Overclock.net banner

What to use to protect 4770K VRM from Liquid Ultra

  • Clear nail polish (Max operating temp unknown)

    Votes: 1 7.1%
  • MX4 paste

    Votes: 4 29%
  • Liquid electrical tape (Max operating temp ~93C)

    Votes: 5 36%
  • Non-conductive thermal pad cut to fit under IHS and around CPU die

    Votes: 4 29%
1 - 15 of 15 Posts

HairyGamer

· Banned
Joined
·
390 Posts
Discussion starter · #1 ·
I'm going to be delidding my 4770K in the next day or two when my CLU arrives in the mail...

Can anybody recommend a method of covering the VRM on the chip to keep the CLU from shorting them out?

I've seen somebody use MX4 smeared onto the VRM but that seems like it would only insulate them and make them run hot while still leaving a chance for the CLU to come into contact with them.

I've seen somebody use clear nail polish to coat them but I'm not sure if that can be conductive in any way (doubt it but you never know, it's nail polish not liquid electrical tape) and I'm doubting its ability to stand up to 85C for any extended period of time.

I've also seen somebody use Fujipoly thermal pads (non conductive) and that seemed like the most "legit" route but I'd have to put off delidding for a week or so if I go that road. I don't mind waiting to do things "right" though. Should I use a non-cunductive thermal pad cut to fit around the die and under the IHS? If so, can anybody recommend something that I can get shipped quickly in the US, preferably Newegg since I get free 2-day shipping from them (which is usually actually overnight
thumb.gif
)

Also, the easiest solution for me (I think) would be to just use liquid electrical tape... it has an operating temp of up to 93C (not sure how hot the VRM gets but I doubt that hot) and I can just brush it on and have a nice non-conductive coating that should peel off without trouble if need be. Anybody have anything to say about this idea?

Thanks OCN geeks
smile.gif
 
I don't know about liquid electrical tape but IMO, the safest method I've seen so far is cover it with non-conductive thermal paste. As long as we do not know how hot these little capacitors run at, I wouldn't put anything else. I'm not gonna cover them tbh, if you do it correctly, nothing will happen. It's not like thermal paste is going to magically move by itself, plus liquid metal paste is a very thin layer, it doesn't spread like other pastes. Even with normal paste I haven't seen any go on the capacitors.
 
Discussion starter · #4 ·
Quote:
Originally Posted by Bartouille View Post

I don't know about liquid electrical tape but IMO, the safest method I've seen so far is cover it with non-conductive thermal paste. As long as we do not know how hot these little capacitors run at, I wouldn't put anything else. I'm not gonna cover them tbh, if you do it correctly, nothing will happen. It's not like thermal paste is going to magically move by itself, plus liquid metal paste is a very thin layer, it doesn't spread like other pastes. Even with normal paste I haven't seen any go on the capacitors.
Thanks for the input - I'll leave them bare and see how it goes
smile.gif
 
Liquid electric tape is what I would use there, I've used it on quite a few motherboards & gpus. Removable, doesn't trap heat, waterproof & can take reasonably high & low temps very well.

Nail polish would keep anything from shorting due to running conductive paste, although I don't know how well it does with higher temps.
 
Discussion starter · #7 ·
Quote:
Originally Posted by FtW 420 View Post

Liquid electric tape is what I would use there, I've used it on quite a few motherboards & gpus. Removable, doesn't trap heat, waterproof & can take reasonably high & low temps very well.

Nail polish would keep anything from shorting due to running conductive paste, although I don't know how well it does with higher temps.
Awesome, I will use Permatex then since it has an operating temp of -54C to 204C and I've always had good luck with Permatex products on my vehicles. I think that would be the most suited material for my needs, as well as the safest route to protect my investment. I'm going to give it a good 12 hours to completely dry before I run the chip.

Thanks for the help
thumb.gif


Permatex liquid electrical tape technical data sheet - for anyone interested in trying this with their chip.


I'll post pics/video and results this weekend.
 
And for the hundredth time, they aren't VRMs, they are just surface mount capacitors. They don't get hot, or else there would be some provision for cooling them already in place.
 
Discussion starter · #9 ·
Quote:
Originally Posted by Forceman View Post

And for the hundredth time, they aren't VRMs, they are just surface mount capacitors. They don't get hot, or else there would be some provision for cooling them already in place.
Why did you post on my thread just to be annoying? Go away dude - I called them VRM in one other thread and posted this before you rudely corrected me.
thumb.gif


They (whatever you wish to call them) get warm, we'll all have to agree on that. You can't say they don't get hot - maybe they just run hot and don't need to be cooled by design.

