Philly's G80 Waterblock Review - Work log **56K Warning**

I am in the process of acquiring 4 sets of G80 waterblocks to do a comparison review. So far, Gabe@ Swiftech and Alex @ Petra's have come thru with a load of "loaners". The guys @ Danger Den and Eddy @ EK are both also contributing lotsa stuff (tubing, extra barbs, a Fluke MM for temps/voltages). I also have the expertise of OPPAINTER (Swiftech) and ROKK (DD's design engineer) to help with ensuring the data is gathered as fairly as possible.

http://i41.photobucket.com/albums/e2...ty-Stuff-5.jpg

http://i41.photobucket.com/albums/e2...ty-Stuff-1.jpg

First off I will be comparing the MCW-60 to the Apogee 1U (with ram sinks on the ramdac, VRM's and memory modules and fans blowing directly on the ram sinks.) then eventually I'll be comparing both of them to the DD NV-88 and EK's soon-to-be-released FC-8800 full coverage blocks. I will also be playing with 226 watt pelts on the Apogee 1U's fed by a Meanwell SP-320-13.5. We will see if the FC blocks are better cost vs performance to something more basic (MCW-60 or Apogge 1U with ram sinks).

I am going to try it all in my current loop (link in my sig) as it is now, then add a second MCP-355 (with Petra's top of course!) to see if the increased flow/head will help a loop trying to cool 4 blocks (CPU/GPU x 2/chipset). I will also compare data with the CPU/NF4 on a seperate loop (MCP-355/EK RES 150/BI GTX240 or the Swifty MCR 220 rad feeding my Storm G4/Maze 4 Expert) and allow my standard loop - PA120.3/MCP355 (Petra's top)/Cyclone 120 res - to cool just the blocks used for the comparison.

Right now I am doing the mundane things like rad prep while I wait on the shipments from Danger Den and EK.

http://i41.photobucket.com/albums/e2...rk-Log-001.jpg

http://i41.photobucket.com/albums/e2...rk-Log-002.jpg

http://i41.photobucket.com/albums/e2...rk-Log-003.jpg

http://i41.photobucket.com/albums/e2...rk-Log-004.jpg

http://i41.photobucket.com/albums/e2...rk-Log-005.jpg

Stay tuned as I get ready to lap my FX-60, Storm G4, Maze 4 Expert, Apogee 1U and MCW-60 blocks to 1500 grit/brasso polish. Then it's off to prep the 8800GTX's and get the right fans in the right place. I am also going to replace my Noise Isolator fan controller with a Sunbeam Rheobus as the Noise Isolator make the fans whine worse than my gf when I tell her no she can't buy another pair of shoes!!

I will post pics of the entire journey and then you can see the final polished review on the front page of XS!!

That's quite a kit there, looks to be an interesting comparison, I'm looking forward to it.

Looking good, well blagged!

woah alot of work ahead mate... good luck :D

really interesting stuuf there ...
the G80 results will be awesome :toast:
Got my fingers crossed and stayin tuned ..
... GOOD LUCK :toast:

Brasso?
I didn't see a spitoon in the pics.

subscribed :D

Good initiative, we really need more GPU block reviews!

Sounds lika a lot of fun. (And work!)

Btw how big are those pelts and where did you get them?

sorry, but I was too busy looking at the Barista expresso maker....Ive always wanted one.:slobber: :D

Looks like one hell of a comparison you've got going. May I ask how you plan to find and compare temps, and if you will also be comparing overclocks? Will you be measuring ambient temp to ensure it's the same during all testing?

I'm looking forward to seeing what you have in store for us. Thanks for your effort. :)

Quote:

---

Originally Posted by Philly_Boy

Stay tuned as I get ready to lap my FX-60, Storm G4, Maze 4 Expert, Apogee 1U and MCW-60 blocks to 1500 grit/brasso polish.

---

I was under the impression that the process used during the WB milling process is far more precise than any lapping? Please correct me if I'm wrong.

Quote:

---

Originally Posted by Istasi

I was under the impression that the process used during the WB milling process is far more precise than any lapping? Please correct me if I'm wrong.

---

According to Swiftech, their blocks are lapped to within +/- 0.0003" across the contact surface... so, basically, they're flatter than anything you'd ever get from a hand-done lapping job. As far as finish (smoothness) goes, anything beyond 600-grit or so is mainly for aesthetics--it's the flatness that really matters.

@ OP, don't you find the sunbeam 120mm fan a tad loud? I have one on and if I let it run with out a grill its a bit too shatering for my ears.

@ eva2000 - yeah. . .got a $h!t load of data to gather. . .how the addition of the GPU's affects the loop, if the second MCP355 helps the temps. . .and what happens when you got to seperate loops. I will average 10 runs of 3D Mark 05 & 06 along with fps readings from FEAR, Oblivion, Quake 4, Prey & HL2 Episode One. I will do this test for a single card as well as SLI.

