to be comparative to other you need to find a way to hold the gap between 2 copper slugs constant (spacer or holder for upper slug). Every material has diffrent heat conductivity and test results depend on layer thickness. This setup is close o professional thermal interface test. only difference s that professional test measures temperature of both slugs near test sample and can automatically plot temp-conductance curve.
Edit: I saw other video where you explain ball bearings spacers. This is the best thermal interface test I have seen on youtube apart from professional tester.
Heh, just saw your edit. Was going to mention the ball bearings. It the best test I could think of on a limited budget.
The weaknesses of my test setup, in my opinion, are:
-Not measuring ambient temperature and humidity
-amazon temperature sensor....and only having the single sensor.
-not monitoring watts/kwh that the hotplate is using.
Forgive me if this is not related but:
When the temperature goes UP, the thing seems to slightly be pushed outwards.
Is there a chance you applied too much of it?
I am asking because the temperature seems to be going up much faster once the puddy is squezzed out.
Good observation on it pushing out. I apply the same amount for every hotplate test, which is to fill the gap and I wipe up any excess before the test starts. This gives each putty a perfect disc measuring 1.90mm thick.
As for why this Putty squeezed out more than others, it is gars for me to know. I just know that it did. One that pushed out even more was the Silicone caulking test, which was kind of boiling or offgassing. I don't recommend using silicone caulking.
@@snarksdomain So far I am using Gelid GP ultimates and they seem to do the work just fine (laptop, 145W gpu, 94W CPU at max turbo, vapor chamber cooling, Asus advantage edition).
I was considering changing them for K5 due to the fact that I had to change Liquid Metal for Kryosheets due to bad contact pressure between cooler and cpu/gpu after changeing the thermal puddy to pads.
With 1.0mm pads the laptop would shut down, with 0.5mm pads my temps were fine but gpu performance dropped by 30%.
Thank you for being infinitely more informative than LTT
Thanks for your vote of confidence and also for your subscription.
I should note that I'll be testing out K5 Pro on my test card anyway since I've received a lot of flack from CSGR regarding the hotplate test.
I should also note that I'm making improvements to my test methodology and equipment.
This hotplate test should be viewed as a reference, but not my final verdict.
Feel free to check out my stream later tonight, where I will be applying Lipoly SH-Putty3 to the test card. Stream usually runs for ~1 hour.
So i was thinking of buying some jeyi thermal putty should I go with it I’m not really to sure I did watch your video and I’m thinking about buying 100g of the thermal putty since it’s the closest to TG-pp10
*Update: I was wrong about the density of TG-PP10. It is somewhere between 3 - 3.2 g/cc. I kept looking at the 1kg container and realizing something was 'off'. There was just no way theybhad fit 833ml (cc) into it as the tub comes around 1/3 to 1/2 full. I emailed T-Global and they responded back that it was a clerical error and 5hat it was around 3, and very likely 3.2 g/cc. This makes the comparative cost of Jeyi a better than I thought previously.
Old info: So this stuff has a higher density than TG-PP10. Jeyi is 3.2g/cc, TG-PP10 is 1.2g/cc.
On a 3090 where I installed copper shims I used 51 grams of TG-PP10. With the higher density of the Jeyi I would have needed 136 grams to do the job (14x10g bottles). With shipping that's like $120 of Jeyi to do the job of $15 of TG-PP10. If we can find a place to buy this stuff in bulk then it might be a viable option.
I just weighed the amount in a Jeyi container and it was at least 10 grams. I should pack the same container full of Jeyi and then with TG-PP10 and weigh them both to confirm the difference in density.
Hey Nathenial, if you're looking for Jeyi putty also check this link out. I can't be 100% certain but I believe this is the same product and almost half price: a.aliexpress.com/_mOVUSCO
What is the best thermal putty you would suggest?
Currently the best performing putties are:
EVGA Putty
Upsiren UX Pro
Penchem TH949-1
Upsiren U6 Pro
@@snarksdomain do you think these thermal putty can outperform the arctic tp3 thermal pad?
@nator1654 It's hard to say. I'll need to get some TP3 pads. Competition will be stiff at the upper end.
The real trick is to have high thermal performance, and still be soft e ought to account for different pads thicknesses required. Thermal Putties have the clear advantage for varying component heights and compressibility.
Putties also lend themselves well to shim applications.
Either way, I hope we end up with a selection of high performing options. That way each person can decide for themselves what they'd like to use.
What about Upsiren U6 Pro 12 w/mk? Can you test it ?
I'll likely need to buy some. I've been told I can expect very similar results as the Jeyi 8100 putty I've tested. It's also possible a sample has been sent already, I just haven't received it yet.
Is there a Viscous solution your recommend over K5 pro?
For sure. Lately I've been hunting for a good alternative to TG-PP10 as it is now obsolete (but still limited quantities of inventory at Digikey).
After testing on the hot plate and a 3070 Ti I really like Penchem TH949-1, followed by Jeyi 8100, Penchem TH855-5, and TH930 (same performance as TG-PP10).
ua-cam.com/video/1vMv20rpte8/v-deo.html
ua-cam.com/video/zSX_7P0HZqc/v-deo.html
what is the best thermal pad?
The Upsiren 3D Graphite beats all the other pads I tested. I'd still go with Upsiren UX Pro Putty as it seems to even match or outperform the 3D Graphite pads.
Updated chart at the end of this video:
ua-cam.com/video/raHKGnJ4LYM/v-deo.html
@@snarksdomain GDDR6?
@Silva Teixeira yep. So if you watch some of my testing videos you'll see I'm using an RTX 3070 Ti with 8 GB of GDDR6X. I've also been using putty on 3080's, 3090's and also on cards with GDDR6 (3070, 3060ti, 6800XT)
@@snarksdomain thank you
Can you test Evga PUTTY PLS?
Yep. I've got 4 tubes of it and will be testing it when I get a chance to do so.
I had considered testing some I pulled off EVGA card but decided it would be best to test new stuff to get most accurate results.
Work is very busy currently so it's hard to find time for testing but I'll make it happen at some point.
@@snarksdomain how does it compare to other putty’s?can you share some data.Dont need a video necessarily
@@hurricane5952 Okay. Just tried it out on the hotplate and the results are looking good. Going to run 2 more tests for confirmation. I'll work on the video shortly here. I'll also test on the 3070 ti in the same manner as my other tests and see how it shakes out.
DO NOT throw out your EVGA putty if you open you card. From what I can see so far it tests in the high end of putties, not the low end.
K5 Pro is absolute garbage, I did try this out on an RTX 3090 once and results are worse than generic 3W/mK pads.
K5 Pro is also very hard to remove, I had to use lots of white spirit and later IPA to remove it.
Yeah, it's definitely less viscous than TG-PP10.
I haven't pulled the copper blocks apart yet as I want to do a confirmation run. Will see how easily it cleans up after 2 cycles.