Any news on the cascade xxx? - Page 8

8-Ball · 05-11-2004, 08:50 AM

Quote:

Originally Posted by jlrii

Ahhtook me 15 min to find a relevent article...HERE ...$$$??? says up to a 40% increase in thermal conductivity HERE

Note that the 40% improvement was the BIGGEST improvement seen, and this was observed when using ethylene glycol as a base, not water.

As I understand it, this technology is more suited to providing lesser coolants with thermal characteristics closer to that of water without many of water's inherant problems, corrosion for example.

8-ball

BillA · 05-11-2004, 08:56 AM

spoon fed, eh ?

close jlrii
the 'trick' is th make the cu into nanospheres, the effective sg of which can be mfgd as desired
- problem is that I could not find a commercial source of cu nanospheres

work wonderfully BTW, their impingment destroys the boundary layer

Groth · 05-11-2004, 09:43 AM

effective sg? Yeesh! Must be a engineering term, I don't know any self-respecting physicist that would be caught dead with such a kludge. And why would you think I had larger particles in mind?

BTW, if they work through impingement destroying the boundary layer, why would all the researchers measure the thermal conductivity with a hot-wire test? No boundary layer in a still fluid...

Lead astray.

BillA · 05-11-2004, 11:24 AM

not being a self-respecting physist, I take many such liberties
and was at a loss as to how to preclude the nanospheres assessment in air, rather than water, yea sg = 1

will look for the papers (3-4 mos back, difficult)

jlrii · 05-11-2004, 11:52 AM

Quote:

Originally Posted by unregistered

spoon fed, eh ?

close jlrii
the 'trick' is th make the cu into nanospheres, the effective sg of which can be mfgd as desired
- problem is that I could not find a commercial source of cu nanospheres

work wonderfully BTW, their impingment destroys the boundary layer

Yeah the metorite comment above reminded me of the nano fluid stuff

jlrii · 05-11-2004, 05:07 PM

"As I understand it, this technology is more suited to providing lesser coolants with thermal characteristics closer to that of water without many of water's inherant problems, corrosion for example."

I believe this refers to the thermal conductivity of straight water with nanoparticles. It references paticals of both Al and Cu.

Quote:
"For example, the use of A12O3particles ≈13nm in diameter at 4.3% volume fraction increased the thermal conductivity of water under stationary conditions by 30% [6].Use of somewhat larger particles (≈40 nm in diameter) only led to an increase of less than ≈10% at the same particle volume fraction [5]; more in accord with theoretical predictions[7]. An even greater enhancement was recently reported for Cu nanofluids, where just a 0.3% volume fraction of 10 nm Cu nanopar-ticles led to an increase of up to 40% in thermal conductivity [8], a result that is more than an order of magnitude above the increase predicted by macroscopic theory. Currently, the origin of such remarkable increases in the thermal conductivity of nanofluids eludes theoretical understanding."

It doesnt mention water in motion

AngryAlpaca · 05-11-2004, 09:57 PM

Yes spoon fed is awesome. I didn't realize what the copper actually did for cooling.

Cathar · 05-11-2004, 10:15 PM

Well the plans for the refined prototype are complete. Unlikely that I'll make too many changes after this next prototype.

My goal with the refined prototype is to exceed the SS across the entire test flow-rate range (2-10LPM), and exceed the XS at anything above 4LPM. The XS has quite phenomenal low-flow performance so I won't set my aim quite that high, but if I do beat it, I'll be a happy person.

AngryAlpaca · 05-11-2004, 11:12 PM

What kind of pressure drop does the XS have? All this talk of flows is kind of useless without it. For all we know, the X-flow could have the best performance at 10LPM (we just can't get to that flow... 37.5 feet... Hehe)

Cathar · 05-11-2004, 11:50 PM

The XS has a moderate-high pressure drop, at about 6mH2O at 10LPM.

