Rockit Cool Rockit 88 Delid and Relid Tool @ [H]

FrgMstr

Just Plain Mean
Staff member
Joined
May 18, 1997
Messages
55,532
Rockit Cool Rockit 88 Delid and Relid Tool

Every since Intel decided to start once again using a thermal interface material in its enthusiast CPUs, we have been looking for better ways to remove the Integrated Heat Spreader (IHS). Rockit Cool has what we consider to be the best tool we have used yet. If safety and ease are priority, this is it.
 
I'm a very happy owner of this tool for Skylake / Kaby Lake processors. It's superbly built, easy to use and RockitCool's customer support is absolutely top-notch.
 
I've enjoyed my rockit, but sadly, the main thread on mine stripped out after it's first use. I did manage to get it to work in combination with a vice since then. I have not contacted them for support regarding the issue, however, so I don't know how they'd handle it. I think mine may have just had some bogus threads :( Regardless, I've been happy with the ease of delidding and relidding that it's offered.
 
I must say the Rockit 88 is a very well designed delidding/relidding tool. Thank you for the video showing how to use the Rockit 88. Very well done.
 
Note that this is an early design of the Rockit88, the newer version doesn't use all three of the assembly thumbscrews on the relid portion. Where the third little thumbscrew would go is another large tapped hole similar to the one in the center you use for the relid process on LGA115x processors. I'm still unsure why they made this change as I think the old design is much better for re-lidding. I can only imagine it was to make the kit compatible with more SKUs?

20180402_123844.jpg

(sorry about the portrait mode picture... meh)

I would like to relate an extremely positive experience with their product team though. When I ordered my kit from them, the triangular shaped plastic hold down plate had a fault in the threads that made using it kind of difficult (and it actually slightly mangled the threads of the big plastic hold down bolt too). I reached out to them and they IMMEDIATELY sent me a replacement bracket and bolt via Priority Mail - no questions asked. You can't ask for better support than that, and they won me over then and there. I've since used my Rockit88 at least half a dozen times and not had a single issue. It's built tough and I see no reason why I wouldn't be able to use it a hundred times over without issue.


I'm really glad to see you do a review on this guy, I think it's a lot easier for people in the states to get hold of than the DeBauer tool. :)
 
Last edited:
Fast forward to today, and I'm having trouble with my EK monoblock due to my relid having a lower z-height than Intel spec: James is custom cutting one of his copper IHSs for me at no extra charge.

Oh yeah, I forgot to mention that in my original reply...

The Rockit88 folks also are now selling a custom copper heat spreader for LGA1151 (Skylake, Kaby Lake, Coffee Lake)... they even include a special relid frame to use with their IHS since it's shaped slightly different. I REALLY want to get one of these and try it out! :D
 
Oh yeah, I forgot to mention that in my original reply...

The Rockit88 folks also are now selling a custom copper heat spreader for LGA1151 (Skylake, Kaby Lake, Coffee Lake)... they even include a special relid frame to use with their IHS since it's shaped slightly different. I REALLY want to get one of these and try it out! :D
I've got a thread about this little mishap of mine, and I'm planning to report back there one I get the new IHS installed, if you're curious about results. RockitCool is shipping mine today, so it should be in workin the week.

https://hardforum.com/threads/warning-delidded-cpu-monoblock-poor-mating.1957658/
 
I'm having trouble with my EK monoblock due to my relid having a lower z-height than Intel spec
Uh yeah. You change spec height on the components that is bound to happen. I am not an overall fan of monoblock solutions. Hard enough to get a solid mating surface without having to focus the block to mate with others as well.
 
Yep! This'll be my first monoblock. If I'd thought it through a bit, I'd have realized that 1.) a glued IHS lowered my z-height and 2.) the monoblock doesn't self-adjust with springs the same way a CPU block does... but I didn't put two and two together before trying it out.

James experienced exactly the same thing when he was testing a delidded CPU with a monoblock, and knew exactly what I needed: a thicker IHS to bring my z-height back into Intel spec.

