Foam dam barriers for Liquid Metal safety insurance guide.

When using liquid metal pastes on laptops that are often transported or carried in backpacks often (e.g. for students or travelers), it is a good idea to have free insurance against liquid metal runoff escaping the CPU/heatsink/GPU areas, getting on the PCB and destroying your hardware. Do not allow conductive balls of doom to run amok in your laptops!

You are STILL going to have to deal with possible dislodging or runoff even if you do use Super 33+ tape or decent Kapton tape. Tape won't prevent runoff at all.

Runoff is spare traces of LM balls that escape from the housing, do NOT get trapped on insulated SMD resistors and slide ACROSS the CPU or GPU housing and off onto the motherboard, where it WILL eventually fry something by shorting. That's where barriers come in. A common barrier first used by an airline pilot over here (who takes his laptop with him, and who NEEDS a barrier) is highly compressible foam.

The foam must be VERY porous and not dense at all and *MUST* compress to the width of a human hair, with your fingers, with MINIMAL resistance, with a cutout shape in the shape of the CPU or GPU, with a few millimeters to spare for proper clearance (you don't want the foam blocking the heatsink from touching the cores). THESE DAMS ARE NOT EVER, EVER TO BE USED IN SUBSTITUTION FOR NAIL POLISH INSULATION COATINGS, OR SUPER 33+/KAPTON TAPES, EVER, but in COMBINATION with them.

The job of the foam is to catch any spare LM balls that escape the chips, and block them from going anywhere else. They will sit on the foam with nowhere to go. If you had a bad mount and the temps started skyrocketing, you will have saved your hardware.

Suitable foam is best sourced locally so you can test it in person, rather than trying to find it online. Hardware stores or hobby shops are excellent places for finding foam. These foam materials are often used in packing sensitive objects.

The cutout foam MUST be very thin. On high pressure heatsinks like desktop heatsinks or AIO's, if foam happens to be used, that is not important, just compressible. But laptop heatsinks have atrocious mounting pressure so ANY resistance from foam will decrease resistance even further and not give you desired temps--you want the temps with foam to be identical to without foam. So do your work.

Let's say your foam is 3mm thick and highly compressible.
You MAY get away with this, but to be safe, you should CUT the foam width in half to 1.5mm. This will be ideal. If you're good and OCD, or just plain pro, go for 1mm. Thinner is better. Just don't have it so thin that it breaks apart.

Don't try to cut an 'uncut' (meaning: square shape of foam, that has not had the CPU or GPU layout trimmed from it yet to trim it, that will be too difficult. Instead cut out your shape first, trim the outside to the exact shape of the CPU and GPU housings, and THEN use extremely precise scissors like titanium scissors, and then start working around both the inside and the outside to trim some of the thickness away. Once you start this, you will see why the "cutout" is much easier to trim than an uncut block

To secure the foam dams so they don't shift and move when applying the heatsinks, put a dab of transparent nail polish (the same stuff you used for insulating the SMD resistors) in each corner and secure it.

Left image is original cutout foam dam, 3mm for BGA CPU. Right three images are trimmed 1mm cutout foam dams for BGA CPU and GPU's.
Trying to use 3mm (or thicker) foam requires VERY low density. The thinner the better.

Some substitutions for foam dams can possibly be thin compressed layers of silicone gasket sealer--the same stuff used for relidding IHS's, although this requires preparation--this MUST sit BELOW the level of the CPU or GPU as this otherwise adds resistance! There are ways to prep this, such as @Mr. Fox 's experiment,--and even extremely thick (VERY thick) pastes like that super old Radio shack thermal paste, or partially dried or old Arctic Ceramique (NOT dried out, just very hard to squeeze) will also work as barriers too, when used carefully around the silicon housings, although this will be messy, and a pain to clean up if you have to redo your LM application. A VERY big pain. Feel free to list tested foam dam alternatives which are safe and work well for you.

Note that trying to use foam dams for LGA processors will require extra planning.

Picture: 3mm cutout BGA foam dam, 1mm cutout CPU and 1mm GPU foam dams.

 

Haven't tested it yet, but I was thinking this would work well:

 

too much density I think

the foam that I got from a dell replacement part box was perfect for this purpose, not sure what they're called in the commercial market

 

Thank you for sharing, take care guys, I lost 3 laptops motherboards and 2 GPU's because of sellers shipping laptops to me with LM without any protections, no taps, no silicon, no foam nothing, I have total lost almost $3000, along with wasted more than 7~8 months of repairing, asking for support and ordering parts internationally. all of that because of LM and unprofessional thermal paste for traveling and shipping activities, and I don't like anyone of you to feel what I felt, face what I faced or lose what I lost.