An Iwaki MD30-RZ can drive the XS at 10LPM. A Swiftech MCP600 at just under 6LPM. A Laing D4 at ~7LPM at 12V, or ~9.5LPM at 16V. An Eheim 1250 at around 5.5LPM. An Eheim 1048 at 4LPM. That's not counting other system resistances, which given the significance of the XS in the loop, have a fairly minimal impact on those flow-rates listed. As you can see, there exists plenty of options to get the block up to the upper end of the flow rate ranges that I test at, and indeed, that is how I manage to test at those flow rates.

I already know what the PD of the newer P2 refinement will be ahead of time, but I may change it depending on the outcome of some experiments with it.

Hmm, I wonder if Dyson will be mad about the block?

jlrii · 05-12-2004, 07:22 AM

You must be somewhat anxious.....I know I am. Like a kid a christmas exept your not sure if Santa's really gonna show up. If things work out for you within the next 3 or 4 months partialy because the timing would work out perfectly for me

. WCing is ripped out of my current rig right now whilst I upgrade.I have all in hand exceptin a block- new rack mount case, 2xD4s...brass ones even, dual Thermochill 120.1s, I have a 5/8 od mandrel bender in hand and I'm going to do much of the main tubing in 1/2 id Cu (except where i can't squeese it or it needs to move) The "P2" would complete it nicely, as I was having no luck at all scanning forums and eBay for a used Cascade and wouldn't touch an RBX or similar if it were free. The whine of my 60mm 6k rpm SK700 air cooler is a constant reminder to get to work on it. Nevermind that I had to derate my oc to run it on air. So wishing you the best here...albiet somewhat selfishly. And also looking forward to how some of the manufactures look in the aftermath of the "storm".

AngryAlpaca · 05-12-2004, 08:50 AM

Why doesn't anyone ever try overvolting the MCP 600? jlrii: The approximate difference between the WW and the Cascade is 1C... Why not just grab the WW? Is that 1C really important enough to make you keep that whiny (and much warmer) HSF?

Jabo · 05-12-2004, 11:18 AM

Cathar, I am seriuosy glad you figured it out finally!
Benefits are quite phenomenal there. Massive increase in coolants velocity with increased pressure combined with substantially higher water to copper surface area. There's one more problem here. It's 3D design of a block using pressure force and velocity increase effects. All I can say is that there's a compromise to be struck here between an amount of coolan't molecules being in contact with copper in any given time, restrictiveness and spot cooling technology. I got inspired by foamed graphite cooling tech

Keep digging m8! You are so close (or maybe you are there already???)

jlrii · 05-12-2004, 01:20 PM

Quote:

Originally Posted by AngryAlpaca

Why doesn't anyone ever try overvolting the MCP 600? jlrii: The approximate difference between the WW and the Cascade is 1C... Why not just grab the WW? Is that 1C really important enough to make you keep that whiny (and much warmer) HSF?

Hmm.....Ummm...yup

...was time for a rebuild, was running a reeal old block, an Innovatek R3, worked well but there's a lot more mhz in this 2500+. I want to do it up right so I can be happy with it for quite some time and have the option to move on as possibly to a 64 platfom at some time. Just using the ex-wc rig for gaming right now, so i found an acoustic solution. Sound is an Audigy ZS Platinum through an old Sound Craftman preamp and PM1600 amp pushing about 600 wrms into a pair of Polk SDA SRS 2 Speakers. I just crank it up about 1/2 way while playing UT and the fan noise seems to go away. NG for listening to music at low volumes as it sits now tho'... drives me nuts. The WW blocks were supposed to be back in stock the end of this month and I had considered getting one but I will hold off till I see what happens with the P2. Once the new box is done I can use the old block in the interim and will pick up a WW if it doesn't work out. I just missed out on the Cascade, I'll be damned if I miss out on this one.

pauldenton · 05-12-2004, 02:46 PM

Quote:

Originally Posted by AngryAlpaca

Why doesn't anyone ever try overvolting the MCP 600?

it doesn't offer as tempting results as the D4... see cathar's test
http://forums.overclockers.com.au/sh...5&pagenumber=3

AngryAlpaca · 05-12-2004, 07:07 PM

Yeah, 60% more flow in an actual system isn't that great.

