Ideally, you'd stencil it on the cooler cold plate in a thin, flat, layer slightly larger than the die, let it dry for at least 15 hours, then mount it to the card. Assuming there are no problems, run a stress test for a while that keeps the GPU edge temp above 60C for at least 30 minutes.

I'd make the stencil out of kapton or residue free masking tape, though if the cooler/block has a raised area for the core, you may have to omit it and manually spread.

 

Also, I live in a humid and tropical climate. Its quite rainy. Also daily temperatures this season range from 24C-31C.
Wouldnt this slow down the drying/curing process?
How would this affect drying/cure time?
Would you advise that i let it dry/cure for atleast 24 hours?

 

I have 7958-SP on my 4090 and 14900K currently.

I didn't do anything different than regular paste. It's difficult to spread, but I warmed it slightly by putting the tube in some warm water, and then applied like standard thermal paste. Applied to the 4090 core, and the 14900K IHS, and spread. Put a dab on the cooler cold-plate side, and mounted. Performed a few heat cycles to let it warm/spread.

Still maintaining identical delta's as day-1 install.

 

They also tried skipping the curing phase on Snarks Domain on YouTube. They sacrificed about 2-3 degrees from what I recall.
Nice to know you have same deltas from day-1, I am looking forward to using this paste.

 

I didn’t get the syringe. I got sheets. It’s doing what paste does and is of a similar consistency in your case so I would assume like it’s regular thermal paste. It’s a biphasic though, it seems to have more adhesion than cohesion, like oil vs mercury (which is low adhesion high cohesion). How thick are your thermal pads? If they’re thin enough you can replace them with it and it won’t just drip out. I’m not sure what that limit is though it’s more than 1mm as it worked for me And my thermal pads were 1mm

 

PTM7958 is only available in paste form unfortunately.
My pads are 1mm.
I plan on using the PTM7958 on GPU and CPU core, then Upsiren UTP-8 inplace of GPU Thermal Pads.

 

Also, I live in a humid and tropical climate. Its quite rainy. Also daily temperatures this season range from 24C-31C.
Wouldnt this slow down the drying/curing process?
How would this affect drying/cure time?
Would you advise that i let it dry/cure for atleast 24 hours?

I'm sure the TIM will absorb some small amount of water from the air, and will absorb more at higher relative humidity, but the solvent that's evaporating here isn't water and the setup time shouldn't be affected much. Main factors will be how thick of a layer you apply and how warm it is.

"Cure time" for this stuff is based on cycles, not simply a function of time. You need repeated heating and cooling fazes over at least a few days in order for this stuff to "cure". Just apply it and use your computer as normal. Your temps will be a bit high initially, but they will continue to come down until such time that it fully "cures".

tl;dr Be patient and use your computer.

BlueBase is not talking about the post application cure time, but the time for the solvent in the paste version of PTM to evaporate/outgas.

The SP versions of this stuff are only liquid because they are thinned with a solvent to reduce viscosity. The application instructions state that this should be given sufficient time to dry out before assembly. acoustic's experience suggest that this process can possibly be abbreviated or skipped, but if one wants to play it safe, the instructions are not hard to follow.

 

I'm sure the TIM will absorb some small amount of water from the air, and will absorb more at higher relative humidity, but the solvent that's evaporating here isn't water and the setup time shouldn't be affected much. Main factors will be how thick of a layer you apply and how warm it is.



BlueBase is not talking about the post application cure time, but the time for the solvent in the paste version of PTM to evaporate/outgas.

The SP versions of this stuff are only liquid because they are thinned with a solvent to reduce viscosity. The application instructions state that this should be given sufficient time to dry out before assembly. acoustic's experience suggest that this process can possibly be abbreviated or skipped, but if one wants to play it safe, the instructions are not hard to follow.

After a basic google search, a company called Caplinq says that drying time for PTM7950-SP (Not 7958) is 24 hours at room temperature.

