Lapping my AIOS waterblock ? Yeaay or Naay

[Cnaydmancangt]

The waterblock base on my riing 360 from TT has turned into a galiumcopper alloy thanks to liquid metal bonding to the copper.. Leaving a silver.gray almost stain that wont come off.. Liquid metal sucks.. it did the same thing to my 4790kIHS inside and outside and to the core die.. And i had it on for less than a week

Never using that stuff again.. After some reading galium aparently bonds to copper to create a new alloy.. Of which im not sure has worse or better or the same thermal conductivity. So i want it off.

The waterblock also isnt polished or smoothed out very well.. It has rings starting direct center expanding outward that looks identical to what a tree looks like when you cut it down. Each ring has grooves i ca feel with my finger specially my nails..

So i figured lapping this block would yeild two advantages.. One to get the new alloy off and back to copper.. And two to smooth the rings down and maybe help with temps.

What do you guys think ?

 

[EarthDog]

I wouldnt bother....no.

That design on the base is intentional.

 

[GTXJackBauer]

ED's right. I would try and see if TT is willing to send you a replacement copper plate if it's possible. It might void your warranty, so idk.

You can see here it has screws but again, there's liquid in there and not sure if there are advisable steps needed to replace something like this.

 

[EarthDog]

Why are we getting another plate?? I dont see anything wrong...

I see complaints about the aesthetic, but the device seems to perform fine.

Stop jumping the gun cnandy...

 

[GTXJackBauer]

I thought this was a complaint about botching the copper plate via the use of a 3rd party liquid TIM, leaving it with a rugged surface.

Maybe I misread both of your comments.

 

[EarthDog]

It's a stain is what i read?

The OP is creating problems about this stain, the new alloy created and its thermal properties. He has another thread thinking his temps were high because of the cooler, but its his chip on auto cranking the voltage on it instead. These threads are related.

The grooves in the plate he is saying is not smooth, i took as the simple machining lines. Pretty sure if the block is lapped, the warranty goes byebye as well id imagine.

I wouldnt cry over stained milk is my take.

 

[GTXJackBauer]

Each ring has grooves i ca feel with my finger specially my nails.

I'm assuming he's tried the TIM removal solution or Isopropyl Alcohol and that it's bonded on the copper plate.

 

[EarthDog]

See my edits...

 

[batboy]

Back in the old days...

Yeah, we used to lap heatsinks/coolers/blocks. We took the chipset coolers and mosfet heatsinks off and lapped them too. We even lapped the CPU core which is part of the reason Intel started using IHS on the processors. Then, we lapped the IHS. But, that was back then. Nowadays, I think water block makers purposely machine the surface concave so when the sink/block is clamped down, the pressure pulls it flat. I vaguely recall Intel doing that with the IHS too. Lapping those surfaces would be counter-productive. That said, a little super fine sandpaper and light lapping probably would not hurt if you really want to mitigate the stain (if it bugs you that much).

 

[Cnaydmancangt]

Ah ok well. I wanted to make sure it wasnt some ****ty machined make it fast design.. The rings are machined like i said they look like tree rings you guys know what tree rings look like right ? I just dont see how tree rings help with thermal transfer it would mean 50% of the surface isnt touching the ihs due to the ring grooves.. Am i making sense ? i know thermal past helps fill in the grooves but still.. It nakes no sense to me. But i do get the concaive thing..

And the voltages run stable at 1.29v at 4000mhz on cineben h the spikes in the max voltage you saw on my pics were were sitting in windows whwn it auto glocks itself, during cinebench it doesnt spike as its staying a conatant speed ny aio just seems 10c higher thrn others ive seen so i figured maybe its this new alloy the galium did to the copper. So id thought id sand it off and fix those machined tree rings.

Its more then a stain if galium actually creates a new alloy with copper like i read.. Its actually a different metal so thats why i was cancerned itd have different or negative thermal properties.

what do you guys think of the evga clc280 AIO it seems to have a bwat cooling of any AIO out there.. Its jist noisier and not flashy like the riing 360.. Its also cheaper too.. By 70$ microcenter said i could swap em when i went today.. Eveb without the box.

