3D Print Your Own Intel Delid Tool for Skylake & Kaby Lake @ [H]

FrgMstr

Just Plain Mean
Staff member
Joined
May 18, 1997
Messages
55,532
3D Print Your Own Intel Delid Tool for Skylake & Kaby Lake - A lot of us have come to find out recently that Intel once again skimped on the Thermal Interface Material inside these processors' Integrated Heat Spreaders. You can fix that issue however, and now there are some ways to do it safely. If you have a 3D printer, you can make the process a lot safer, and the tools needed much less expensive.
 
Last edited:
Another great video. I'm too much of a coward to do something like this to my brand new CPU, but I might consider it a couple years down the road.
 
In the articles and subsequent discussions, I believe you mentioned that one of the crucial steps was re-lidding the processor. The first tool used assisted in the re-lidding process as I understood. Does this 3d printed tool design assist with re-lidding?
 
I just have to say, I am enjoying the DIY project related posts and articles.

Physically tinkering with my computer has always been enjoyable yet recently the fact that OC'ing has been more mainstream has taken some of that fun away (AIO's, factory hardware support, tons of enthusiast hardware and case options). It used to feel like we were getting away with something by squeezing out some extra performance.

Keep up the good work!
 
The good days when this
17327x300.jpg

was the OC tool of choice.
 
Kyle, this may be asking too much of you but would it be possible since you do seem to have several new Kaby's if you could show us the process from start to finish. From the delidding, the scraping of the stuff you have to remove, maybe a few hints on how to be careful, application of the liquid metal and the adhesive you used to reapply the metal cpu lid. I personally would feel comfortable having a blue print so to speak. There are a few on the web but 1 - they are not in 4k ( your 4K is amazing ) 2 - Those aren't HardOCP branded or narrated by yours truly and 3 - I think a lot more people would do this if they had a start to finish.

Something to consider. Or, you may already have this planned.

Also, I think the 4k is a great tool. Would love to see more videos maybe one a month.
 
Kyle, this may be asking too much of you but would it be possible since you do seem to have several new Kaby's if you could show us the process from start to finish. From the delidding, the scraping of the stuff you have to remove, maybe a few hints on how to be careful, application of the liquid metal and the adhesive you used to reapply the metal cpu lid. I personally would feel comfortable having a blue print so to speak. There are a few on the web but 1 - they are not in 4k ( your 4K is amazing ) 2 - Those aren't HardOCP branded or narrated by yours truly and 3 - I think a lot more people would do this if they had a start to finish.

Something to consider. Or, you may already have this planned.

Also, I think the 4k is a great tool. Would love to see more videos maybe one a month.

Hasn't he already done that? The video posted last week showed the entire process of delidding, scraping off adhesive, applying liquid metal, and the reapplying of the IHS. What more do you need to see?
 
I guess I'm gonna have to find a kind soul who will 3D print one of these tools for some Paypal $$$. :D
 
Does this 3d printed tool design assist with re-lidding?
Kyle, this may be asking too much of you but would it be possible since you do seem to have several new Kaby's if you could show us the process from start to finish. From the delidding, the scraping of the stuff you have to remove, maybe a few hints on how to be careful, application of the liquid metal and the adhesive you used to reapply the metal cpu lid. I personally would feel comfortable having a blue print so to speak. There are a few on the web but 1 - they are not in 4k ( your 4K is amazing ) 2 - Those aren't HardOCP branded or narrated by yours truly and 3 - I think a lot more people would do this if they had a start to finish.

Something to consider. Or, you may already have this planned.

Also, I think the 4k is a great tool. Would love to see more videos maybe one a month.
All that is actually already there in the series of video is you watch.
 
As an Amazon Associate, HardForum may earn from qualifying purchases.
Hy Guy`s

I`m the creator of the 3D Printed Delid tools, i just found the posts on HARDOCP website.

First of all Thanks to Kyle for the great Presentation of the Tool :)

Just some in addition, for the responses i got (and also your video) it seems that the Kaby Lake has more / better glue than the Sky Lake, so it is better to screw the vice even slower than you did to give the glue time to peel off.

In Addition, PLEASE DONT USE A HAMMER ! :eek:, it`s not designed to do it this way.

Even with screwing my skylake very fast i noticed that the PCB was bending a little bit while it was under pressure so with a hammer, i don`t know what may happen to the PCB inside:cry:

To the C-Clamp, i also recommend this method only limited, especial when the mounts are very small.
the reason is if someone is printing it with poor quality (and i already saw such prints), it may brake the tool apart.
So if you really use a C-Clamp at least please use some sort of small woodplate (or similar) that has the size of the tool to distribute the force equally.

I had an Ivy delid tool printed .. ended up using a razor blade to finish the job.
could you please share the information why it didn`t worked for you ?
As i only have a Skylake CPU i could delid i also had no Ivy-Bridge CPU to do so, so any information for fixing the tool would be nice :)



sincerly Chri
 
  • Like
Reactions: Meeho
like this
Hy Guy`s

I`m the creator of the 3D Printed Delid tools, i just found the posts on HARDOCP website.