How much improvement does the D4 allow?

pauldenton · 05-12-2004, 08:34 PM

Quote:

Originally Posted by AngryAlpaca

Yeah, 60% more flow in an actual system isn't that great.

How much improvement does the D4 allow?

err no - at 16V its 60% more than 10V.....
mcp600
12.0v 0.98A 11.8W 5.8LPM
16.0v 1.41A 22.6W 8.0LPM
which is a 38% improvement....

the D4 gave:
12V approx 19W 7.5LPM
16V approx 24W 10.5LPM
http://forums.overclockers.com.au/sh...5&pagenumber=7

why would you go to the effort of overvolting an mcp600 (seperate psu, separate control circuit, plus an unknown effect on the life etc...)
when a stock D4 delievers virtually the same performance without the hassle. the attraction of overvolting the D4 is iwaki performance cheaper and more compact.....

Cathar · 05-12-2004, 09:05 PM

An MCP600 oc'ed to 15-16V gives US-spec Iwaki MD15-R sort of performance. A 15VDC power-pack is a pretty cheap thing, and the MCP600 + 15VDC powerpack would be considerably less than the Iwaki MD15-R.

Not that overclocking of DC pumps is a terribly good idea if you care about warranty or pump life-span (I don't and I don't try to RMA a product that I broke - I accept the consequences of my actions), I still wouldn't recommend running the MCP600 at more than 15V as it may not reliably start up above that voltage.

I use 2 x D4's in series for my testing, attached to a lab bench PSU where I can run the voltage from 5V (Eheim 1046 sort of performance) right up to 16V (US-Spec Iwaki MD-40RZ sort of performance).

I tell ya, the 2 x D4's in series at 16V provide plenty of push, hitting around 19PSI or so if you want to try to stop them from flowing. Excellent cheap pumping power.

AngryAlpaca · 05-12-2004, 09:46 PM

Oh oops.

Quote:

MCP600 + 15VDC powerpack would be considerably less than the MCP600.

?!?!? 19PSI? I find that hard to believe, well, a little...

Cathar · 05-12-2004, 09:55 PM

Quote:

Originally Posted by AngryAlpaca

Oh oops. ?!?!? 19PSI? I find that hard to believe, well, a little...

Fixed the "Oops".

A Laing D4 at 16V dead-heads at around 6.8mH2O.

2 in series = 13.6mH2O

13.6 x 1.42 ~= 19.3PSI

Here's my plots of my measurement for the Laing D4's PQ curve at 12v and 16v: here.

05-11-2004, 11:50 PM	#150
Cathar Thermophile Join Date: Sep 2002 Location: Melbourne, Australia Posts: 2,538	The XS has a moderate-high pressure drop, at about 6mH2O at 10LPM. An Iwaki MD30-RZ can drive the XS at 10LPM. A Swiftech MCP600 at just under 6LPM. A Laing D4 at ~7LPM at 12V, or ~9.5LPM at 16V. An Eheim 1250 at around 5.5LPM. An Eheim 1048 at 4LPM. That's not counting other system resistances, which given the significance of the XS in the loop, have a fairly minimal impact on those flow-rates listed. As you can see, there exists plenty of options to get the block up to the upper end of the flow rate ranges that I test at, and indeed, that is how I manage to test at those flow rates. I already know what the PD of the newer P2 refinement will be ahead of time, but I may change it depending on the outcome of some experiments with it. Hmm, I wonder if Dyson will be mad about the block? Last edited by Cathar; 05-12-2004 at 02:28 AM.