Sorry im not crying overspilled milk as someone said here.. Ive spent 4000$ on my system and im not happy with 70c temps on watercooling... Every 2700x ive seen the temps are 10c colder then mine when their on water and sure my ambient temps are higher... but still.. It jist seems too high

brw here is the waterblock plate..

see the rings... the finish on the block just seems meh to me.. And see what the liquid metal did after less than a week ? Cpu IHS its only touching in 80% of the center too.. The block is touching maybe 40% of the IHS.. either the IHS SUCKS or the block sucks.. Regardless though can you guys see the finish ? Its nit mirror its terrible.. Clearly machined and not polished like their W4 cpu block. The rings have such deep grooves i can see them on the cpu ihs like a fingerprint..





I just took the olate off i wanted to see how the fins were done. Seems typical pretty good to me. Although there was a ton of air in the pump. only half filled with liquid.. I noticed this AIO had air bubbles in it when i got it but i was told its normal... Hrmm.

Since i got the plate off.. I think ima lap it.. I get why the center is touching vs the sides due to concaive ihs.. but clearly the sides on my ihs arent touching the block at all.. If they were there shouldnt be blobs of AS5 itd be see through like the center.. This plate isnt doing a good job ... Also inside the pump there are chunks or slivers of metal in the liquid.. Seriously..





Okay just finished lapping the block, used 600-1000-2000- then 3000 polisher.

Look at the difference..

And my temps dropped 5c btw, instead of 62-63c in cinebench everything on auto, its now 57-58c and autoclocks to 4050mhz vs 3950. Thats a huge difference. I knew this plate was poorly done, not polished or anything with machine marks i could feel with my skin !!, that and the Gallium from the liquid metal probly didnt help either.. Look at the pic i circled the tree like swirls im talking about..

 

[GTXJackBauer]

Well done.

I have to say, I'm not surprised being a TT product. I'm sure most of their AIO's aren't like that but it seems you got a not ready bottom plate. I would have reached out to TT and made them send me one or get a refund on that AIO. I know my Corsair AIOs I've had over the years were smooth as butter.

 

[Cnaydmancangt]

I dunno man when i put this AIO on my 4790k vs my 17 year old dangerden maze 3.. which is you know just a 3 loop maze.. no 1000000 little micro fins.. Temps only went down 5-6c which isnt much considering the fin design... Oh and im running prime 95 now.. 5-10 min in temps are only 63-64c just went to 64.. 5-10 min in prime95 small ffts before temps were 70c


TBH i feel like calling thermaltake and telling them its complete BS how bad this copper plate was.. You see the deep machined circles right ? Some of the ())() swirls are from the camera.. but the reason the camera did that was because of the deep circles on the block creating an optical illusion which is why i circled the ones in red. Thermaltakes W4 cpu block you can buy.. their own pictures brag about the polished nickle plated bottom.. Why didnt mine have a polished bottom, for $175.. meh


EDIT: wanna know something funny.. Look at the picture i took of the fins inside... Notice how the flat part of the plate there is polished and shiny... lol the inside of the copper plate was better looking then the outside...

 

[batboy]

I guess my gut feeling for AIO loops is partially justified. I always felt they were substandard to a good custom loop. I'd say that a 5 degrees cooling improvement was worth the effort.

 

[Cnaydmancangt]

I called thermaltake, the support rep told me to make a ticket, and i flat out told them i took the plate off, the pump was half empty, the liquid had silver metal slivers in it, then i lapped the plate and included before and after pics of the plate, pump, and the temp drops. I told them i mentioned this on some forums and im hoping they will rectify the issue somehow regardless of me voiding the warranty.. Because in the end they cant stop me from returning it.. If i wanted i could simply make the drive to micro center and return it and get a different setup

 

[EarthDog]

Its amazing the conclusions being arrived at with anecdotal musings.

1. The grooves on the block are normal. Depending where one looks that is MORE surface area to transfer heat.
2. Since no empirical testing was done, who knows if its a mount difference or tim application difference etc.