First of all Thanks to Kyle for the great Presentation of the Tool :)

Just some in addition, for the responses i got (and also your video) it seems that the Kaby Lake has more / better glue than the Sky Lake, so it is better to screw the vice even slower than you did to give the glue time to peel off.

In Addition, PLEASE DONT USE A HAMMER ! :eek:, it`s not designed to do it this way.

Even with screwing my skylake very fast i noticed that the PCB was bending a little bit while it was under pressure so with a hammer, i don`t know what may happen to the PCB inside:cry:

To the C-Clamp, i also recommend this method only limited, especial when the mounts are very small.
the reason is if someone is printing it with poor quality (and i already saw such prints), it may brake the tool apart.
So if you really use a C-Clamp at least please use some sort of small woodplate (or similar) that has the size of the tool to distribute the force equally.


could you please share the information why it didn`t worked for you ?
As i only have a Skylake CPU i could delid i also had no Ivy-Bridge CPU to do so, so any information for fixing the tool would be nice :)



sincerly Chri

I think it was just a matter of the print not being structurally sound. It collapsed on me rather than delidding the chip.
 

Attachments

  • IMG_6185.JPG
    IMG_6185.JPG
    142.8 KB · Views: 181
  • IMG_6181.JPG
    IMG_6181.JPG
    152.4 KB · Views: 229
Oh okay, that definitly looks like a printing issue, too much perimeters and bonding problem between them :/

Chri
 
Possibly printed at too low of a temperature for the type of filament being used.
 
I think it was just a matter of the print not being structurally sound. It collapsed on me rather than delidding the chip.

That looks like your hotend temperatures were too cold for the material. You could also try changing the print orientation next time so the clamp forces are inline with the layers for more strength.
 
That looks like your hotend temperatures were too cold for the material. You could also try changing the print orientation next time so the clamp forces are inline with the layers for more strength.

Most likely. Honestly I just sent the plans to a friend at a university and he had it printed on campus, don't know any more details.
 
It looks like your extrusion multiplier for the initial layer height is set too high in your slicer, or, your Z offset is too far from the bed. That will cause delamination of the layers as well.

While PLA can be strong, I'd think this is the kind of tool you want to use ABS or PETG primarily with where strength is required.
 
These videos bring a whole new aspect to [H]. Anyone can write words of what their experience is with such-an-such tool yet a video showing the tool in action speaks volumes. Thanks for taking the time to create these videos and keep up the good work, Kyle. (y)
 
These videos bring a whole new aspect to [H]. Anyone can write words of what their experience is with such-an-such tool yet a video showing the tool in action speaks volumes. Thanks for taking the time to create these videos and keep up the good work, Kyle. (y)
Yeah, this stuff needs more than words IMO. Thanks for the props.
 
I'm eagerly awaiting this, lol. I want to see if you go the safe route with light tapping or if you'll go the full WHACK route with one swing. The latter would be hugely entertaining but may be the worst idea.

I vote full WHACK. Go big or go home.
 
It looks like your extrusion multiplier for the initial layer height is set too high in your slicer, or, your Z offset is too far from the bed. That will cause delamination of the layers as well.

While PLA can be strong, I'd think this is the kind of tool you want to use ABS or PETG primarily with where strength is required.

ABS is too weak when printing normal, got already some pictures with slightly damaged tool`s,
PLA is much more harder than ABS, just print two parts and knock them on the desk, you hear the difference ;)

Chri
 
Not sure I agree with that assessment. PLA tends to be more brittle. Most prototype tools and parts are printed in ABS for its strength, especially if you vapor bath your parts. PETG has the best of both worlds (structural rigidity, easy to print). Check out Tom's on Youtube where his strength tests compare ABS to PLA and Nylon, HIPS, PETG. He tests layer adhesion, strength, temperature exposure.

As a side note, I hate printing in ABS. Bed warping is super irritating, especially on larger printed parts where you run the bed upwards of 110c and keep the fans set to minimum.
 
Why not cut a shallow groove into the piece of wood? That way it wouldn't slip around or tilt as much when you strike it. Perhaps also a larger hammer (4 pound sledge) with a slower strike.

Just thoughts.
 
Do you think a rubber mallet would work directly on the tool itself? (might not be a hard enough hit though :-/ ) A deadblow hammer would work better on that wood so it wouldn't bounce off of it so hard.
 
Last edited:
It's 40% and 100% infill, not backfill. Backfill is usually material that gets dumped into a hole at a construction site.
 
Last edited:
You're destroying, or at least trying to destroy processor that are better than what I'm running Kyle.

I need to start a website so I can have an excuse to buy shit I don't need.
 
It's 40% and 100% infill, not backfill. Backfill is usually material that gets dumped into a hole at a construction site.
Noted and corrected in the article. Not going to shoot the video again! Thanks!
 
Kyle was like Tim the Toolman when he transitioned from the broken 40% tool to the 100%. More Powa!
 
You're destroying, or at least trying to destroy processor that are better than what I'm running Kyle.

I need to start a website so I can have an excuse to buy shit I don't need.

Oh but he He does need it! All in the name of science!
 
Back
Top