 

It might be too dense, but it seems pretty thin (1/8"). If it is not too expensive to buy for testing (or if someone already has it on hand) it would be worth trying. If the temps are crazy high, it will be an obvious indication it is too dense.

 

@Mr. Fox method works perfect, @Mobius 1 shipped his KM1 to me and he did this method and the laptop arrived without any problems.

 

I already have it. It was like $3 for that bag at an ACE Hardware. It's for sealing, so it's definitely not very porous. It's 3mm in its normal state and can compress down to about .5mm when applying some finger pressure. I would probably leave it at 3mm for sectioning around the IHS, and cut it in half (1.5mm) for placing around the GPU die.

The thicker versions of this in the store were actually much more porous, like the ones in the OP. But they would need to be cut into thinner sections as well.

 

You may end up needing to cut it down for the CPU as well. Even if it feels very compressible between your fingertips. There is not much air space between the CPU retention bracket and heat sink at the hinged end of the bracket. Try it with full thickness first. You will feel how much resistance when tightening down the heat sink. It may feel very soft between your fingers, but when compressed evenly the length of the foam using the heat sink it becomes more resistance to compression. You will be able to tell right away by the temps if it is too stiff. For only $3 it is definitely worth trying. The self-adhesive part would also be nice.

 

Good point.

I was thinking of sticking it on the motherboard around the socket as well. That way the retention bracket is basically sandwiched between two layers of foam.

 

Indeed. Trimming it down to 1mm (width of a CD) completely prevents the foam from interfering with anything at all. Then you have to make sure you don't use too much nail polish, since if that hardens, it can actually be thicker than the foam

 

Yeah I don't think one needs to go overboard on the clear nail varnish. 1 or 2 coats seems to be enough, spread thinly and letting it dry between coats. It doesn't break down easily from what I've seen, and is supposedly temperature resistant up to 120C.

 

I usually require LM to be removed before transport when I buy a system (I know I’m going to end up redoing it myself just to be sure) and when selling systems I usually remove it unless the buyer wants it just to be safe.

Most times I end up using nanogrease except for certain situations on laptops where the temps are just out of bounds or I’m trying to tame the fan noise.

 

Always good advice. Even with foam dam barriers ensuring 100% success of no damage, you can still have high temp issues from LM shifting and leaving hot spots. The foam dams however ensure a 99% chance of no damage or magic smoke from transport.

 

where did that 1 % go from the second to the third sentence?

btw, i went out shopping with the lady last weekend to start off her new balcony greenery project (happy gf/wife, happy life i always say! so totally worth the investment ) and in one of the gardening centers i found a whole pile of foam just lying around waiting to be thrown in the trash. so naturally, i grabbed a large sheet of it right away lets see what i can do with that

 

https://www.amazon.com/uxcell-Singl...rd_wg=YQM6u&psc=1&refRID=7WJFHBQBS9RCPK8ZFE7X

While trying to find a perfect match for the description from Falkentyne's post I stumbled upon this stuff. It seems perfect for the job, it's 1mm thick and shockproof. What do you guys think?

Thanks for leading me here Falkentyne, so far this does seem like the best solution. It seems like the super 33+ tape works as well, but not as good as the foam.

thanks

 

1mm thick means nothing if it isn't compressible with no resistance.
The space between a CPU housing and the CPU slug raised up is like 0.2mm !
If this cannot be compressed to that level with no effort, it will completely ruin heatsink pressure.

 

I've had good luck so far with liberally applying conformal coating and using heat resistant electrical tape around the cpu ang gpu die. Is the foam really necessary?

 

If you plan on carrying the laptop around with you in a backpack or airplane, absolutely YES.
If you use it only as a home desktop replacement and don't ever have it carried or bumped while vertical or sideways, then you can get away with the Super 33+ tape. I would apply a few extra layers however just for safety (reducing the gap around the CPU BGA housing to the areas of the heatsink not touching the CPU is always a good thing...less space means less chances of stray conductive balls of doom roaming around running Amok like Thor the God of Thunder.

 

Best explanation I could have hoped for. lol. Well honestly I'm in no hurry to repeat the hell i went through doing it to my laptop last time. But if I ever have to do another system I'll take this threads advice and use the foam. For now I'll just pray nothing frys in my laptop mid transport.

 

How did it work out? I'm curious because it is inexpensive and easy for anyone to get.

 

I have not tried it yet. I decided not to go the liquid metal route for now.

