Show us your IHS with a large dollop of lapping

[Jleo]

I have been reading a thread or 2 where the members are lapping their IHS (Integrated heat spreader)

Why?

Some of the IHS are concave, they have a distinct ridge, this is creating the need for a large amount of thermal compound or the need to get your fine wet n dry sand papers out and start lapping. This is happening across the Amd range so show us your IHS

Lapping Guide (this guide applies to IHS's also, but be very carefull with the Cpu pins)

http://www.3dvelocity.com/reviews/pcviper/ultrakit_3.htm

Lucky me nice flat edges on the X2 IHS no real problems here.

Not the story for my old Amd64 4000, obvious imperfections

I didn't lapp my cpu IHS, I lapped my Scythe Ninja instead since the X2 I got has a good IHS

 

[Jleo]

4 of my 6 A64/Opterons don't even have an IHS anymore

Did you freeze 'em or drown 'em? LOL

IHS

IHS = Integrated heat spreader

 

[Jleo]

How can you tell if one core is hotter than the other?

I thought there was only one thermal probe.

Each core must have its own sensor, I know Everest Ultimate displays 3 sensors results I take it 2 are core temps and one is the socket temp sensor.

 

[Jleo]

I's all bad, i got a cornrow now.

LOL, I's won't be sending you to buy my Conroe heaven knows what you'd come back wit

 

[Jleo]

What I never got with lapping was the obsession with shiny.

You want it flat, and unless you have it properly set up in a jig lapping by hand ain't going to do this.

The whole point of thermal compound is to fill in the minute imperfections in the two contact surfaces.

Flat is good, shiny is irrelevant.

Shining with a lapping/cutting paste not a household wax polish LOL

HS 1
A flat and shining surface = near scratchless surface = less TIM needed = more metal to metal contact = better heat absorption = better heat dissipation = efficient

HS 2
A flat surface that doesn't shine = more scratches = more TIM needed = less metal to metal contact = less absorption = less dissipation = less efficient

 

[Jleo]

This always makes me laugh. From an Engineering point of view surface waviness is in the order of a 1000 times more important for contact than roughness.
You not going to get a HSF really flat by hand lapping

I am glad I don't have to. I only lap to get rid of the rough surface I have to rely on the HS manufacturer to have a least made it near flat to start with

This always makes me laugh. From an Engineering point of view surface waviness is in the order of a 1000 times more important for contact than roughness

This wouldn't be laughable if the engineering application required the to surface's to stay together from a structural pov. Its is more effective to have gripping surfaces than smooth surfaces, blimey can you imagine the London Eye/Eifel Tower constructed with flat shiney surfaces on all the ajoining beams/supports, we would never have seen it in its upright glory my apologies if I am stating the obvious
Did you ever see a new pair of vice grips with flat shiney jaws? hehehe

Edit
How can this be I have seen the iron/steel being smelted/produced what is done to it to determine either finish?

 

[Jleo]

Normal stock is rolled, its as rough as ar**hol**. You can buy smooth, its actually described as ground stock from the stockist. Its ground on the biggest surface grinder you can imagin I've seen it at work. There are thousands of specifications for surface finish, fits and tolerances. It a large part of Engineering.

Good to know, my general engineering knowledge never required that depth, but I can see the relevance to more aesthetically eye pleasing engineering projects. My Leatherman for example cool piece of engineering never get tired holding it and looking at it even use it the odd time too Does that sound insane

 

[Jleo]

I used to work with Renishaw and helped design the first gen of Nikon non contact CMM's. Does that use Ruby/Diamond probes like a Talysurf or is it non contact?

Okay the rows out, he just used 5 words/terms that I didn't understand. LOL

 

[Jleo]

The last one I was involved with the chassis was made of an incredibly stable composite aggregate material. Used lasers and optics to get in the nano range and the lab had to be held at +-1C, breathing on a componet can change it. It was used for anything from Rolls Royce engine parts to hedge shears, I kid you not.

I've came across that, here is a very similar version I think it was made by Blourown a German company Click here [right click and select open]

 

[Jleo]

And thats all i was hoping to do, never said it was purely flat or that i had the machinery to do it, sometimes the sinks or IHS is totally off and a little lapping helps a lot.

Agree, my Scythe Ninja had a very defined rippled milled finish, it is no more

 

[Jleo]

More hand lapping and this is from guys who use the good old lathe, they seem very happy with the results they get plus it makes their lathes more efficient. Good enough for them good enough for me, what do you think?
Mini Lathe

 

[Jleo]

I really don't think I need to reply to this after what the other two guys have said.

To summarise though:

Shiny doesn't matter.

Flat does.

Hand lapping is rubbish.

Some people like to be contrary just for the sake of it, if your lapping is rubbish learn how to lapp better

On the shiney non shiney aspect this is for the know it all's that claim it doesn't matter.

Copper oxidizes slowly in air, corroding to produce a brown or green patina. At higher temperatures the process is much faster and produces mainly black copper oxide.

Oxidation.... reduces copper's ability to transfer heat.

So which is more effective?

Shiney meaning less oxidised

So it don't matter how little difference it makes you are wrong when you say it doesn't matter.

 

[Jleo]

It's not that shiny is not flat, but that shiny doesn't mean flat. It just means shiny.

Who said shiny means flat? bring him er now and tie him to the main mast and get me the cat o' nine tails yar tiven' varmin LOL

 

[Jleo]

You've been watch Pirates of the Caribbean again haven't you?

Big LOL

How can you live with yourself carrying that dirty dirty money around? LOL

Fact of the matter for me, remember, ME.. is that i've seen it reduce temps a fair bit and in quite a few instances over generations of cpus, it does work when done right and usually in cases where the manufacturer may have not done such a good job in the first place.

Justintime my thoughts exactly, if we keep lapping we'll convert them all, O lord forgive the non believers, shikes my hands getting sore lapping how are you fairing out? If I keeps this up I'm going to buy one of those megabuck machines can't wait 40 years to get good at this it'll kill me for sure. I wonder if I changed the word lapping to laughing and held up the Heatsink and laughed at it for 5 hours would it work better? I know one of non believers has and answer wait till you see. To be sure though you can't knock it to you tried. No answers, they either have tried it and got crap results or don't know how to do it LOL

Free to all non believers
Lapping for beginners Learn to lapp like a pro

 

[Jleo]

The hole is for gases to escape so the IHS would adhere properly when put on, thats what i've heard anyhow.

That must be why the Amd64 cpu doesn't have a hole, they don't seal the Ihs all the way round instead.