microfin wb made from akasa ak360/370

leejsmith · 03-11-2003, 12:26 PM

i would appricate any comments / tips you guys might have on this wb i made from the base of the akasa ak360 copper cooler.

would removing even more of the fins on the outer edges at and at the same time making the inside of the top smaller improve performance ?

i also think if the initial copper cooler didnt have the gap in the center that would improve the wb but i dont have the ability / tools to make my own copper base plate with the micro fins.

i have tested it and the setup worked very well. I have a 900l/h pump. heater core almost the same as the chevvy running on a p42.4

room temp 17.9C
cpu 19C
water 18.3C
m/b 19C

after 1 hour

room temp 18.3C
cpu 21C
water 20.2C
m/b 26C

i ran sandra cpu test 60 times
room temp 18.3C
cpu 26C
water 22C
m/b 27C

after 5 mins back at idle

room temp 18.4C
cpu 21C
water 20.9C
m/b 26C

hope you guys can help

Lee

bigben2k · 03-11-2003, 12:42 PM

Quote:

Originally posted by leejsmith
hope you guys can help

Lee

Help with what? What's the problem?

You should note that your CPU temps are probably off: getting CPU temps within 1 deg of the water would be a miracle.

leejsmith · 03-11-2003, 12:54 PM

Quote:

Originally posted by bigben2k
Help with what? What's the problem?

would removing even more of the fins on the outer edges at and at the same time making the inside of the top smaller improve performance ?

i also think if the initial copper cooler didnt have the gap in the center that would improve the wb but i dont have the ability / tools to make my own copper base plate with the micro fins.

You should note that your CPU temps are probably off: getting CPU temps within 1 deg of the water would be a miracle.

the cpu temps are from asus probe.

bigben2k · 03-11-2003, 03:02 PM

That depends: with the gap, how tall are the fins? If they're more than 5mm, you'd have a hard time improving anything.

leejsmith · 03-11-2003, 03:43 PM

the taller fins are 8mm and the lower are 4mm

the gap is 1.5mm

overall fin area as 45mm x 30mm

the base is 3mm

i use 3/8 barb

leejsmith · 03-13-2003, 05:54 AM

so are you saying there is nothing i can do to make this any better ?

wymjym · 03-13-2003, 06:58 AM

You can always improve but,
if you are getting the cpu within 4 degrees of the water in your worst environment (benching-gaming) you have the best cooling water block I have heard off.
I'm not familiar with the PIV temp swings but most AMD cpus (in well configured water systems) see the cpu temp around 8~10 degrees above the water temp.
wj

leejsmith · 03-13-2003, 07:04 AM

i am doing some testing on the temp probes i used to make sure the readings are correct.

see below for my testing results.

info at overclockers.com on the wb

MadDogMe · 03-13-2003, 08:25 AM

Thermaltake make a 1U 'skived' copper HS without the central gap, get them at theoverclockingstore.co.uk ...

For more detailed pics without the fan search for some of Volenti's posts in this forum~section...

*****************************************

I'd make the fin area a little bit bigger than the cores size but no bigger, I think reducing the height of the fins would help as well, 3 to 5mm maybe?. Also making a 'jet' so it concentrates the water and makes it the right width like Cathars 'whitewater wonder!' would help

...

The Thermaltake 1U is a firm favourite of mine at the moment, it's a H2o block maker/modders dream!

, I'd like to cross cut one into 'pins' and incorporate it in a radial design. soon, soon, but I'm busy at the mo' :shrug: ...

leejsmith · 03-13-2003, 08:38 AM

yes that looks perfect.

just out of interest how would you make a top ?

i dont have any way to make a copper top so i used acrylic but the more i read on this subject the more reasons i find not to use acrylic .

I have a p4 so if i cut the fins down to be a little bigger than the p4 heat spreader and about 4mm in height that should be ok.

bigben2k · 03-13-2003, 08:40 AM

Quote:

Originally posted by leejsmith
the taller fins are 8mm and the lower are 4mm

the gap is 1.5mm

overall fin area as 45mm x 30mm

the base is 3mm

i use 3/8 barb

That's good, and no, you probably can't make it much better.