It will be interesting to see how the full coverage blocks fare against the MCW-60 with ram sinks. I will be especially interested to see how they fare when I turn it up a notch and compare the FC blocks to the Apogee 1U's with a 226 watt pelt fed by a Meanwell SP-320-13.5 PSU.

I've never condensate proofed a video card before, but after reading about it and talking to OPPAINTER it doesn't seem tremendously difficult. A few coats of conformal spray on the back of the card, some clear nail varnish and dielectric grease on the back of the card and around the IHS followed by a few layers of closed cell neopreme. Styrofoam insulation and neopreme on the Apogee 1U's and I think I'll be set. I'll mod the original "775" hold down bracket from the Apogee 1U's to act as a rear hold down to put even pressure on the neopreme.

Anyway, pm me if you have any advice. . .I can use all the experienced clocker tips I can get. I cut my teeth on your I4Memory OC'ing guide, so if you have anything to add feel free!

@ clokker - no spitoon's, dude. . .gonna use the brasso as the final mirror polish for the IHS of the FX-60 and the Apogee 1U's.

@ andersson.j - they are 226 watt pelts I got from Petra's along with the cold plates.

@ poonani - yeah, I am a serious coffee snob. I use my Barista to make cafe mocha's and carmel maccciato's like 3-4x a day. Better than spending four bucks a piece @ Starbucks. They wind up costing me about 80 cents a piece to make. The Barista paid for itself in the first 6 months I owned it!

@ EnJoY - I have a Fluke mm to read ambient temps as well as voltages, etc. I will stick a temp sensor about 2" off my lowest intake fan and call that ambient. I'll adjust the room temp so all tests are run at the same ambient. I will run my FX-60 oc'ed to 3 ghz (24 hour dual prime stable) and I will be comparing temps as well as max overclocks of my volt modded 8800GTX's in single and SLI mode. See my comment to eva2000 for the data gathering/averaging.

So. . . if one block yields lower temps and a higher clock, then obviously it will score better in the 3D Mark 05/06 tests and get higher fps in the games. That's what the tests are all about! Much more real world as far as an enthusiast is concerned. If you buy one of these blocks after reading my review, you will pretty much know what to expect from it once you put it into your rig's water loop.

@ Istasi & Petra - After checking on Petra's statement (sheesh is he ever wrong? :slap: ) I do stand corrected. The Apogee 1U's base is considerably less shiny than the Storm's or MCW-60, so I may have a go at those blocks as well as hand lapping my FX-60's IHS. I think the proc will have much better surface to surface contact (and hopefully better temps!) once I expose the underlying copper in the IHS. Sheesh, I wish someone would tell us how to remove the IHS from the 8800GTX's GPU. Would love to do some naked GPU testing. . .

@ Timmay - I haven't actually mounted the Sunbeams yet, but if I cannot tame them with the Sunbeam Rheobus then I'll swap them out for Yate Loons, which I *know* are quiet. They will be on my top blow holes and the rear exhaust on the mobo tray on my MM U2-UFO case. Click on "The Wraith" link in my sig for a few pics of the rig.

Thank you one and all for the encouragement. I will be doing this test out @ Chilly1's house in Las Vegas. I'll be helping him to bang out his backlog of units by day and gathering data for this review by night. Anyway, off to dinner and some hand lapping while Grey's Anatomy is on. I'll post a few photo's of the FX-60 before and after later tonight.

Looks awesome man :)
There's really a lack of good gpu reviews out there. Keep up the good work (and the pics ;))

Great job just assembling that equipment. You are going to be a very busy guy! I like expresso machines too, but I won't go off topic :)

Quote:

---

Originally Posted by Istasi

I was under the impression that the process used during the WB milling process is far more precise than any lapping? Please correct me if I'm wrong.

---

I got two Apogee 1U Server blocks tonight that I am putting on my 8800GTXs and I did find that they are not completely flat. I drew an X with a marker and then started to lap them. The middle of the copper surface where the X crossed came off right away but it took some time to get the outer edges of the X to disappear. I finished with 1200 grit and they are now very flat and pretty. :banana:

Quote:

---

Originally Posted by BlackHawk2383

I got two Apogee 1U Server blocks tonight that I am putting on my 8800GTXs and I did find that they are not completely flat. I drew an X with a marker and then started to lap them. The middle of the copper surface where the X crossed came off right away but it took some time to get the outer edges of the X to disappear. I finished with 1200 grit and they are now very flat and pretty. :banana:

---

I just lapped my FX-60 up to 600 grit. . .all I had on hand. I am having a dificult time sourcing 1000 and 1200 or 1500 grit sandpaper locally. Is it only available online? I will attack the Apogee 1U's tomorrow.

Quote:

---

Originally Posted by Philly_Boy

I just lapped my FX-60 up to 600 grit. . .all I had on hand. I am having a dificult time sourcing 1000 and 1200 or 1500 grit sandpaper locally. Is it only available online? I will attack the Apogee 1U's tomorrow.