 

OK. Next time I have to take apart the P870DM-G for something I will go get some and try it. I saw they had some at TrueValue Hardware that is already thin and labeled something like " low density for tiny gaps" and that might work nice. If it does, having one side self-adhesive would be good.

 

Mr. Fox, sorry for the late reply. I wanted to let you know that the foam window seal doesn't work well for LM dam. It's not compressible enough. I ended up picking up this much softer A/C foam weather stripping and it works wonderfully. I cut it cross-section using a pair of Cutco Super Shears scissors and an X-Acto knife, then used the scissors to carve out the middle and trim it further down to about 3mm. Couldn't get it any thinner because the foam would break apart, but it compresses very easily so I don't think it matters at that point.

For the CPU, I placed the dam on the IHS retention bracket. For the GPU, I sealed off around the die using nail polish and Super 33+ electrical tape, then placed the foam on top of that.

I actually forgot to secure the dams using nail polish under the corners as @Falkentyne had suggested . Hope they didn't move when I put the heatsink back.

 

Awesome. Thanks for the update. If your temps are good, don't worry about if the foam GPU dam moved a little. If it moved too much and ended up sandwiched in the wrong spot it would cause contact interference and you won't be seeing good GPU temps.

 

Yup, I got an 8C drop on the GPU after applying Conductonaut on it and doubling up on the paper clips under the retention arms. I also got a 5C drop on the CPU using your shim between the IHS and heatsink sandwiched with Conductonaut. Before LMing everything I was using Prolimatech PK-3 thermal paste. I also flattened out the heat plates and pipes using my hands and a wedge as best I could.

Something interesting I learned is that if a copper surface is too smooth, as was the case with my heatsink and the shim, it's LM-phobic and any attempt to spread the LM just moves the entire ball around. I had to use the gray scouring pad from the CLU package to buff the copper slightly in order for the Conductonaut to spread.

Off-topic, but I found an upgraded version of my heatsink, which is less "unified" than the stock heatsink and has more heatpipes running over the GPU core:

Spoiler: stock vs. upgraded

I'm hoping this drops my GPU temps even further, since the GTX 1080 in this unit runs closer to its thermal limits than the CPU does.

(Again sorry for the O/T)

 

@Mr. Fox or anyone else who knows, is the retention bracket on a desktop CPU socket made of aluminum?

LM on the IHS being so close to it makes me a little worried. When I install my new heatsink, I think I'm gonna cover the retention bracket with Kapton tape.

 

That's a very good question.
I always thought they were made of steel. But knowing how everyone cuts corners these days...that being said, on desktops, you should only use LM between CPU core and IHS anyway.

 

They are made of steel. You can confirm that with a magnet. I have never seen one made of aluminum. I always bridge the air space between the IHS and retention bracket with Kapton tape anyway. Not to protect the bracket, but to have no place for stray liquid metal to go down toward the socket. With the foam dam and Kapton tape it is totally encapsulated, so it does not matter if a bead or drop drips off the side of the IHS.

I think he is referring to his laptop with a desktop CPU, and not a desktop. It make perfect sense to use liquid metal on both sides of the IHS in a laptop due to their chintzy little heat sinks and wimpy fans.

 

Thanks for the info, guys. Really appreciate it. Puts my mind at ease now knowing that it's steel.

@Mr. Fox good observation about the magnet, and I will follow your suggestion of the Kapton tape bridge for next time.

@Falkentyne Mr. Fox is right, this is the laptop in my sig. I'm using a 0.5mm copper shim between the IHS and HS with Conductonaut on both sides, and that netted me a 5C drop compared to normal paste with no shim.

 

It's definitely steel - aluminum doesn't have the 'springiness' to perform this function (I'm an engineer and work in a metal shop).

 

Sorry for replying to an old thread but i'm looking to do this and have some foam that will fit the bill. I can't see it surviving any sort of heat however. These things rest under the heatsink, aren't they going to melt?

 

Not likely. They never melted on me. I did not use any kind of special heat-resistant foam.

 

I have VERY VERY VERY thin extremely porous foam i trimmed down to 0.2mm andit lasted fine through 85C temps. You are fine. Foam will not combust. Even if slightly "melts" slightly (for lack of a better word) it still protects fine. I've tested several kinds of foam and the most thin most compressible "weak" foam works best especially when trimmed to 0.2mm. The foam that does NOT work is 'thick' non compressible foam or any foam that offers resistance to pressing. If you can take a square block of the foam at 3mm thickness and compress it to the width a human hair with your fingers without any resistance, that will work great. But on BGA weak heatsinks, I would trim the foam to 0.2mm.