PTM7950-SP | The most thermally conductive stencil printable Phase Change Paste | PTM7950SP

PTM7950SP: Ultra-conductive phase change paste for thermal interface applications. Minimizes resistance, excels in reliability tests, & offers cost-effective application. Stencil printing compatible (down to 0.1mm), but requires drying step (24hrs or 5min @ 100°C). Ideal for power modules...

www.caplinq.com

Do you think there are advantages to drying PTM7958-SP for 24 hours? or would you recommend staying with the 15 hours?

Also, Snarks Domain on YouTube tried applying the paste without the drying phase, and sacrificed about 2-3 degrees. I would have preferred to keep the 2-3 degrees though.

 

That is crazy. I would not even bother with the stuff.

 

Depends on your use case. It’s not as good as Liquid Metal, but it’s better than everything else. Liquid Metal has some issues though, so if you’re not sure of what things are made of, or you are but it’s the wrong thing its the best stuff. me, I wasn’t sure.

 

Snark's PTM7958SP looks a lot more wet in his video than the syringe I have.

I'm not sure about this whole curing phase. Admittedly, I didn't look up instructions for it -- in my mind, it's a paste, so I handled it just like I would KPx. Might have been the wrong move, though.

For the GPU, I spent quite some time after applying/spreading the paste before putting the card together as I was doing some clean-up from my first time messing with thermal putty. Maybe that helped slightly?

 

After a basic google search, a company called Caplinq says that drying time for PTM7950-SP (Not 7958) is 24 hours at room temperature.

I may have been looking at the TDS for 7950SP. Do whatever the instructions for the product you have says to do, within reason. Most end-users won't have custom laser-cut stencils, so a more simplistic tape square, or a much less precise manual spread will have to suffice. Applying heat (don't blow directly at the TIM, heat the object from the opposite side, use IR, or put in a higher temp ambient) is almost certainly an option to dry it faster.

Also, Snarks Domain on YouTube tried applying the paste without the drying phase, and sacrificed about 2-3 degrees. I would have preferred to keep the 2-3 degrees though.

Interesting, and not surprising. The thermally conductive component of any paste is the filler...the binders or solvents that make them liquid are only there to make sure they can be applied and that they fill small enough voids. A phase change material doesn't need or benefit from the later as it's phase change properties take care of that.

Snark's PTM7958SP looks a lot more wet in his video than the syringe I have.

I'm not sure about this whole curing phase. Admittedly, I didn't look up instructions for it -- in my mind, it's a paste, so I handled it just like I would KPx. Might have been the wrong move, though.

For the GPU, I spent quite some time after applying/spreading the paste before putting the card together as I was doing some clean-up from my first time messing with thermal putty. Maybe that helped slightly?

It's not a normal paste, it's the paste form of a phase change material. If it's solid phase wasn't solid/dry most of the benefits would be lost. And if those benefits were possible in normal pastes without downsides like drying time and more limited storage life, normal pastes would all be like this.

https://www.caplinq.com/index.php?option=com_filecabinet&cid[0]=941&lang=en&productId=3481&task=download&tmpl=component

The material is formulated to be stenciled on with a squeegee, so is quite low viscosity when fresh, and recommended pot life for the paste, once opened, is very low. It's very possible that the tube you had was older, was stored differently, and didn't need as much time for what was left of the solvent to evaporate. I'm not even sure tube-size formats are officially supported by Honeywell and probably imply quite a bit of solvent loss simply in the process of someone pouring them into a tube.

The remaining solvent would surely evaporate and escape during thermal cycling, but might not leave the TIM interface exactly as expected. It's possible you'd see some improvement with a new application. Probably not worth redoing, unless you encounter issues at some point. As always, there is margin for error, but ideally, PCM is dry before mounting the cooler.

Not sure what I got then. I bought it from a Hong Kong company. Seems to work though. Was a sort of blue plastic sheet.

You likely have the standard PTM7950 pads, which are essentially a pre-dried version of the paste. The plastic is just there to hold it together and protect it before application.