---

Try Pepboys.

Quote:

---

Originally Posted by bruce+

Try Pepboys.

---

I will do that tomorrow. . .thanks!

yeah another vote for Pepboy's that is where I found mine.


Hey Philly when you have all this together, can I stop by to check it out first hand?

whats going to happen with 7 packs of MC14 ramsinks?? thats alot of cash on just ramsinks...

Quote:

---

Originally Posted by phelan1777

yeah another vote for Pepboy's that is where I found mine.

Hey Philly when you have all this together, can I stop by to check it out first hand?

---

Sure. . .we'll have to make a plan to hang out when I get back from Las Vegas with the cascade. Run some benches, game a little. . .sounds like a plan for between the holidays!

Quote:

---

Originally Posted by Sneaky

whats going to happen with 7 packs of MC14 ramsinks?? thats alot of cash on just ramsinks...

---

Some I had, some are "loaners" from Swiftech for the review.

OK. . .I now know why lapping is so important. Theres no way my FX-60 IHS was flat. Take a looky:

This is the station in front of the hi def TV watching CSI (Grey's Anatomy was a rerun). The pooch (my gf's Shih-Tsu) is parked at my feet.
http://i41.photobucket.com/albums/e2...rk-Log-009.jpg

After one sheet of 320 grit. You can still see the silk screening on the IHS:
http://i41.photobucket.com/albums/e2...rk-Log-012.jpg

After the 2nd sheet of 320 grit. That silk screening is still there albeit more faint this time.
http://i41.photobucket.com/albums/e2...rk-Log-013.jpg

After 3rd sheet of 320. Now we're getting somewhere!
http://i41.photobucket.com/albums/e2...rk-Log-015.jpg

After one sheet of 400 grit to polish it a bit.
http://i41.photobucket.com/albums/e2...rk-Log-016.jpg

After 500 grit. This was the finest grit I had laying about. Gotta get some 1000, 1200, and 1500 tomorrow @ Pep Boys, a local automotive store.
http://i41.photobucket.com/albums/e2...rk-Log-024.jpg

lapped IHS is niceeee ... :D
let us see you in the FX60-mirror ... :D

Can I also lap an Intel IHS on the C2D? I mean like if I would wound the cpu somehow or somethin :stick: ...
... gee Im such a Intel n00b :D

Hey, Philly, I don't think you will find 1500 but you should find 2K, I don't remember seeing 15k grit.

so it should be 800/1000/2000

Quote:

---

Originally Posted by phelan1777

Hey, Philly, I don't think you will find 1500 but you should find 2K, I don't remember seeing 15k grit.

so it should be 800/1000/2000

---

On my way out to get that right now. . .thanks for the tip!!

Philly don't lap the Swiftech blocks... they are FLAT and you will make them worse. Swiftech has the best bases in the business.

Go to any hobby store or automotive paint place, they have more sandpaper options than you'll ever need. I think I've got everything from 1000 to 12k from well stocked hobby stores.

I wonder if anybody would be brave enough to post adverse or no affects from lapping. Seriously, looking at all of the pictures you can't help but round off the corners significantly. Has to negatively impact performance.

Quote:

---

Originally Posted by nikhsub1

Philly don't lap the Swiftech blocks... they are FLAT and you will make them worse. Swiftech has the best bases in the business.

---

Dude, I totally respect that. . .but wouldn't it be prudent to get them up to flawless mirror finishes? I am stretching the sandpaper over 3/8" plate glass and being muy very careful as I lap. The FX-60 will look flawless once I get the 800/1000/2000 grit paper on it. Your caution is that I may wind up with Apogee 1U's with bases that are not as flat as they came from the factory?

I now wish I would have taken pics of my Apogee 1U blocks last night. They were alot better than my MP-05 but not perfect...which is what I want. You can now watch TV in their reflection. I can't see lapping hurting anything if you do it correctly and take you time.

Quote:

---

Originally Posted by Philly_Boy

Dude, I totally respect that. . .but wouldn't it be prudent to get them up to flawless mirror finishes? I am stretching the sandpaper over 3/8" plate glass and being muy very careful as I lap. The FX-60 will look flawless once I get the 800/1000/2000 grit paper on it. Your caution is that I may wind up with Apogee 1U's with bases that are not as flat as they came from the factory?

---

mirror does not equal flatness. swifty bases are gauarnteed flat to +/- 0.0003".
they may not be shiny, but if they wer ethey would produce pretty much flawless geometry reflections.

make sure that you lap that FX-60 for flatness (hard to impossible with hand lapping kit) not for reflectiveness.

Quote:

---

Originally Posted by Senater_Cache

mirror does not equal flatness. swifty bases are gauarnteed flat to +/- 0.0003".
they may not be shiny, but if they wer ethey would produce pretty much flawless geometry reflections.

make sure that you lap that FX-60 for flatness (hard to impossible with hand lapping kit) not for reflectiveness.

