Big Typhoon - Lapping Pics - Temp Curves - Before and After

malicious

Supreme [H]ardness
Joined
Apr 8, 2005
Messages
4,772

Welcome

Ok, so this thread has become my little project of installing a laped BigT on my new xp-mobile 2400+ (45w). We'll see what the "after" temps look like and what the overclocking increase is ... if any.

This thing is huge!


Before Temps:
Runing volcano 7+ w/ Panaflo M1A 120 adapted to 80 interface
This is running Prime95 Priority 10 untill the temps leveled


Before: WoW That Sucks!


After about 6 hours of 400, 600, and 800 grit papers


Nearly completed 2000 grit





After 10,000 grit polish





bent some fins with the air compressor

hand sander

polishing grit/gunk

stock fan


PERFORMANCE RESULTS: 12 Degree C Lower - Stock 1.8 Full Load ----- 50C -> 38C



INSTALLED:



OVERCLOCK TEMPS: Full Load Between 41C and 42C at 2.25 w/ 1.65vcore(max limited)
See above for stock 1.8ghz full load at 38C



FS: Ti-83 Plus & Ti-89
[url]www.easypckits.com
for all your lapping needs
[/URL]
 

CRaZYMoFo

Limp Gawd
Joined
Jun 18, 2005
Messages
489
Does lapping normally take that long? o_O
Im planning to lap my BT as well. Awaiting temp results :D
 

malicious

Supreme [H]ardness
Joined
Apr 8, 2005
Messages
4,772
Starfox said:
Lapping NEVER took me more than an hour at the MOST.

That's because you never laped a BigT. The stock Tt surface is the worst I've ever seen. Honestly, I still didn't get past the orginal surface. I can still see hair line scratches in the original direction. They were deep to begin with.

Also, I must mention that my base was not flat. I spun the HSF on the glass so it never skiped once. The center of my sample was for sure sunken compared to the parimeter.

With regard to the other guy thinking of laping a BigT. I would say go for it. It takes some work but the performance gain will probably be great. I'll be installing mine tomarrow. Have to remove the mobo and all. Unfortunately I don't have any pre-lap temps as I will be replaceing my very antiquated volcano 7+. Also, I'm going to be runing it with a Panaflo M1A vs the stock 120mm fan. (86.6cfms vs. stock 54.4cfms)... I'll leave the stock fan on there .. get temps .. change to M1A.. get temps ... to see how much increaseing cfms will change the "power" of the cooler.

 

malicious

Supreme [H]ardness
Joined
Apr 8, 2005
Messages
4,772
We'll that's it for tonight folks... time for little herbal refreashment.
 

Ace123

Anal Rooter
Joined
Oct 3, 2003
Messages
404
Heres what 15 minutes will yield if you know how to lap a heatsink properly. all the unecessary sanding will yield nothing but time lost. 600 grit and a good buffer (I work automotive) with some 3m coarse cut compound will make this thing so shiny it will make a blind guy squint. Then hit it with some swirl remover. mmhmm.
he is right though, the pad on this thing is horrible stock.
 

TheRapture

Supreme [H]ardness
Joined
May 31, 2000
Messages
6,871
Looks like I will be lapping mine, if it gets better, hell, I will whistle Dixie loud and clear till the cows come home.... :p
 

malicious

Supreme [H]ardness
Joined
Apr 8, 2005
Messages
4,772
Ace123 said:
Heres what 15 minutes will yield if you know how to lap a heatsink properly. all the unecessary sanding will yield nothing but time lost. 600 grit and a good buffer (I work automotive) with some 3m coarse cut compound will make this thing so shiny it will make a blind guy squint. Then hit it with some swirl remover. mmhmm.
he is right though, the pad on this thing is horrible stock.
Ha ... I got PWNED Oh well
 

wwswimming

n00b
Joined
Mar 15, 2005
Messages
48
shiny does not equal flat

i've worked about 22 years in Silicon valley, about half that time designing heat sinks.

this doesn't mean i know everything about heat sinks. it just means i sit in rooms and offices with other engineers getting paid to talk about heatsinks (when we're not shopping on Newegg).

anyway, you need shiny AND flat, to get good contact between the heatsink surface and the top of the chip.

if you got a friend with a machine shop, i suggest taking your heat sink in to measure both flatness and surface finish.

anyway, i really got to admire your effort. kind of builds up the old arm muscles ?!

what kind of music did you listen to - or did you have the TV on.

sh*t. 6 hours. God.

BTW - if the surface you were working on was FLAT - like a granite slab you find in a metrology lab in a machine shop - the time was not wasted, if it improved the flatness.
 