Doesn't a thicker IHS also reduce the thermal efficiency of the setup, even if not to the same extent as having poor contact from being to short?
 
Doesn't a thicker IHS also reduce the thermal efficiency of the setup, even if not to the same extent as having poor contact from being to short?
I can't imagine that being the case. The poor contact condition was a visible air gap. The thickness of the IHS only needed to be increased by 0.25mm in order to bring the whole package back within Intel's dimensional spec. Of course I won't know until I get the thing and remount, but I'll report my findings then.

It may not be exactly as thermally efficient as a stock-thickness IHS from Intel, but I don't think it'll be a measurable difference with only 0.25mm more copper between the die and coldplate.
 
You really can't overhype the benefits of delidding. I had a local shop delid and relid my 3770k with one of these, and I'm still working through AS5 burn-in, and temps dropped by ~20C.

As someone who's been overclocking for like 15 years and manually delidded his Opteron 165 with a razor blade, these tools are pretty amazing. When the time comes to rebuild, I plan to pick up one of these for the respective socket and pop the top myself like the good old days.

Thanks for the solid write-up Kyle!
 
Intel needs to eat a bag of you know what over this awful thermal solution. About to finally get my 7700k done as it's been one of the thermal lemons and that's on water.
 
I bought their Rockit 99 tool for socket 2066 chips, and was really impressed how well it worked. I used it on a 7820X, with Conductonaut in place of the Intel TIM. I've got my ROG Rampage VI Apex running at 4.7 GHz with temps as expected with a delid (15-20C less than before - using a NZXT Kraken X62 with Kryonaut).

My first choice was der8auer's Delid-Die-Mate-X, but I couldn't find anyone selling it in the US. It's seriously over-engineered and is very beautiful, but it woulda cost about $130 US with shipping.

I'm glad I chose the Rockit 99. I sweated bullets leading up to the delidding, with visions of having to buy a new processor. But really the Rockit 99made it dirt simple.
 
The Rockit tool is a must. I tried to do a 7700 with a razor blade & borked it. Got a tool for the second one, worked fine, probably going to last a long time. No problems with my noctua cpu fan fitting, either. Temps never go more than 30 -35c above room temp (approx 25c), even when getting beaten around by Civ 6 or AS:eek:rigins (I hit 100% cpu on my 5 ghz 7700k with them). Don't think I'll ever go wet, air seems to do me just fine.
 
Weird coincidence that this posted just now. I'm actually considering this IF my 8 hour RealBench stress test doesn't pass, I'll know when I get home. I have a 8700k with a custom loop and it's frustrating that I can get the chip stable at 5.0GHz @ ~1.264 volts but temps during AVX Prime get up to the high 90s. Temp is definitely my bottleneck not voltage.

I've backed off to 4.9GHz at 1.232 volts which makes for better temps; but if that doesn't pass I'll just delid up the voltage for 5.0Ghz and perhaps slightly beyond.
 
I bought their Rockit 99 tool for socket 2066 chips, and was really impressed how well it worked. I used it on a 7820X, with Conductonaut in place of the Intel TIM. I've got my ROG Rampage VI Apex running at 4.7 GHz with temps as expected with a delid (15-20C less than before - using a NZXT Kraken X62 with Kryonaut).

My first choice was der8auer's Delid-Die-Mate-X, but I couldn't find anyone selling it in the US. It's seriously over-engineered and is very beautiful, but it woulda cost about $130 US with shipping.

I'm glad I chose the Rockit 99. I sweated bullets leading up to the delidding, with visions of having to buy a new processor. But really the Rockit 99made it dirt simple.

Totally agree. I got the 99 as well after checking out the DDMX and considered sending it into Silicon Lottery for delid service. Though I was also quite nervous (first timer) but it was a breeze with this tool particularly the relidding. Now I have a new 7820x which I grabbed in the ebay sale last week and im super happy with the decision since its already paid for itself and its simple so Im no longer nervous about it all. Its like any other pc building task. I cant recommend it enough and Im glad its getting some love on [H].
 