---

Now you all have me worried. I use three fingers evenly over the foam (so I don't accidentally bend a pin) with light yet even pressure. Every 25 strokes I rotate the proc 90 degrees. Is there a "better" way to actually do the lapping? Straight back and forth...or figure eights...or circles? I slowly go back and forth in mostly a straight line.

Man, that FX is like my prized possession. . .and I don't wanna hork it just to make it look pretty.

Quote:

---

Originally Posted by Philly_Boy

Now you all have me worried. I use three fingers evenly over the foam (so I don't accidentally bend a pin) with light yet even pressure. Every 25 strokes I rotate the proc 90 degrees. Is there a "better" way to actually do the lapping? Straight back and forth...or figure eights...or circles? I slowly go back and forth in mostly a straight line.

Man, that FX is like my prized possession. . .and I don't wanna hork it just to make it look pretty.

---

i do circles while still rotating the processor 90* after every 20 or so circles

Quote:

---

Originally Posted by Philly_Boy

Dude, I totally respect that. . .but wouldn't it be prudent to get them up to flawless mirror finishes? I am stretching the sandpaper over 3/8" plate glass and being muy very careful as I lap. The FX-60 will look flawless once I get the 800/1000/2000 grit paper on it. Your caution is that I may wind up with Apogee 1U's with bases that are not as flat as they came from the factory?

---

Exactly. I'm personally not fond of the "lapping mania" that is all over forums.
If u have a problematic block, hs and u want to fix it, u can only do this with machines and not by hand.
With swiftech u don't need to do anything.
Lapping is purely for aesthetics, if u ask me... :p:

Looking forward to your results! :)

Not to be a :banana::banana::banana::banana::banana:, but PLEASE dont use brasso, after all that effort of lapping and stuff.

It acts like a isolator between metals...The mirror finish is just for show...you're not going to do any good if you get that mirror finish with brasso...its going to be worst.
Try to lap it with smaller grit sandpaper (???) and skip brasso.

Just my 2 cents.

I've heard figure eight are best but I have no idéa what that's based on.

I don't think it matters much if the corners get slightly round as long as it's pretty flat over the core where the heat is.

Quote:

---

Originally Posted by Philly_Boy

Dude, I totally respect that. . .but wouldn't it be prudent to get them up to flawless mirror finishes? I am stretching the sandpaper over 3/8" plate glass and being muy very careful as I lap. The FX-60 will look flawless once I get the 800/1000/2000 grit paper on it. Your caution is that I may wind up with Apogee 1U's with bases that are not as flat as they came from the factory?

---

No, mirror finish is NOT THE GOAL OF LAPPING EVER! It does not improve performance. Flatness is what matters, and lack of scars or nicks or imperfections. If you do lap the Swiftech blocks I guarantee you they will be less flat then before you touched them.

I have 800, 1000 and 2000 grit in hand now. . .I'll finish lapping the CPU tomorrow first thing and post pics of the final result. I really do want to have a go at the Apogee 1U's also.

One set back occurred today. I found out that one of my 8800GTX's was defective. One card did 10477 in 3D Mark 06 and the other could only manage 2390. I rma'd it to NewEgg and should have the replacement in hand by a week from Monday. So I will do all the single card testing while i wait.

Lapping a FX60 is dumb, just take the IHS off...
Http://www.sspmustang.com/OT/FS/fx60.JPG

Quote:

---

Originally Posted by fhpchris

Lapping a FX60 is dumb, just take the IHS off...

---

Nudity! :mad:

Quoting with image = loss...

Nudity for the win.

Quote:

---

Originally Posted by fhpchris

Lapping a FX60 is dumb, just take the IHS off...

---

Yeah, that is the next step. . .I have to consult the oc'ing gods to see if I need to remove it once the cascade is set up. Remember, FX-60's are cold bugged @ around -60 to -70c. Once Chilly tells me what to expect from the Extreme Frostbyte he's making for me I will make that decision.

@nikhsub1 - ok....ok....I will leave the Aprgee 1U's as they are. I did feel as if I made the IHS of the FX-60 flatter, but I hear you and others (Gabe from Swifty included) that those bases are "dead flat".

Looks good Philly! Nice Job!

I remember in one thread Cathar said lapping up to 800 grit using only water yielded best results. Also, instead of polishing with Brasso which acts as an insulator, I would finish polish with some AS5. Both will help anti-oxidation, but AS5 obviously is the best thermally, plus polishing with it will fill any micro-scratches that could later become air-pockets.

Also, it has been said before (and is also on the instruction manual) Swiftech blocks are machine lapped much more perfect than possible by hand.

Then again, having a shiny block might be worth losing a few degrees. :D I wouldn't stress over it.

Look forward to results. Subscribed.

Quote:

---

Originally Posted by entropy

I remember in one thread Cathar said lapping up to 800 grit using only water yielded best results. Also, instead of polishing with Brasso which acts as an insulator, I would finish polish with some AS5. Both will help anti-oxidation, but AS5 obviously is the best thermally, plus polishing with it will fill any micro-scratches that could later become air-pockets.