The only thing you are able to change is the inlet geometry and orientation. You are otherwise limited by the HS properties.

leejsmith · 03-13-2003, 08:44 AM

Quote:

Originally posted by bigben2k
That's good, and no, you probably can't make it much better.

The only thing you are able to change is the inlet geometry and orientation. You are otherwise limited by the HS properties.

thats why i chose the skived copper hsf they should be a good base to work with.

so would the thermaltake skived base be better as it does not have the gap in the middle.

bigben2k · 03-13-2003, 08:47 AM

Probably not: 4mm is still very effective. The limitation here is in the fin spacing: it's optimized for air, not water.

leejsmith · 03-13-2003, 08:50 AM

what would be the best gap between each fin for water.

i found on the first attempt on the wb you could see loads of bubbles between the fins and no matter how long i left it or shook the wb they never moved. Thats why i lowerd every other fin.

bigben2k · 03-13-2003, 09:04 AM

Cathar figured that the best fin to channel width ratio is ideally between 0.75 and 1.5. If you measure your fin gap, I think you'll find it to be a number of times bigger than the fin.

For the bubbles, an additive would probably fix that: what are you using now?

If you just want to test it, a couple of drops of Jet-Dry should take care of it, but that's just for a test: Jet-Dry might not be the best thing to use, permanently.

Also, like I said, you can play with the inlet geometry and orientation, but you should be already pretty close to the maximum you're going to get out of that heatsink.

leejsmith · 03-13-2003, 09:12 AM

Quote:

Originally posted by bigben2k
Cathar figured that the best fin to channel width ratio is ideally between 0.75 and 1.5. If you measure your fin gap, I think you'll find it to be a number of times bigger than the fin.

For the bubbles, an additive would probably fix that: what are you using now?

If you just want to test it, a couple of drops of Jet-Dry should take care of it, but that's just for a test: Jet-Dry might not be the best thing to use, permanently.

Also, like I said, you can play with the inlet geometry and orientation, but you should be already pretty close to the maximum you're going to get out of that heatsink.

i only have some cooling system conditioner and anti corrosive stuff from a local car super store. if and when i install it into a case i will get wetter water.

what do you think about using acrylic for the top is that a bad idea ?

bigben2k · 03-13-2003, 09:16 AM

Quote:

Originally posted by leejsmith
i only have some cooling system conditioner and anti corrosive stuff from a local car super store. if and when i install it into a case i will get Water Wetter .

what do you think about using acrylic for the top is that a bad idea ?

Are you using distilled water? (see the article).

An acrylic top is fine.

leejsmith · 03-13-2003, 09:18 AM

i am using de-ionised water.

thanks for your help btw.

Myten · 03-13-2003, 11:22 AM

deionised water isnt good enough. you _need_ to get some additives in that mixture if youre planning to keep your system. cause eventuallt organisms _will_ start to grow. waterwetter is a popular product to prevent this from happening. WW also reduces waters viscosity which should increase coolingperformance.

bigben2k · 03-13-2003, 11:40 AM

Quote:

Originally posted by Myten
WW also reduces waters viscosity which should increase coolingperformance.

Not true, but otherwise correct.

We use Water Wetter for its anti-corrosion properties, period.

leejsmith · 03-13-2003, 12:26 PM

Quote:

Originally posted by leejsmith
i only have some cooling system conditioner and anti corrosive stuff from a local car super store. if and when i install it into a case i will get wetter water.

what do you think about using acrylic for the top is that a bad idea ?

i am not just using the water i do have some additives in there to.

theetruscan · 03-13-2003, 12:30 PM

Quote:

Originally posted by bigben2k
Not true, but otherwise correct.

We use Water Wetter for its anti-corrosion properties, period.

You're wrong. Sorry.