Ace123

Anal Rooter
Joined
Oct 3, 2003
Messages
404
Just get a peice of flat steel like I did. Wrap a peice of wet sandpaper around it and rub the heatsinks thermal patch on it in circular motions. not only will you rub out all the uneven spaces and the machining lines but it will also be flat. No messy machine shop bills and you get to keep all the thrills.
 

pstang

[H]ard|Gawd
Joined
Jan 26, 2005
Messages
1,601
Starfox said:
Lapping NEVER took me more than an hour at the MOST.
some hsf are flatter than others... and that is what takes the most time is getting it flat.... the 2 thermalright's (AX7 and XP-90) i've done took between 3 and 4hrs... of course i learned my lesson from the AX7 not to use too low of a grit sand paper ;)
 
Joined
Oct 2, 2003
Messages
2,173
wwswimming said:
shiny does not equal flat

i've worked about 22 years in Silicon valley, about half that time designing heat sinks.

this doesn't mean i know everything about heat sinks. it just means i sit in rooms and offices with other engineers getting paid to talk about heatsinks (when we're not shopping on Newegg).

anyway, you need shiny AND flat, to get good contact between the heatsink surface and the top of the chip.

if you got a friend with a machine shop, i suggest taking your heat sink in to measure both flatness and surface finish.

anyway, i really got to admire your effort. kind of builds up the old arm muscles ?!

what kind of music did you listen to - or did you have the TV on.

sh*t. 6 hours. God.

BTW - if the surface you were working on was FLAT - like a granite slab you find in a metrology lab in a machine shop - the time was not wasted, if it improved the flatness.
Perhaps you could answer a question for me then. I read somewhere that using polishes on the base of a heatsink actually inhibits its performance. This is due to the microscopic polishing particles getting stuck in the small grooves you have on the base of the heatsink (it isn't completely flat, after all) to help make it shiny. Within these grooves is where your thermal compound, such as arctic silver 5, should be.

The above seems logical, and if it is right, I wonder why so many people polish the bases of their HSFs after lapping. Sure, it is shiny, but no one ever sees it, and it counteracts your hour(s) of lapping work.

Like I said before, I am not sure if this is true, but I would definetly lean towards it being true rather than false.
 

malicious

Supreme [H]ardness
Joined
Apr 8, 2005
Messages
4,772
I'll be installing the BigT today. This will only provide a comparision between my current volcano 7+ and the laped BigT. No pre-lap temps were taken. However, I will have data comparing 54.4cfms(stock) vs. 86.6(M1A).
 

botld92z

Limp Gawd
Joined
Aug 3, 2005
Messages
155
wwswimming said:
if you got a friend with a machine shop, i suggest taking your heat sink in to measure both flatness and surface finish.
I'm going to cheat and lap mine with a CNC mill, lol. It'll take longer to set up the piece in the jaws than it will to actually lap it... :eek:
 

Starfox

2[H]4U
Joined
May 6, 2002
Messages
2,220
You guys that are making it shiny are idiots.

THE OBJECT of lapping is not to get BLING BLING, it is to make the surface of the heatsink flat so that it has better contact with the core of the cpu.

When you apply thermal compounds like artic silver, they fill in all of the microscoping pots and crevices on the heatsink and core creating a flush contact to more effectively trasnfer heat.

But now instead of having a thermal compound, you idiots are fucking brasso and automotive wax there. Good going!

At least it is blingy blingy, especially on a part of the computer that you will NEVER see.
 

wwswimming

n00b
Joined
Mar 15, 2005
Messages
48
seeing your question about heat sink compound - well, yeah, it's the measurable temperature difference that counts.

and, admittedly, these processors can take a lot of abuse. i once ran my abit ic-7 max3 for about a half an hour - without a fan. it was burn your finger hot. my vision is poor - once i noticed the thing was unplugged, i plugged it in, and it has worked ever since, running pro-e on it right now.

in engineering calc's. that get presented at design reviews, air is always considered a "demon" in the thermal circuit. for example, the last place i worked, the thermal conduction was only considered within like a quarter inch radius of where 2 screws connected 2 surfaces. when i represented air as a conductor in thermal circuits, it was frowned upon by the group manager. i'm talking about like for presentations to other engineers working for military customers. paid for by your tax dollars. Thank you ! those hour long breaks we took every afternoon about 4:00 PM were necessary for our psychological health - if we have any. :-D

anyway, i just kind of do what tom's hardware does, like when they ran a p4 northwood up to 5 GHz using liquid nitrogen. they say use arctic silver ceramique, i use arctic silver ceramique.

most of it does get squeezed out. yes, theoretically, particles from the grease (i could never use that term in a design review) could increase the distance between the metal surfaces.

but, ultimately, whatever works (i used toothpaste once. maybe more than once). if the difference in temperature is less than a degree, it's a sign that you're giving the CPU more TLC than i gave the CPU on that abit ic7 max3.
 

zer0signal667

2[H]4U
Joined
Jan 11, 2002
Messages
3,847
Starfox said:
You guys that are making it shiny are idiots.