Also, it has been said before (and is also on the instruction manual) Swiftech blocks are machine lapped much more perfect than possible by hand.

Then again, having a shiny block might be worth losing a few degrees. :D I wouldn't stress over it.

Look forward to results. Subscribed.

---

Thanks for the kind words of advice. I will not be lapping my Apogee 1U's as flat is waaaay more important than pretty. I use AS5 already to fill in the micro scratches and won't use brasso after hearing of it's insulating properties.

Here's the FX after some 800 grit:
http://i41.photobucket.com/albums/e2...rk-Log-025.jpg

Another 800 grit shot:
http://i41.photobucket.com/albums/e2...rk-Log-026.jpg

Now we're making some progress after the 2nd sheet of 1000 grit.
http://i41.photobucket.com/albums/e2...rk-Log-027.jpg

Here's the finished product:
http://i41.photobucket.com/albums/e2...rk-Log-028.jpg

I can see that the corners are no longer flat. Maybe I went too fast.
http://i41.photobucket.com/albums/e2...rk-Log-029.jpg

http://i41.photobucket.com/albums/e2...rk-Log-031.jpg

WOOT baby!!! EK is about to drop their new block on us!!:toast: :woot: :toast:
Here's what my 8800's are going to look like in a week when I put a set thru their paces:

http://i41.photobucket.com/albums/e2...800gtx1qa7.jpg

http://i41.photobucket.com/albums/e2...800gtx2dq6.jpg

As my cousin says in Boston. . ."Thoese blocks are wikked haaad core!"

Quote:

---

Originally Posted by Philly_Boy

Yeah, that is the next step. . .I have to consult the oc'ing gods to see if I need to remove it once the cascade is set up. Remember, FX-60's are cold bugged @ around -60 to -70c. Once Chilly tells me what to expect from the Extreme Frostbyte he's making for me I will make that decision.

@nikhsub1 - ok....ok....I will leave the Aprgee 1U's as they are. I did feel as if I made the IHS of the FX-60 flatter, but I hear you and others (Gabe from Swifty included) that those bases are "dead flat".

---

Quote:

---

Remember, FX-60's are cold bugged @ around -60 to -70c.

---

You are kidding right? you MIGHT get to -50c if you are even lucky and I am talking 190-240 HTT here, so you better be ready to use some high multis...

Quote:

---

Originally Posted by fhpchris

You are kidding right? you MIGHT get to -50c if you are even lucky and I am talking 190-240 HTT here, so you better be ready to use some high multis...

---

Do you have any links to threads that discuss this? I didn't realizethat the cold bug occurred at -50 or so and the HTT was so limited. I guess I'll be learning how to throttle back the cascade to keep idle temps above that threshold. My mobo has dividers made for running higher multi's & low HTT and "over dividing" the ram to 10/9 or 8/7 or 6/5, etc to still get the best performance from the ram. Thanks for the info.

Prepping the Swiftech MVR-220 rad with Yate Loon D12SL fans (these are the purple framed YL's with blue LED's. . .only Petra's has them).
http://i41.photobucket.com/albums/e2...-Log-002-1.jpg

http://i41.photobucket.com/albums/e2...-Log-003-1.jpg

Remounting the lapped-very-nicely FX-60:
http://i41.photobucket.com/albums/e2...-Log-001-1.jpg

Remounting the Storm:
http://i41.photobucket.com/albums/e2...-Log-004-1.jpg

http://i41.photobucket.com/albums/e2...-Log-005-1.jpg

I moved all the aluminum Evercool fans on to the PA 120.3, and threw a few very quiet Sunbeam's on the top blowholes and rear mobo side exhaust. I will have four of the D12SL's on the secondary rad (be it the Swifty MCR220 or the new BI GTX 240 when it arrives in a few days) in a push/pull configuration. I tightened up all the fan wiring and installed a Sunbeam Rheobus to tame the fan noise. It is **FAR** better than the POS "Noise Isolator" I had before! And the blue LED's (to denote the fans are @ 7v or below) fit in with the black, silver and blue theme of the case!

http://i41.photobucket.com/albums/e2...-Log-006-1.jpg

I will shut the system down now and start prepping my lone 8800GTX (til the other one gets back from rma. . .:( ) with the Swifty MC14 ramsinks and a MCW60.

Tonight I'll begin gathering data!:woot: :toast: :woot:

Waiting for results...

Quote:

---

Originally Posted by JoeBar

Waiting for results...

---

It'll be about 2-3 weeks before the final results are tabulated and collated into something we can all understand. The EK blocks just shipped today. . .so it'll be 12-14 days before I even have them to play with.;)

Stay tuned as it unfolds. . .

does anyone know if the G80 clip for the MCW60 works also with MCW50 block ?