I have heard conflicting information on this, so I will not claim to know why, but water wetter improves cooling when added to water (the common claim I have seen is reduced surface tension, but I don't know).

http://www.redlineoil.com/redlineoil/wwti.htm - look about 1/2 way down.

bigben2k · 03-13-2003, 12:45 PM

Quote:

Originally posted by theetruscan
You're wrong. Sorry.

I have heard conflicting information on this, so I will not claim to know why, but water wetter improves cooling when added to water (the common claim I have seen is reduced surface tension, but I don't know).

http://www.redlineoil.com/redlineoil/wwti.htm - look about 1/2 way down.

...and that would be true, if you were using it in an engine's cooling system, where the part you're trying to cool is so hot, that it boils the coolant. (which WW is very good at reducing, hence the improved cooling).

But we're talking about a PC here, not an engine block.

Water temps hardly go over 30 C, and we're able to maintain the CPU temp about 10 deg C above that (worst case), so 40 deg C is the highest temp the coolant will ever face.

So if the additive contains, for example, mineral oil, then the water's viscosity will be ever so slightly higher (averaged out), which will increase the thickness of the boundary region, and reduce the cooling effectiveness.

But it won't corrode.

If you came across the WC101 article, see my comments about it.

theetruscan · 03-13-2003, 01:04 PM

OK, so I searched long an hard for this one (5 seconds of googling is a lot for a lazy bastard). And here, from http://www.procooling.com/articles/h...er_coolin2.php That's right, procooling, we have the following statements:

"I could bring up a table of the heat capacity and conductivity of the other additives, but this has been well discussed in other articles and forum threads. Simply put, pure water is the best at moving heat if you are measuring a static system. If you are using a chiller to go below ambient, consider methanol as an additive to reduce the freezing temperature not Antifreeze. Antifreeze is very thick at low temperatures. At normal temperatures, the only performance known enhancement is a surfactant such as Water Wetter™ or soap. However, remember that pure methanol is poisonous and flammable. Silicone tubing may allow methanol fumes to migrate out and reach unpleasant concentrations in a poorly ventilated room."

Water Wetter "Reduces surface tension and increases heat transfer efficiency because of reduced skin effect."

Oh, and your comment that water wetter works by having coolant not boil makes no sense. Coolant boiling is catastrophic. The Glycol/Water mixture shouldn't be boiling, the deal is that it transfers more heat.

MadDogMe · 03-14-2003, 04:29 AM

It's supposed to work like you say T~E, it reduces surface tension, but whether it does provide any noticable cooling enhancement I don't know for sure. I know people (BillA for one) have had problems with WW varients 'slimeing up the tubing, but again I don't know if this is down to incorect mixture. If anything it's the anti~growth anti~corrosion aspects/claims that are 'ambiguos' IMO ...

*******************************************

If those bubbles don't move I doubt you're getting any water flowing in those places

, lowering the hight would be a good idea to boost velocity through the block, and I can't see those fins carrying heat up very high, not that high anyway!, 3~4mm would be good for that thickness I think...

For the top I personaly would use brass and solder copper/brass tubing onto it in place of barbs, it does'nt have to be so 'thick' if it's not threaded. Lexan is VERY hard to get hold of in the UK, it's VERY expensive to buy it by the sheet as well (£150+!!, for 4'x8'x5mm, it'd need to be thicker for threading as well!), some bus shelters are made from Lexan, the vandal~proof 'glass' ones

, maybe take your rechargable jigsaw with you next time you take a bus anywhere!

. I would'nt use plexiglass as it's even more brittle/prone to crack, if you do be careful

(edit!, noticed you DO

)...

Have you found a pic of the T/T 1U P4 HS yet?, seen how dense the fins are?, they're about 50~50 fin to water ratio I believe!. I'd like to try fin heights between 3~5mm to see what's best on a T~Bred sized core...

PS, the only thing you realy need is a 'jet'/orrofice to spread the inlet~water across the width of the CPU die, it's probably not spreading out to the edges at the moment...