THE OBJECT of lapping is not to get BLING BLING, it is to make the surface of the heatsink flat so that it has better contact with the core of the cpu.

When you apply thermal compounds like artic silver, they fill in all of the microscoping pots and crevices on the heatsink and core creating a flush contact to more effectively trasnfer heat.

But now instead of having a thermal compound, you idiots are fucking brasso and automotive wax there. Good going!

At least it is blingy blingy, especially on a part of the computer that you will NEVER see.
Here's a perfect example of how NOT to express your opinion.
 

mastercheeze

2[H]4U
Joined
Jul 23, 2004
Messages
2,796
wwswimming said:
BTW - if the surface you were working on was FLAT - like a granite slab you find in a metrology lab in a machine shop - the time was not wasted, if it improved the flatness.

I lapped my ninja today - used the marble countertop in my bathroom. Laid the sandpaper on top of it, add a sprinkle of water, add a tiny drop of liquid dishsoap, run it back and forth about 10 times, turn 90 degrees, repeat, turn, repeat, etc etc

400 -> 600 -> 1000 -> 1500

It was funny because as I looked at the base after about 15 minutes of 400 I noticed an oval shape in the middle of the heatsink that wasnt getting sanded as much. Temps are maybe 1-2C lower. I use Ceramique, so it should drop over the next week as the TIM cures. Also using a yate-loon on the ninja.
 
Joined
Oct 2, 2003
Messages
2,173
prtzlboy said:
just clean it when you're done polishing it.

Crisis averted.
Cleaning with alcohol still doesn't necessarily clean out all of the microscopic particles left behind. AS5 pretty much permanently changes the color of the bottom of a HSF, for instance. Even if you clean it well with alcohol, it stays there.

Other compounds such as brasso and polishing agents are no different, if not worse. These agents are specifically designed to stay in the small crevices that the human eye cannot see.

Just because it looks clean, it doesn't mean it is.

Overall, it is best to have a flat surface. For metal objects, if a surface is flat, it is usually shiny. A mirror-quality finish is not necessary, however, and reflectivity certainly isn't an indication of flatness (although it can help you gauge the flattness of your base by comparison of object and image).

I would strongly recommend NOT using any sort of polishing cream on the base of your HSF. Chances are it will hamper your thermal transfer.
 

zer0signal667

2[H]4U
Joined
Jan 11, 2002
Messages
3,847
MGDMN said:
What r u referring to Zer0? I know your a mod but I see no rule against voicing your opinion.

You guys that are making it shiny are idiots.
There's no need for stuff like this. There's nothing wrong with voicing your opinion, as long as it's done in a civil manner.
Please continue with the thread, this shouldn't turn into a lecture... :p
 

MGDMN

2[H]4U
Joined
Sep 30, 2003
Messages
2,433
zer0signal667 said:
There's no need for stuff like this. There's nothing wrong with voicing your opinion, as long as it's done in a civil manner.
Please continue with the thread, this shouldn't turn into a lecture... :p
Your right, it shouldnt.

 

malicious

Supreme [H]ardness
Joined
Apr 8, 2005
Messages
4,772
mastercheeze said:
I lapped my ninja today - used the marble countertop in my bathroom. Laid the sandpaper on top of it, add a sprinkle of water, add a tiny drop of liquid dishsoap, run it back and forth about 10 times, turn 90 degrees, repeat, turn, repeat, etc etc

400 -> 600 -> 1000 -> 1500

It was funny because as I looked at the base after about 15 minutes of 400 I noticed an oval shape in the middle of the heatsink that wasnt getting sanded as much. Temps are maybe 1-2C lower. I use Ceramique, so it should drop over the next week as the TIM cures. Also using a yate-loon on the ninja.

gotta get it flater
 

malicious

Supreme [H]ardness
Joined
Apr 8, 2005
Messages
4,772
the 12c drop was going from a volcano7+ to a laped Big Typhoon

....................................... :rolleyes:
 

seanmcd

[H]ard|Gawd
Joined
Jan 20, 2003
Messages
1,433
TehRoflcopter said:
Rofl, no, he's not talking about Tea, lol.
He very well could be talking of "tea" (which is what they called "it" back in the day)

I'm thinking of sanding (I really hate the word lapping) my XP-90, as it's fairly rough and my temps are too high with a dual core Opteron, and I am running out of things to do while I avoid working on the house/yard
 
Top