Quote:

---

Remounting the lapped-very-nicely FX-60:
http://i41.photobucket.com/albums/e2...-Log-001-1.jpg

---

You may have applied too much AS5 if two surfaces are flat. just my :2cents:

Quote:

---

Originally Posted by bruce+

You may have applied too much AS5 if two surfaces are flat. just my :2cents:

---

I put 75% as much last time and it didn't cover the whole area of the IHS. I used a tad more this time to ensure total and even coverage.

Quote:

---

Originally Posted by nikhsub1

No, mirror finish is NOT THE GOAL OF LAPPING EVER! It does not improve performance. Flatness is what matters, and lack of scars or nicks or imperfections. If you do lap the Swiftech blocks I guarantee you they will be less flat then before you touched them.

---

you tell em nikhsub :mad:

Quote:

---

Originally Posted by Philly_Boy

I can see that the corners are no longer flat. Maybe I went too fast.

---

:brick:

Don't lap with sandpaper if you want flat things. You can try to stretch it flat all you wont it wont do you much good.

Buy a proper lapping plate:
http://www.lapmaster.com/accessories...s/default.html

I've yet to run into problems lapping with sandpaper on plate glass... nor have countless others, so *shrug* I don't think everyone who wants to lap needs a lapping plate... though it would be a killer thing to have if you want to be regarded as the absolute lapping god :banana:

Quote:

---

Originally Posted by epion2985

you tell em nikhsub :mad:



:brick:

Don't lap with sandpaper if you want flat things. You can try to stretch it flat all you wont it wont do you much good.

Buy a proper lapping plate:
http://www.lapmaster.com/accessories...s/default.html

---

Just because you don't "see" or "feel" you have a problem doesn't mean you don't. Remember some posts above people think that they lap swiftech blocks smoother then they come, which we know it just not true.

Actually if you don't want to buy a plate, which is a good investment in this hobby, you can always walk down to your local machine shop and they will lap it for you for a modest fee much better then you ever could.

OK we got the 8800GTX prepped and in the waterloop; Check it out:

Say goodbye to the stock cooler:
http://i41.photobucket.com/albums/e2...-Log-009-1.jpg


Now we have to remove the 13 screws to remove the cooler:
http://i41.photobucket.com/albums/e2...rk-Log-010.jpg


Off it comes. I have done this once already replacing the crappy thermal tape with T411.
http://i41.photobucket.com/albums/e2...rk-Log-011.jpg


Next up it cleaning the IHS and other hot bits to prep them for the MCW-60/ram sinks.
http://i41.photobucket.com/albums/e2...-Log-012-1.jpg


All nice-n-clean.
http://i41.photobucket.com/albums/e2...-Log-013-1.jpg


I ran into a problem in that the ram sinks I had planned to cover the VRM's with did not sit flat when trying to
cover two sets of VRM's. So, I broke out the handy dandy Dremmel and made one into three!
http://i41.photobucket.com/albums/e2...rk-Log-014.jpg


I cut little pieces of T411 to attach them and now they look mighty trick!
http://i41.photobucket.com/albums/e2...-Log-015-1.jpg


I heated the thermal tape with the high setting of the gf's hair dryer for 30 seconds and pressed them firmly
onto the mem modules. Not one is loose. Not one!!
http://i41.photobucket.com/albums/e2...-Log-016-1.jpg


AS5 on the IHS. Don't flame me over my application method til you see the temps I get! This is the
approximate area the MCW-60 will contact the IHS.
http://i41.photobucket.com/albums/e2...rk-Log-017.jpg


MCW-60 attached
http://i41.photobucket.com/albums/e2...rk-Log-018.jpg


Another view to show tubing clearance.
http://i41.photobucket.com/albums/e2...rk-Log-019.jpg


Lookin nice-n-purdy in the loop.
http://i41.photobucket.com/albums/e2...rk-Log-020.jpg

Here's a preliminary result. From 70C idle with stock cooler and stock thermal tape to 51c idle with T411
to 37c idle on water. Guess what I'll be doing later tonight?
http://i41.photobucket.com/albums/e2...C-GPU-Idle.jpg

Quote:

---

Originally Posted by epion2985

Just because you don't "see" or "feel" you have a problem doesn't mean you don't. Remember some posts above people think that they lap swiftech blocks smoother then they come, which we know it just not true.

Actually if you don't want to buy a plate, which is a good investment in this hobby, you can always walk down to your local machine shop and they will lap it for you for a modest fee much better then you ever could.

---

I will remember that next time I have a lapping project. There is a machine shop within a mile of my house!

woah sweet.. happy clocking :)

Quote:

---

Originally Posted by fhpchris

Lapping a FX60 is dumb, just take the IHS off...

---

damn that is my old old FX60..... i don't even have my pic anymore but saw someone posted in here lmao

Quote:

---

Originally Posted by eva2000

woah sweet.. happy clocking :)

---

I actually got it to 715/1140 before I had to drain the loop and pack it up to ship it to Las Vegas. I am going out to meet FUGGER and Chilly1. I'm going to try to help Chilly get some of his backlogged units built and shipped. I'll be gathering data for my little G80 block review in my spare time. . .