03-11-2003, 12:26 PM	#1
leejsmith Cooling Savant Join Date: Mar 2003 Location: notts uk Posts: 408	microfin wb made from akasa ak360/370 i would appricate any comments / tips you guys might have on this wb i made from the base of the akasa ak360 copper cooler. would removing even more of the fins on the outer edges at and at the same time making the inside of the top smaller improve performance ? i also think if the initial copper cooler didnt have the gap in the center that would improve the wb but i dont have the ability / tools to make my own copper base plate with the micro fins. i have tested it and the setup worked very well. I have a 900l/h pump. heater core almost the same as the chevvy running on a p42.4 room temp 17.9C cpu 19C water 18.3C m/b 19C after 1 hour room temp 18.3C cpu 21C water 20.2C m/b 26C i ran sandra cpu test 60 times room temp 18.3C cpu 26C water 22C m/b 27C after 5 mins back at idle room temp 18.4C cpu 21C water 20.9C m/b 26C hope you guys can help Lee

03-11-2003, 03:02 PM	#4
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	That depends: with the gap, how tall are the fins? If they're more than 5mm, you'd have a hard time improving anything. __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

03-13-2003, 08:47 AM	#13
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	Probably not: 4mm is still very effective. The limitation here is in the fin spacing: it's optimized for air, not water. __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

03-13-2003, 09:04 AM	#15
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	Cathar figured that the best fin to channel width ratio is ideally between 0.75 and 1.5. If you measure your fin gap, I think you'll find it to be a number of times bigger than the fin. For the bubbles, an additive would probably fix that: what are you using now? If you just want to test it, a couple of drops of Jet-Dry should take care of it, but that's just for a test: Jet-Dry might not be the best thing to use, permanently. Also, like I said, you can play with the inlet geometry and orientation, but you should be already pretty close to the maximum you're going to get out of that heatsink. __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

03-13-2003, 11:22 AM	#19
Myten Cooling Neophyte Join Date: Sep 2002 Location: sweden Posts: 19	deionised water isnt good enough. you _need_ to get some additives in that mixture if youre planning to keep your system. cause eventuallt organisms _will_ start to grow. waterwetter is a popular product to prevent this from happening. WW also reduces waters viscosity which should increase coolingperformance. __________________ "to fully understand something and to misunderstand the same thing, dont always exclude one another - kafka"

03-11-2003, 03:43 PM	#5
leejsmith Cooling Savant Join Date: Mar 2003 Location: notts uk Posts: 408	the taller fins are 8mm and the lower are 4mm the gap is 1.5mm overall fin area as 45mm x 30mm the base is 3mm i use 3/8 barb

03-13-2003, 05:54 AM	#6
leejsmith Cooling Savant Join Date: Mar 2003 Location: notts uk Posts: 408	so are you saying there is nothing i can do to make this any better ?

03-13-2003, 06:58 AM	#7
wymjym Cooling Neophyte Join Date: Sep 2002 Location: near Austin Posts: 96	You can always improve but, if you are getting the cpu within 4 degrees of the water in your worst environment (benching-gaming) you have the best cooling water block I have heard off. I'm not familiar with the PIV temp swings but most AMD cpus (in well configured water systems) see the cpu temp around 8~10 degrees above the water temp. wj

03-13-2003, 07:04 AM	#8
leejsmith Cooling Savant Join Date: Mar 2003 Location: notts uk Posts: 408	i am doing some testing on the temp probes i used to make sure the readings are correct. see below for my testing results. info at overclockers.com on the wb

03-13-2003, 08:25 AM	#9
MadDogMe Thermophile Join Date: Aug 2002 Location: Just shut up ;) ... Posts: 1,068	Thermaltake make a 1U 'skived' copper HS without the central gap, get them at theoverclockingstore.co.uk ... For more detailed pics without the fan search for some of Volenti's posts in this forum~section... ***************************************** I'd make the fin area a little bit bigger than the cores size but no bigger, I think reducing the height of the fins would help as well, 3 to 5mm maybe?. Also making a 'jet' so it concentrates the water and makes it the right width like Cathars 'whitewater wonder!' would help ... The Thermaltake 1U is a firm favourite of mine at the moment, it's a H2o block maker/modders dream! , I'd like to cross cut one into 'pins' and incorporate it in a radial design. soon, soon, but I'm busy at the mo' :shrug: ...