There is probably a non-SP version of PTM7958, but as it's a semi-custom version of 7950 made for OEMs, I haven't seen the pad version and the paste version would be more useful to an OEM that is stenciling their own patterns for their own hardware.

 

I may have been looking at the TDS for 7950SP. Do whatever the instructions for the product you have says to do, within reason. Most end-users won't have custom laser-cut stencils, so a more simplistic tape square, or a much less precise manual spread will have to suffice. Applying heat (don't blow directly at the TIM, heat the object from the opposite side, use IR, or put in a higher temp ambient) is almost certainly an option to dry it faster.



Interesting, and not surprising. The thermally conductive component of any paste is the filler...the binders or solvents that make them liquid are only there to make sure they can be applied and that they fill small enough voids. A phase change material doesn't need or benefit from the later as it's phase change properties take care of that.



It's not a normal paste, it's the paste form of a phase change material. If it's solid phase wasn't solid/dry most of the benefits would be lost. And if those benefits were possible in normal pastes without downsides like drying time and more limited storage life, normal pastes would all be like this.

https://www.caplinq.com/index.php?option=com_filecabinet&cid[0]=941&lang=en&productId=3481&task=download&tmpl=component

The material is formulated to be stenciled on with a squeegee, so is quite low viscosity when fresh, and recommended pot life for the paste, once opened, is very low. It's very possible that the tube you had was older, was stored differently, and didn't need as much time for what was left of the solvent to evaporate. I'm not even sure tube-size formats are officially supported by Honeywell and probably imply quite a bit of solvent loss simply in the process of someone pouring them into a tube.

The remaining solvent would surely evaporate and escape during thermal cycling, but might not leave the TIM interface exactly as expected. It's possible you'd see some improvement with a new application. Probably not worth redoing, unless you encounter issues at some point. As always, there is margin for error, but ideally, PCM is dry before mounting the cooler.



You likely have the standard PTM7950 pads, which are essentially a pre-dried version of the paste. The plastic is just there to hold it together and protect it before application.

There is probably a non-SP version of PTM7958, but as it's a semi-custom version of 7950 made for OEMs, I haven't seen the pad version and the paste version would be more useful to an OEM that is stenciling their own patterns for their own hardware.

I bought my PTM7958-SP from Moddiy. Hopefully, it comes with instructions on how to apply. If it doesnt come with instructions, then I am going to dry for 24 hours just to be absolutely sure its dried.

After applying the paste, would it be beneficial to store the syringe in the refridgerator?
I know heat speeds up the drying of the PTM7958, so placing it the refridgerator should prevent it from drying.
Add to that the increased humidity in the refridgerator, which slows evaporation.
Could the cold and the increased humidity in the refridgerator help the PTM7958-SP in the syringe to maintain its original consistency?

 

I'd like to see a comparison between PCMs like PTM795x and graphene pads like KryoSheet as they seem to have the most similar use cases for enthusiasts.

 

I use Kryosheet in my personal system. I actually just ordered some PTM7950 for my server clusters and network appliances, having never used it before personally. I'll do a comparison in a couple weeks once it arrives and I'll be sure to tag you in it.

 

I'd like to see a comparison between PCMs like PTM795x and graphene pads like KryoSheet as they seem to have the most similar use cases for enthusiasts.

The pads were very meh for a while. They may have improved though

I bought my PTM7958-SP from Moddiy. Hopefully, it comes with instructions on how to apply. If it doesnt come with instructions, then I am going to dry for 24 hours just to be absolutely sure its dried.

After applying the paste, would it be beneficial to store the syringe in the refridgerator?
I know heat speeds up the drying of the PTM7958, so placing it the refridgerator should prevent it from drying.
Add to that the increased humidity in the refridgerator, which slows evaporation.
Could the cold and the increased humidity in the refridgerator help the PTM7958-SP in the syringe to maintain its original consistency?