I have never done anything serious to an IHS prior to buying the RockitCool.
The thing is amazing, simple to use and has worked as advertised.
I've done 6 CPUs, no problem. Resealed each with Red RTV using a 16 gauge needle on a 1cc Insulin syringe. No issues.

Really simple,well built and designed tool .
 
I need to find a source for the syringes... I've just been using a toothpick to apply the RTV to the edges of the IHS (yes, I still make sure to leave an air gap). Works just fine but the syringe / needle method would be a lot less messy methinks...

I already bought a 100 pack of those plastic razor blades Kyle uses in his delid videos. Those things make cleaning the glue off the PCB and IHS a freaking breeze!
 
Very good tool, quality of the material is top-notch. Have used it to delid my i7 7700K and will probably delid my new 8700K too once I move it from my DAN A4-SFX to the Ghost S1.
 
I need to find a source for the syringes... I've just been using a toothpick to apply the RTV to the edges of the IHS (yes, I still make sure to leave an air gap). Works just fine but the syringe / needle method would be a lot less messy methinks...

I already bought a 100 pack of those plastic razor blades Kyle uses in his delid videos. Those things make cleaning the glue off the PCB and IHS a freaking breeze!

Syringes:https://www.amazon.com/Tuberculin-S...F8&qid=1522866397&sr=8-1&keywords=tb+syringes

You put a wad of the RTV on a paper plate, suck up about 1/2 a syringe full, which will be more than enough to
lay down a nice bead for a single CPU.

16 gauge blunt tip needle to fit: https://www.amazon.com/QTY-1-1-Disp...2866454&sr=1-3&keywords=16+gauge+blunt+needle

Rather than buying plastic razor blades, I just cut a brand name Q-Tip shaft at a 45 degree angle. It will scrape the old sealant off easily.
 
As an Amazon Associate, HardForum may earn from qualifying purchases.
Does anyone have info regarding how much cooling delta you get from using traditional pastes (like Noctua NH1, Arctic MX4, etc) after re-lidding?

I just am not thrilled with many aspects of the metal solutions, as wonderfully conductive as they are.
 
Does anyone have info regarding how much cooling delta you get from using traditional pastes (like Noctua NH1, Arctic MX4, etc) after re-lidding?

I just am not thrilled with many aspects of the metal solutions, as wonderfully conductive as they are.
Anecdotally from my experience, you will generally net about 75 to 80% of the gain with metal TIMs.
 
I just am not thrilled with many aspects of the metal solutions, as wonderfully conductive as they are.

Can you elaborate on what you don't find thrilling with LM?

I built an 8700K system a few months ago, with the intent to delid and OC it. I got the Rockit 88 and some Grizzly Conductonaut,but haven't gone through with it yet.
 
Can you elaborate on what you don't find thrilling with LM?

I built an 8700K system a few months ago, with the intent to delid and OC it. I got the Rockit 88 and some Grizzly Conductonaut,but haven't gone through with it yet.

Just the fact that it is electrically conductive and a little fussier in application than the easy peasy pastes. Nothing deal-breaking, just aspects I'd prefer to avoid if the delta wasn't very big.
 
I ordered a kit along with the copper IHS and the Liquid tim they sell. Plan on delidding my 7700k and the conductive nature of the tim also concerns me. Most of the vids I've seen, including Kyle's, used nail polish (not sure what kind exactly and im not sure if some of them are conductive themselves), but I've also seen people just put tape over the contact points during application of the tim or not protect them at all. My concern with not putting anything over the contact points is that the tim may eventually "drip" while it is vertically installed in a pc. I also dont have the steadiest of hands so I may just spray the tim everywhere during installation.
 