03-13-2003, 08:38 AM	#10
leejsmith Cooling Savant Join Date: Mar 2003 Location: notts uk Posts: 408	yes that looks perfect. just out of interest how would you make a top ? i dont have any way to make a copper top so i used acrylic but the more i read on this subject the more reasons i find not to use acrylic . I have a p4 so if i cut the fins down to be a little bigger than the p4 heat spreader and about 4mm in height that should be ok.

03-13-2003, 08:50 AM	#14
leejsmith Cooling Savant Join Date: Mar 2003 Location: notts uk Posts: 408	what would be the best gap between each fin for water. i found on the first attempt on the wb you could see loads of bubbles between the fins and no matter how long i left it or shook the wb they never moved. Thats why i lowerd every other fin.

03-13-2003, 09:18 AM	#18
leejsmith Cooling Savant Join Date: Mar 2003 Location: notts uk Posts: 408	i am using de-ionised water. thanks for your help btw.

03-13-2003, 01:04 PM	#24
theetruscan Cooling Neophyte Join Date: Jan 2003 Location: OR/CA/NY Posts: 81	OK, so I searched long an hard for this one (5 seconds of googling is a lot for a lazy bastard). And here, from http://www.procooling.com/articles/h...er_coolin2.php That's right, procooling, we have the following statements: "I could bring up a table of the heat capacity and conductivity of the other additives, but this has been well discussed in other articles and forum threads. Simply put, pure water is the best at moving heat if you are measuring a static system. If you are using a chiller to go below ambient, consider methanol as an additive to reduce the freezing temperature not Antifreeze. Antifreeze is very thick at low temperatures. At normal temperatures, the only performance known enhancement is a surfactant such as Water Wetter™ or soap. However, remember that pure methanol is poisonous and flammable. Silicone tubing may allow methanol fumes to migrate out and reach unpleasant concentrations in a poorly ventilated room." Water Wetter "Reduces surface tension and increases heat transfer efficiency because of reduced skin effect." Oh, and your comment that water wetter works by having coolant not boil makes no sense. Coolant boiling is catastrophic. The Glycol/Water mixture shouldn't be boiling, the deal is that it transfers more heat.

03-14-2003, 04:29 AM	#25
MadDogMe Thermophile Join Date: Aug 2002 Location: Just shut up ;) ... Posts: 1,068	It's supposed to work like you say T~E, it reduces surface tension, but whether it does provide any noticable cooling enhancement I don't know for sure. I know people (BillA for one) have had problems with WW varients 'slimeing up the tubing, but again I don't know if this is down to incorect mixture. If anything it's the anti~growth anti~corrosion aspects/claims that are 'ambiguos' IMO ... ******************************************* If those bubbles don't move I doubt you're getting any water flowing in those places , lowering the hight would be a good idea to boost velocity through the block, and I can't see those fins carrying heat up very high, not that high anyway!, 3~4mm would be good for that thickness I think... For the top I personaly would use brass and solder copper/brass tubing onto it in place of barbs, it does'nt have to be so 'thick' if it's not threaded. Lexan is VERY hard to get hold of in the UK, it's VERY expensive to buy it by the sheet as well (£150+!!, for 4'x8'x5mm, it'd need to be thicker for threading as well!), some bus shelters are made from Lexan, the vandal~proof 'glass' ones , maybe take your rechargable jigsaw with you next time you take a bus anywhere! . I would'nt use plexiglass as it's even more brittle/prone to crack, if you do be careful (edit!, noticed you DO )... Have you found a pic of the T/T 1U P4 HS yet?, seen how dense the fins are?, they're about 50~50 fin to water ratio I believe!. I'd like to try fin heights between 3~5mm to see what's best on a T~Bred sized core... PS, the only thing you realy need is a 'jet'/orrofice to spread the inlet~water across the width of the CPU die, it's probably not spreading out to the edges at the moment...

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)