That said metal shavings shouldnt be in there, and id be floored if Tt took that thing back after you took it apart and lapped it. Dont complain if they don't.

 

[GTXJackBauer]

Yup, you should have made that phone call prior with your benchmark/temp evidence before voiding the warranty on that thing. I'd also be surprised if they did anything at this point.

P.S. - Don't say our forum or names and throw us under the bus. LOL!

 

[Cnaydmancangt]

I didnt say the forum name, i said i posted in forums. I dont care about my warranty i have a replacement plan through microcenter that i paid $20 for and im still within 30 days to return it as well. I have no need to return it now anyway. But if i wanted too i could easily. Im just curious what thermaltakes response will be on the swirls.. Again if i wanted a new AIO i could drive 20 min away and swap it out in 5 minutes

Its amazing the conclusions being arrived at with anecdotal musings.

1. The grooves on the block are normal. Depending where one looks that is MORE surface area to transfer heat.
2. Since no empirical testing was done, who knows if its a mount difference or tim application difference etc.

That said metal shavings shouldnt be in there, and id be floored if Tt took that thing back after you took it apart and lapped it. Dont complain if they don't.

Depending on one looks lol, well you must be blind. A spiral surface like that doesnt equate to more surface area touching.. Think about that for a second.. You got 2000 ridges and 2000 ditchs.. You take a flat surface and lay it on top.. 2000 ridges are touching the flat surface, 2000 ditches arent. You shave all the ridges down to be flat with the ditch and what do you have then ? More surface area as its all even now.. where the hell did you conclude that ? Why is every cpu block and gpu block polished, yet an AIO block isnt ?

If we went by what you saidd you might as well take a metalic brush like this to your IHS and heatsink base im sure 10 million deep scratchs will add more surface area as well... "LOGIC"

As for testing, and mounting.. Its kinda hard to MISS mount a flat block on a flat IHS with 4 screws tightend lightly and then in a X pattern especially a mounting design with no springs... I also used 2 different pastes. Ive been mounting blocks on pcs for 20 years dont insult me man..... I ran all the same test at the same auto settings 20 times, including prime 95 the temps are 5-6c colder.. My room temp hasnt changed. The point of thermal paste as you know is to fill is in the grooves and scratches and micro scratchs you cant see to make for a better contact area. Thermal paste was doing its job with those swirls, but there was LESS hard surface area..

Now maybe your more surface area logic is measuring the distance from the ridges to the ditchs and adding that up.. youll have more "square" footage if that makes sense.. But there is less surface area touching the flat IHS.. The thermal paste will also fill in the swirls making that surface area increase Nigh

Explain to me why this block is mirror polish and the AIO wasnt.. The fin design is the same. If the swirls were better they would have done it to this block as well. Im betting the swirls are a machining process to make things cheap and mass produce them. Why waste time polishing the heatplate on an AIO

https://www.amazon.com/gp/product/B06XYTBKD8/ref=ox_sc_act_title_1?smid=ATVPDKIKX0DER&psc=1


View attachment 199875

 

[Maviryk]

Just chiming in here. Admin was correct that it was poor machine work on that copper plate surface.

Looks like ThermalTake has not improved on their quality in the last decade.

Another note, just because something has a mirror finish, does not mean that its flat, and vice versa.

 

[Cnaydmancangt]

Just chiming in here. Admin was correct that it was poor machine work on that copper plate surface.

Looks like ThermalTake has not improved on their quality in the last decade.

Another note, just because something has a mirror finish, does not mean that its flat, and vice versa.

True, but the surface i lapped it on was flat enough and well the results speak for themselves. Im not going to spend my time doing


tests taking pics and vids and make a whole review on this.. But like i said temps went down, and im happy. I like this AIO and the RGB matchs the setup i have. I dont really want to switch it, which is why i basically said F it im lapping it lol



edit: side note.. i cant find any reviews for the various brands of waterblocks.. its alll AIOS.. does no one review seperate waterblocks together anymore ?

 

[Blaylock]

For the record those swirls you are talking about are NOT due to poor machining. These same machine marks are found on several other major manufacturers' heatsinks. It's not indicative of poor quality.