I ordered a kit along with the copper IHS and the Liquid tim they sell. Plan on delidding my 7700k and the conductive nature of the tim also concerns me. Most of the vids I've seen, including Kyle's, used nail polish (not sure what kind exactly and im not sure if some of them are conductive themselves), but I've also seen people just put tape over the contact points during application of the tim or not protect them at all. My concern with not putting anything over the contact points is that the tim may eventually "drip" while it is vertically installed in a pc. I also dont have the steadiest of hands so I may just spray the tim everywhere during installation.

Just put some clear nail polish on there if you're afraid of damaging them. I suggest using a paintbrush and being conservative with the liquid metal tim. The rockit tool really makes everything pretty easy.
 
So I was wondering. Do I have to trace the entie pattern of the previous silicon that kept the IHS in tact? I was hoping to just do the wings of the original IHS to hold it in place w/the ultra black permatex suggested on the rockit88 site.
 
So I was wondering. Do I have to trace the entie pattern of the previous silicon that kept the IHS in tact? I was hoping to just do the wings of the original IHS to hold it in place w/the ultra black permatex suggested on the rockit88 site.
Nope! You technically don't have to re-adhere the IHS at all, since the socket holds it down, but it makes the CPU a lot easier to handle. The only no-no (from what I've read) is to completely seal in the airspace all the way around.
 
Nope! You technically don't have to re-adhere the IHS at all, since the socket holds it down, but it makes the CPU a lot easier to handle. The only no-no (from what I've read) is to completely seal in the airspace all the way around.
good. Hopefully this goes accoring to plan then. gonna test to see if it works soon. Should I wait for 24 hours or is 1 hour enough time for it to adhere?
 
good. Hopefully this goes accoring to plan then. gonna test to see if it works soon. Should I wait for 24 hours or is 1 hour enough time for it to adhere?
If you're using RTV, I'd let it go 24 hours. You wanna make sure the adhesive is fully cured before you subject it to stresses from the socket clamp. I've never used RTV, so take that for what it's worth, it's just my gut instinct.
 
If you're using RTV, I'd let it go 24 hours. You wanna make sure the adhesive is fully cured before you subject it to stresses from the socket clamp. I've never used RTV, so take that for what it's worth, it's just my gut instinct.
Everyone should do whatever they are most comfortable with, of course, but 24 hours for this application is way overkill IMO. In the case of the commonly used Permatex Red or Black gasket maker keep in mind this stuff is used quite a bit for things like engine manifolds, both intake and exhaust. These are located in high-heat areas (especially so with exhaust) and subjected to large amounts of pressure and/or vacuum. The 24 hour fully cured time on the Permatex directions is if, after application in a "normal" ambient temperature and humidity environment, you do absolutely *nothing*. It will cure much more quickly when exposed to higher temps (like starting up a motor). Also, the smaller the amount (and we are talking about tiny dabs on an IHS vs "creating" a new gasket to seal an entire exhaust manifold) the quicker the compound will cure. I have experience building PC's and have been working on cars and motorcycles since back when fuel injections was a rare and strange thing and nobody had ever heard of a computer except in science fiction.

I freely admit I do NOT yet have experience marrying the two (putting RTV onto a PC component) but will be doing so soon as I finally delid/relid my ancient 4770K. Today when I, or friends I'm helping, reassemble components using RTV in a car we wait about as long as it takes to properly button everything up before starting it up. I plan on doing the same with my PC. Keep in mind that this stuff is used not just by professional mechanics (not shadetree like me) but assembly lines as well, where time is very definitely money but failed repairs and warranty/recalls are even more so. If the RTV sealing an exhaust manifold on an SBC can handle the temps and pressures within an hour of being applied I'm not at all worried about a few dabs holding down the IHS on my PC.

Just my $0.02 worth, don't blow it all on wine and women in the same place.
 