 

Not my image but this is the same exact model. Am I correct in assuming the foam barrier would go around the heat-resistant rubber pad that is already surrounding the GPU? The foam I have is extremely porous open-celled foam, i'm pretty sure I could shave it down to 0.3mm at least, if not thinner. Sorry for asking so many questions, but I think this is something I want to fully grasp before I even think of attempting it lest I goof everything up permanently.

 

It goes right on top of the pad. What sort of pad is on that thing anyway? Never seen anything like that.

The thinner the foam is, the better. The goal is to 'block' any tiny air spaces that exist between the area 'outside' of the GPU slug, between the GPU housing and the heatsink, where liquid metal can slide along either the black coating thing, or the outside of the heatsink. Even a barrier 0.2mm thick (as long as it offers NO resistance to being compressed to the width of a human hair) will block liquid metal nicely.

 

It feels like either very dense foam or even rubber. It's shorter than the die and I can only guess that it's some sort of thermal barrier for components around or under the die.

 

I had that black plastic around my GPU as well. It doesn't protect the SMD components around the GPU die from LM, which can still flow under it, so just rip it off and apply nail polish or Kapton/Super 33+.

 

This.
This.

Here is what I found out btw from my own tests.
Nail polish results in lower temps (and SHOULD absolutely *SHOULD 100% be applied on any delids to coat any resistors inside the CPU; I think this applies more to HEDT processors as I don't think the 7700K and 8700K have any exposed SMD's, although there are trace marks, and I have no idea if those NEED to be insulated or not), but when Liquid metal is applied on nickel plated copper, you are fine just using nail polish (IHS delids) or nail polish+foam dam on direct die BGA. But if you are using a copper heatsink, then it's best to use Super 33+ tape+foam dams. It's not because of nail polish not working, but rather because of the risk of increased oxidiation if the heatsink contact is not high pressure and has ANY flaws in it at all. A convex heatsink will cause air (even a TINY BIT) to get between the cpu slug and heatsink which will cause increased oxidation, which leads to accelerated absorption of the gallium into the copper. The layer of super 33+ tape actually decreases the amount of air space between the heatsink and the cpu housing area. there is about 0.1mm of space, maybe slightly more but not 0.2mm exactly (seems to be less than 0.2mm of space between the CPU BGA surface and the housing. I don't know if this is the same for LGA chips or not however. Anyway if you use the super 33+ tape + the dams, you decrease the amount of air that can get around the heatsink which is good.

If the heatsink is perfectly flat and not convex and the pressure is semi decent (For BGA Garbage), then you can get by with just nail polish + dams.

 

Liquid metal should improve the temps a little bit, but I think you may still have some thermal challenges with the single pipe heat sink and small radiator. It also appears none of the VRM chips or memory chips have any cooling.

This is on all Clevo Pascal GPUs that I have seen. It is there to provide some protection for the surface-mounted components around the die.

It is just a thin black plastic film. It is not rubber and has no elastic qualities to it.

Yes, I agree. It's a nice idea, but it doesn't stick well enough to work and always seems to lift up around the die. That actually forms a pocket that is likely to trap liquid metal. If not for the fact that it doesn't stick well it would be nice to have.

 

You can also use kneaded eraser to fill up the area around the CPU socket to keep liquid metal out, same as protecting against condensation on chilled systems. This is how I have my CPU socket on my desktop with the water chiller. You can set the foam dam on top of that. Just make sure the kneaded eraser is not taller than the IHS. It will squish down, but you don't want it between the heat sink and IHS because that will interfere with contact.

 

Why is nobody advocating for proper conformal coating? I'd trust it more than nail polish myself.

I've been itching to repaste my laptop recently as the stock paste seems to be losing it's luster after almost 1 year of use.

 

Just use kapton tape or similar tape + a very small blub paste inside IHS. Why make it so difficult?

 

I've repasted my laptop with LM 3 months ago. I used a foam/sponge barrier (around 3 mm thickness) like this:



After applying the heatsink it compressed easily.



7 days ago i opened up the laptop (almost 3 months later), lifted the heatsink and the barrier remained compressed (almost half the initial thickness). I used it as is again, applied slightly more LM (smallest drop) and there's almost no effect in temperatures. I guess i don't have to repaste the cpu anytime soon. The barrier works great!

 

Hi, good work I do have a question.
When you used a barrier that thick, did you do a test without the barrier (best to do prime95 small FFTs with AVX disabled (e.g. CPUSupportsAVX=0 in local.txt), to get a baseline idea of what the temps should be like?