Philly Boy,

I don't know who Chilly1 is, but I understand you are trying to get him back on his feet. I am sure its a worthy cause because of your involvement. Please let me know if I can help, but I guess I would need to understand the cause. Chilly1 seems very reputable.

Chilly1 is a phase-change god. Without him phase-change wouldn't be what it is today. Just browse the Phase Change section for a while.

Yes. He's as good as they come in terms of reputation. I still don't understand what happened to him and why he needs a contribution.

Quote:

---

Originally Posted by IanY

Yes. He's as good as they come in terms of reputation. I still don't understand what happened to him and why he needs a contribution.

---

Maybe the link in Philly's sig?

Quote:

---

Originally Posted by EnJoY

Maybe the link in Philly's sig?

---

Yeah, hit the linky in my sig and read the thread. I am at the airport now on my way to Las Vegas to help him out.

I don't know the guy at all, but what the h3ll, people who are down and out sometimes can use a boost.

Philly I've done the blob of AS5 in the center and the smear it around smooth. I always get the same temps no matter what way I do it.

Philly, you were at PHL INT? What terminal did you depart from? I was working down there today, currently on a contract for US AIR

Quote:

---

Originally Posted by phelan1777

Philly, you were at PHL INT? What terminal did you depart from? I was working down there today, currently on a contract for US AIR

---

I was at terminal C28 flight 797 to Las Vegas that left at noon. Where were you?

nice :) love cooling solutions for g80's... nice temps.

subcribed.

Quote:

---

Originally Posted by Philly_Boy

I have 800, 1000 and 2000 grit in hand now. . .I'll finish lapping the CPU tomorrow first thing and post pics of the final result. I really do want to have a go at the Apogee 1U's also.

One set back occurred today. I found out that one of my 8800GTX's was defective. One card did 10477 in 3D Mark 06 and the other could only manage 2390. I rma'd it to NewEgg and should have the replacement in hand by a week from Monday. So I will do all the single card testing while i wait.

---

in addition to what nikhsub1 said i'd like to add that 600 grit paper is as far as you should go if you really want to lap as that finish will give the best results consistently according to Cathar's testing :toast:

Quote:

---

Originally Posted by dinos22

in addition to what nikhsub1 said i'd like to add that 600 grit paper is as far as you should go if you really want to lap as that finish will give the best results consistently according to Cathar's testing :toast:

---

I wasn't satisfied with the result from 600 grit so I carefully went to 800, then 1000, then 2000 grit. I saw a 2c gain in temps for lapping the proc.

I am a bit stalled as my new 8800GTX is due to arrive today and I'll get back into gathering data. Patience, my friends. . .I'll be back into this by the weekend.

bah I just stripped my 170, core to block contact.

I understand that the newer c2d ihs are almost welded on nowa days.

Here are some anecdotal findings:

Ambient temp was 21.1C for all tests.

Just CPU and NF4 chipset in loop:
CPU idle (stock clock/v-core)=23c; loaded=34c (Orthos 24 hour);
CPU idle (overclocked with 1.5v v-core)=27c; loaded=36C
Best CPU oveclock=3060 mhz

Adding one 8800GTX (stock) with MCW-60 and MC14 ramsinks to the loop:
CPU idle (stock)=25c-26c; Loaded=35c;
CPU idle (oc'ed with 1.5v v-core)= 28c loaded=38c
Best CPU overclock=3060mhz.
Altho the ram sink above the ramdac on the 8800GTX got quite hot to the touch the other ram sinks were cool to the touch with a 120mm fan blowing directly from the back of the card towards the front.

I took the 8800GTX out of the loop, stripped it down and replaced the MCW-60 with a Danger Den NV-88 full coverage block and mounted it back into the loop:
CPU idle (stock)=31c; Loaded=41c
CPU idle (oc'ed)=36c; Loaded=44c
Best CPU oc=2929mhz
I could not get the CPU to go any higher than 2929 mhz. Anything above that failed Orthos within minutes.

I am going to try adding a second MCP355 pump (Petra's top) and T411 tape to the memory modules to see if that helps temps/max CPU OC when using the NV-88. I might also experiment with switching out the Storm for a less restrictive DD TDX block. I'll have more to follow along with some 3D Mark 06 runs when my rig gets back from Vegas this coming Wednesday. My second card is due back late next week from RMA. Then I can see what adding 2 blocks into the loop does to temps.

So what happened? did you test the EK blocks?

I am @ CES in Las Vegas this week coming. I am looking to acquire a few other blocks to make the review more complete. I am almost done testing the NV-88's and the EK's are up next.