I thought refrigerators and freezers are both self defrosting now. So They pull water out of the air inside them and deposit it on a pan underneath them. So they both dehumidify. I worry about the fridge Safety wise. Food is kept in it. The move would likely be to place the thing in a glass jar then put it in the fridge to prolong life. Stuff even migrates through glass though it improves matters a lot. Also i don’t know what solvents are in it but it’s possible they’re really aggressive ones like methyl-ethyl ketone and xylene and stuff that you really don’t want getting into stuff you eat. A metal tube in a glass jar wouldn’t hurt either if you happen to have an old cigar tube lying around.
scientists keep refrigerators that say “NOT FOR FOOD” on them for a reason. Usually they smell awful though so its not tempting.

 

The pads were very meh for a while. They may have improved though

I thought refrigerators and freezers are both self defrosting now. So They pull water out of the air inside them and deposit it on a pan underneath them. So they both dehumidify. I worry about the fridge Safety wise. Food is kept in it. The move would likely be to place the thing in a glass jar then put it in the fridge to prolong life. Stuff even migrates through glass though it improves matters a lot. Also i don’t know what solvents are in it but it’s possible they’re really aggressive ones like methyl-ethyl ketone and xylene and stuff that you really don’t want getting into stuff you eat. A metal tube in a glass jar wouldn’t hurt either if you happen to have an old cigar tube lying around.

You're correct. No properly functioning modern refrigerator should be humid. Just as your air conditioner pulls moisture from the air its cooling, a refrigerator does the same.

 

The pads were very meh for a while. They may have improved though

I thought refrigerators and freezers are both self defrosting now. So They pull water out of the air inside them and deposit it on a pan underneath them. So they both dehumidify. I worry about the fridge Safety wise. Food is kept in it. The move would likely be to place the thing in a glass jar then put it in the fridge to prolong life. Stuff even migrates through glass though it improves matters a lot. Also i don’t know what solvents are in it but it’s possible they’re really aggressive ones like methyl-ethyl ketone and xylene and stuff that you really don’t want getting into stuff you eat. A metal tube in a glass jar wouldn’t hurt either if you happen to have an old cigar tube lying around.
scientists keep refrigerators that say “NOT FOR FOOD” on them for a reason. Usually they smell awful though so its not tempting.

You're correct. No properly functioning modern refrigerator should be humid. Just as your air conditioner pulls moisture from the air its cooling, a refrigerator does the same.

We could remove the refridgerator humidity from the equation.
Since temperature is a major influence, then shouldnt the cold temperature help maintain the PTM7958SP syringes original consistency?

 

Refrigerators aren't more humid inside, they are less, even if they don't have defrosters, because cold air holds less moisture. Regardless, ambient humidity is completely irrelevant to a sealed container as it should be air tight.

You should keep anything that's been refrigerated, that you don't want getting wet, sealed until it reaches room temp. This is so water from the air outside the fridge doesn't condense on the material when you open it up, not because of the humidity in the fridge.

Since temperature is a major influence, then shouldnt the cold temperature help maintain the PTM7958SP syringes original consistency?

Lower temp almost always increases shelf-life and the Honeywell documentation says as much.

 

With regards to shelf life, the only Honeywell documentation I could find through a shallow search is this document that claims a 12-month shelf life when stored at 19-24C. It seems to be an extremely well-documented product.

 

Why would you guys get PTM7958 in paste when sheet of PTM 7950 is cleaner and evenly applied works wonders.
When I got the sheet from ModDIY they said its 3 years storage in dark cool temps.

 

Why would you guys get PTM7950 in paste when sheet is cleaner and evenly applied works wonders.
When I got the sheet from ModDIY they said its 3 years storage in dark cool temps.

Because the paste performs slightly better, and I don’t like the sheets.

More product variety is a good thing.

 

I can't find any source for a 15-hour cure time other than a reddit post from a year ago.

 

Honeywell's datasheet says 15 hours at ~25C or 3.5 hours at 50C for a 10-12mil (~0.3mm) thick application for PTM7950-SP.

I suspect 7958 behaves very similarly, but I've never seen the documentation for that specific product.