Everyone should do whatever they are most comfortable with, of course, but 24 hours for this application is way overkill IMO. In the case of the commonly used Permatex Red or Black gasket maker keep in mind this stuff is used quite a bit for things like engine manifolds, both intake and exhaust. These are located in high-heat areas (especially so with exhaust) and subjected to large amounts of pressure and/or vacuum. The 24 hour fully cured time on the Permatex directions is if, after application in a "normal" ambient temperature and humidity environment, you do absolutely *nothing*. It will cure much more quickly when exposed to higher temps (like starting up a motor). Also, the smaller the amount (and we are talking about tiny dabs on an IHS vs "creating" a new gasket to seal an entire exhaust manifold) the quicker the compound will cure. I have experience building PC's and have been working on cars and motorcycles since back when fuel injections was a rare and strange thing and nobody had ever heard of a computer except in science fiction.

I freely admit I do NOT yet have experience marrying the two (putting RTV onto a PC component) but will be doing so soon as I finally delid/relid my ancient 4770K. Today when I, or friends I'm helping, reassemble components using RTV in a car we wait about as long as it takes to properly button everything up before starting it up. I plan on doing the same with my PC. Keep in mind that this stuff is used not just by professional mechanics (not shadetree like me) but assembly lines as well, where time is very definitely money but failed repairs and warranty/recalls are even more so. If the RTV sealing an exhaust manifold on an SBC can handle the temps and pressures within an hour of being applied I'm not at all worried about a few dabs holding down the IHS on my PC.

Just my $0.02 worth, don't blow it all on wine and women in the same place.
Good info. One thing I might say though is that we're not using it in a gasket application here. It may still be a little rubbery while still strong enough to hold in exhaust gases, but if it's a little rubbery when you clamp it down under your CPU socket, it may result in the IHS being deflected.

Again, just firing from the hip here, but it's fun stuff to talk about. =)
 
Good info. One thing I might say though is that we're not using it in a gasket application here. It may still be a little rubbery while still strong enough to hold in exhaust gases, but if it's a little rubbery when you clamp it down under your CPU socket, it may result in the IHS being deflected.

Again, just firing from the hip here, but it's fun stuff to talk about. =)
If the intent is to use the sealant to have a perfectly aligned IHS/CPU die marriage prior to reinsertion and clamping into the mobo, then agreed - movement could occur if the sealant isn't fully cured; your point is quite valid. This would also dictate that the builder be using one of the newer jigs such as the Rocket tool/jig in the title or something similar (including homebuilt) to assist with such alignment during the curing (oooh, I'm bustin' rhymes here!). Given that the original delidders rarely even bothered to reseal the IHS and relied only upon the mobo's CPU socket clamp I'm not too worried about a small amount of deflection that I can (hopefully) account for by strategically placed digital manipulation (my finger). Especially for a Haswell. No doubt for the newer CPU's this may be a larger concern; given all of those that have gone before me with Haswell's I'm OK with my decision.

N.B. - if I *did* need such alignment and had access to such a tool (I'll be using a vise for my delid since I have one and don't have any other delid projects in the forseeable future) I'd still probably pass on the 24 hour wait time and instead opt for a different sealant like Permatex's The Right Stuff (grey), which is ready to go in one minute. Since I have Red RTV on hand and I'm a cheap bastard and really have no other need for "TRS" that's what I'll be using. :)
 
This is the rockit thread so if you are using rtv, you don't need anything else no crazy glue etc. Use a toothpick and spread some sealant on the lip of the IHS, throw it together on the relidder. In 15 minutes it will hold but I say keep it under the relidder for an hour or more then you can reinstall it and start oc'ing.

I've been using the rockit for a couple years now, first with a 6700k and now with a 7820x.

https://hardforum.com/threads/delidded-6700k.1920121/#post-1042721901
 
well under windows 7 it sits(stock cooler as well) at a nice 27 degrees idle(its been testing in a dell optiplex running for a few hours now). It did do a strange boot then turn off once but after that its been golden. I did wait 2 hours and it was on with no wiggle at all when clamping it in

Gotta admit this went smooth as butter.
 
Maybe this has been answered, but I can't seem to find it :( What TIM did you use? Finally setup my 8700K and thinking of doing this now :p
 
Back
Top