I envy you guys that have the opportunity/find the time to attend CES.

wow, the DD block increased the cpu temps very very much

is it due to the fact that it is cooling more components on the vga or due to the pressure drop

I'm going to wager that the difference is more from the added heat than the pressure drop. We'll be able to see with more detail once the second pump is added to the loop.

EDIT - So with all of these results you're using a dual 120mm radiator? I'm not sure if I just missed it, but which 120.2 is it, the MCR?

Keep up the great work!

Quote:

---

Originally Posted by AllAgainstPaul

Keep up the great work!

---

Well, my rig showed up at my doorstep damaged and broken. It looks like it's a total loss. The case is damaged as are many of the hardware components. Luckily it was insured. . .but I really won't get the true value of the rig back. I am pissed. . .this is a huge set back.

I will try to settle with FedEx and then convert to an Intel Conroe based rig. I will have to regather the data all over again.

This sux big time!!

Wow that's some terrible luck Philly :( GL with the insurance claim. I look forward to more testing. The initial results are... surprising. I can't help but LOL at how bad the full cover block is. I mean, I knew they were bad, but a 100mhz CPU loss is horrible.

Quote:

---

Originally Posted by Philly_Boy

Well, my rig showed up at my doorstep damaged and broken. It looks like it's a total loss. The case is damaged as are many of the hardware components. Luckily it was insured. . .but I really won't get the true value of the rig back. I am pissed. . .this is a huge set back.

I will try to settle with FedEx and then convert to an Intel Conroe based rig. I will have to regather the data all over again.

This sux big time!!

---

Damn! I feel it for you... I'd be pissed off too... Yeah, GL whit the insurance claim...

I hope this thing goes fast because i realy enjoy your tests. And shame on fedex to:)

Quote:

---

Originally Posted by phelan1777

Hey, Philly, I don't think you will find 1500 but you should find 2K, I don't remember seeing 15k grit.

so it should be 800/1000/2000

---

there is 1500 grid.. i have, very hard to come by. allso there is 2500 and 3000. but who really needs them anyways:cool: 2000 is more than adequate for the jop.

and Philly really intresting review you got going here :toast: most definitely will be lurking around.
btw are going to be able to keep all that stuff is so you are one lucky bastard keep up the great work.

Philly,

Sorry to hear about the shipping damage. I just hope that you had it insured for more than it is worth to allow for price increases and your labor.

Quote:

---

Originally Posted by Solarfall

there is 1500 grid.. i have, very hard to come by. allso there is 2500 and 3000. but who really needs them anyways:cool: 2000 is more than adequate for the job.

---

What the???? OSH has 1500 grit, 99cents for a 8.5"X11" sheet!!!

Quote:

---

Originally Posted by Philly_Boy

Well, my rig showed up at my doorstep damaged and broken. It looks like it's a total loss. The case is damaged as are many of the hardware components. Luckily it was insured. . .but I really won't get the true value of the rig back. I am pissed. . .this is a huge set back.

I will try to settle with FedEx and then convert to an Intel Conroe based rig. I will have to regather the data all over again.

This sux big time!!

---

Could we get a few photos to see how badly damaged they are?

Very sorry to hear about the rig, Philly. :(

FedEx needs to pay up.

that's really too bad :(

please show us some pics of this mess

Is there anywhere else US residents can purchase the EK 8800GTX waterblock besides Petra's? I really want to order my parts ASAP. :(

Quote:

---

Originally Posted by darcon

Is there anywhere else US residents can purchase the EK 8800GTX waterblock besides Petra's? I really want to order my parts ASAP. :(

---

EK directly I think would be your best bet.

incase you all didn't see, so far Philly has revived most if not all the hardware that was inside the case and is currently benching to start his reviews.

I got a chance to help him take down the rig and see the damage first hand.

None of it is like torn metal, or holes where there should be none. There are broken screws, dented plates, and corners and scratches in the powder coating.

Quote:

---

Originally Posted by darcon

Is there anywhere else US residents can purchase the EK 8800GTX waterblock besides Petra's? I really want to order my parts ASAP. :(

---

If Petra has them in stock, then you'll get in 2-3 days. Otherwise you get them direct from EK. . .and that's 2-3 weeks shipping time. You decide. . .

On a side note. . .my rig is priming now @ 3ghz. . .everything is out of the case and sitting beside me. I'll have it this way til the reviews lighten up.

Gotta finish the G80 waterblock review. . .then it's the FuZion vs the Storm G4 vs the ApogeeGT with AMD and Intel test beds. . .then perhaps it'll be a memory face off to see if those new Corsair Dominators really Dominate against offerings from OCZ, Patriot, and Crucial. Finally I'll be doing presentation of Quad FX from an enthusiast point of view. I think Quad FX is being given a bad rap. . .sure it's more expensive than a Kentsfield set up, but I think it clocks better than we're being led to believe. I guess we'll find out in a month or so. Sheesh somewhere along the way will be the Tri State Bench Fest and after that the first annual Philly Phrag Phest. Needless to say the winter will be busy!