This is VERY Important for anyone using foam that thick. I tried foam of that thickness (different color) and I noticed a 3C temperature increase when using foam like that on the CPU. I had to trim the foam down substantially (to about 0.2mm) to bring the temps down. The foam was interfering with heatsink contact pressure.

Laptops usually have very atrocious mounting pressure. So it's important to make fully sure (by doing tests without the foam barrier) that anything that thick is not hurting the temps. You must remember that on BGA jokebooks, the surface distance between CPU surface height and the green housing it sits on is about 0.1 to 0.2mm at most. So anything thicker than that can potentially interfere with contact if its a thicker foam and several 0.xmm thickness!

 

I actually did that. I wanted to make sure it won't hurt the temperatures. The reason it's not messing with temperatures is because of heatsink pressure. I've increased the heatsink pressure by adding washers to the screws. I added a couple to some screws and single washer on others to maintain a flat heatsink pressure. I did several tests before using LM and after. Add a couple more tests with or without sponge barrier.

Also, the sponge is really compressible. I can compress the 3 mm thick sponge to 0.1 mm without any effort. I also took the liberty of checking the temperature resistance of the sponge myself. It passed (short test) but it won't be staying there for long as it can't regain it's full thickness & i also noticed a slight discoloration just after 3 months. I'll have to look for a more permanent solution sooner or later. But hey it works and i just hope it won't catch fire.

I actually did not need the foam barrier but the constant travel and too much vibrations made me a bit paranoid so i decided on adding the dam for some peace of mind.

 

Good work. It's important to let people know this precisely though, because they may look how awesome the sponge looks and then wind up using it and get horrible temps. Plus sponges are strange things. Some sponges can harden like a brick so its important to get the right one. That's why I always recommended the correct packing or window foam (if you can find it). Once you've seen it and felt it you then go "wow this will work awesome".

I never noticed a problem using the "thicker" black foam with temps I tested on the GPU heatsink. The GPU had higher pressure because the core has a larger uplift from the surface, compared to the weak BGA CPU's.

 

Couldn't agree more. I did extensive research on LM before using it. In my case the heat sink is made up of aluminium with copper shims just above cpu & gpu. There's not a lot of space between the copper and aluminium and cpu die touches the edge of the copper. Slightly more/wrong application of LM and it'll ruin the aluminium. I can't explain it properly so here are a couple of pics from months ago when i applied and cleaned up the first application later (sanded with 1200 grit just once or twice):





About the GPU part, i didn't take the risk. The copper shim is very tiny and there's a good chance the LM will contact the aluminium mountings. Here I've highlighted the GPU die contact area.



I use eGPU anyway (Gigabyte 1070 Mini ITX) which gives me PCIe 3.0 x8 speeds. Perhaps the best eGPU solution out there.

I wanted to do much more mods but stuff is limited here in India. Not much you can buy here.

I don't post much in forums since everything's already available. I just lurk.

EDIT: Fixed image

 

Did you get the foam from leftover shipping or packing material (like I did) or did you buy it someplace? Would be nice for others to know where if you purchased it. Maybe share a link or something.

It is not important for the foam to regain its original shape and the discoloration doesn't matter. All it needs to do is act as a gasket to serve as a barrier so you can handle the laptop like an ordinary laptop... toss it around in the backpack, fling it onto the bed or couch, or slam it down on the desk when you're grumpy or having a bad day, etc. ...without having to wonder if liquid metal is going to get dislodged and mess something up. I think using the foam barrier is a smart idea even if you don't actually need to. It's really cheap insurance.

Nice job on it. You did great.

 

I kind of got lucky with that foam. Like i said i never used the foam dam but read about it in clevo or alienware subforums i think.

I went to a furniture shop one day and the store owner wanted to show me some more of the stuff i asked for so he took me to his warehouse where i found this piece of foam lying between two plywood sheets. I thought it'd work great for the foam barrier and took it (with his permission of course). If i'm ever going to run out of this piece of foam i'm going back to shop furniture's and hope to get lucky once again.

Thanks for the tip! I was getting a bit paranoid about the shape and discoloration. That'll save me some time since i'll not open up the laptop again. It takes a lot of time to access the cpu by dismantling everything in my laptop. Thanks!

 

Do you have a source or if possible a name or direct link for this foam? How dense is it overall? I'd like to try it. You said a furniture shop. You mean something like Lowes? I tried Michaels before but i don't know if they would have anything nice there. By the way, I can tell from the texture of that foam that it is completely perfect. That's the same grade of foam I had found left over from some item I bought, just a different color).