Unless you have some Intel credentials I'd stop trying to sound like such a knowitall.
rolleyes.gif
 
The caps on the PCB won't create heat, at least nothing compared to the hot die a few millimeters away, but I would think everything under the IHS is going to be hot while the rig is running with that little furnace of a die in there.
I've corrected the VRMs / SMD capacitor mistake a few times too, but most are thinking they are VRMs so I just stopped correcting instead of keeping on saying it. Doesn't really matter to most folks as long as they aren't getting knocked off or shorted
 
Discussion starter · #11 ·
Quote:
Originally Posted by FtW 420 View Post

The caps on the PCB won't create heat, at least nothing compared to the hot die a few millimeters away, but I would think everything under the IHS is going to be hot while the rig is running with that little furnace of a die in there.
I've corrected the VRMs / SMD capacitor mistake a few times too, but most are thinking they are VRMs so I just stopped correcting instead of keeping on saying it. Doesn't really matter to most folks as long as they aren't getting knocked off or shorted
Yeah, I figured everything under the IHS was quite warm seeing as the CPU runs so hot - physics are funny like that.. Thanks for backing that up because I didn't understand how they couldn't be getting hot sitting about 2mm from an 85C CPU die.

The VRM mistake is likely coming from some bad information online - I was calling them VRM due to something that JJ (Asus Rep) said in an OCing video about Z87 and Haswell which I believe he may have done with Newegg IIRC.

I understand that they are just capacitors but as you said, I really can care less what they are called so long as I don't knock them off while delidding or short them out with the CLU if I were to let the IHS slide or something along those lines.
 
Quote:
Originally Posted by HairyGamer View Post

Why did you post on my thread just to be annoying? Go away dude - I called them VRM in one other thread and posted this before you rudely corrected me.
thumb.gif


They (whatever you wish to call them) get warm, we'll all have to agree on that. You can't say they don't get hot - maybe they just run hot and don't need to be cooled by design.

Unless you have some Intel credentials I'd stop trying to sound like such a knowitall.
rolleyes.gif
I responded here because unless people stop calling them VRMs new people will think they are VRMs and the merry-go-round will continue. If bad info keeps getting spread (like the 1.52V max voltage thing from a few years ago) then it keeps providing people bad info. I'm sorry you took it personally, it wasn't meant as an attack, but there are tons of posts in the delidding threads about the capacitors and what to cover them with.

And I didn't mean the hundredth time you said it wrong, if that's what you got offended about, just that they have been incorrectly called VRMs by a lot of people.
 
Discussion starter · #13 ·
Just got my PC back together after a successful Haswell delidding.

I used a razor blade for 85% of the work and then a thin plastic card to finish it off as safely as possible.

I used liquid electrical tape (rated for ~95C) on the capacitors to the left of the die to protect them from contact with the CLP.

Pics to come tomorrow, I really need to get to bed because I have to go to work in ~6 hours.

Anyway - the good news is that I went from 86C to 54C!! That's a sick 32C drop
thumb.gif


This is awesome! Can't wait to try to OC from scratch again tomorrow - I was stable at 4.5 but I went back stock because of temps just before delidding.

A few pics for now, more tomorrow - maybe the video too.


Liquid tape on capacitors and anything that could be perceived as a nick although they were really hardly even scratches.

If you look closely you can see the scratch on the top right of the pic, covered it with liquid tape since I'm paranoid.

 
i know this is old but i am about to delid my 4770k. i have a few questions.

what tool did you use to spread the liquid electric tape onto the mini capacitors? i watched a youtube video where a guy used a toothpick and it looked way too sloppy (your job looks as clean as i would want it)

did you use the liquid tape to re adhere the lid to the chip? or are you just using the socket bracket to hold in place? i want to re adhere the lid in case i need to remove the cpu in the future i dont have the hassle of trying to keep it together.

i know its been a while, but do you have a link to where you saw a guy use fujipoly on the caps (as stated in your first comment)? i just re did my 295x2 with grizzly kryonaut on the GPU's, Grizzly minus 8 pads on the Ram, and Fujipoly Sarcon XR-m on the vrm. i have a very small amount of 1mm (may not be enough) and a decent amount of 1.5mm Fujipoly that i could use.

i figure i would just use my thumb and smash the fujy poly over the caps to cover them. but not sure the best way to apply it.

or maybe squish fujipoly onto caps, then use electric tape to glue fuji down to cpu (not cover the fuji, but outline the fuji)? not sure if that would hinder the performance of the fuji pads
 
1 - 15 of 15 Posts