05-12-2004, 07:22 AM	#151
jlrii Cooling Savant Join Date: Feb 2004 Location: Massachusetts, USA Posts: 158	You must be somewhat anxious.....I know I am. Like a kid a christmas exept your not sure if Santa's really gonna show up. If things work out for you within the next 3 or 4 months partialy because the timing would work out perfectly for me . WCing is ripped out of my current rig right now whilst I upgrade.I have all in hand exceptin a block- new rack mount case, 2xD4s...brass ones even, dual Thermochill 120.1s, I have a 5/8 od mandrel bender in hand and I'm going to do much of the main tubing in 1/2 id Cu (except where i can't squeese it or it needs to move) The "P2" would complete it nicely, as I was having no luck at all scanning forums and eBay for a used Cascade and wouldn't touch an RBX or similar if it were free. The whine of my 60mm 6k rpm SK700 air cooler is a constant reminder to get to work on it. Nevermind that I had to derate my oc to run it on air. So wishing you the best here...albiet somewhat selfishly. And also looking forward to how some of the manufactures look in the aftermath of the "storm". Last edited by jlrii; 05-12-2004 at 07:49 AM.

05-12-2004, 11:18 AM	#153
Jabo Cooling Savant Join Date: Feb 2004 Location: Hertfordshire, England Posts: 164	Cathar, I am seriuosy glad you figured it out finally! Benefits are quite phenomenal there. Massive increase in coolants velocity with increased pressure combined with substantially higher water to copper surface area. There's one more problem here. It's 3D design of a block using pressure force and velocity increase effects. All I can say is that there's a compromise to be struck here between an amount of coolan't molecules being in contact with copper in any given time, restrictiveness and spot cooling technology. I got inspired by foamed graphite cooling tech Keep digging m8! You are so close (or maybe you are there already???) __________________ Museum piece but .... Medusa cooled Last edited by Jabo; 02-19-2005 at 02:08 PM.

05-12-2004, 09:05 PM	#158
Cathar Thermophile Join Date: Sep 2002 Location: Melbourne, Australia Posts: 2,538	An MCP600 oc'ed to 15-16V gives US-spec Iwaki MD15-R sort of performance. A 15VDC power-pack is a pretty cheap thing, and the MCP600 + 15VDC powerpack would be considerably less than the Iwaki MD15-R. Not that overclocking of DC pumps is a terribly good idea if you care about warranty or pump life-span (I don't and I don't try to RMA a product that I broke - I accept the consequences of my actions), I still wouldn't recommend running the MCP600 at more than 15V as it may not reliably start up above that voltage. I use 2 x D4's in series for my testing, attached to a lab bench PSU where I can run the voltage from 5V (Eheim 1046 sort of performance) right up to 16V (US-Spec Iwaki MD-40RZ sort of performance). I tell ya, the 2 x D4's in series at 16V provide plenty of push, hitting around 19PSI or so if you want to try to stop them from flowing. Excellent cheap pumping power. Last edited by Cathar; 05-12-2004 at 09:50 PM.

05-11-2004, 08:56 AM	#142
BillA CoolingWorks Tech Guy Formerly "Unregistered" Join Date: Dec 2000 Location: Posts: 2,371.493,106 Posts: 4,440	spoon fed, eh ? close jlrii the 'trick' is th make the cu into nanospheres, the effective sg of which can be mfgd as desired - problem is that I could not find a commercial source of cu nanospheres work wonderfully BTW, their impingment destroys the boundary layer

05-11-2004, 09:43 AM	#143
Groth Cooling Savant Join Date: Mar 2003 Location: MO Posts: 781	effective sg? Yeesh! Must be a engineering term, I don't know any self-respecting physicist that would be caught dead with such a kludge. And why would you think I had larger particles in mind? BTW, if they work through impingement destroying the boundary layer, why would all the researchers measure the thermal conductivity with a hot-wire test? No boundary layer in a still fluid... Lead astray.

05-11-2004, 11:24 AM	#144
BillA CoolingWorks Tech Guy Formerly "Unregistered" Join Date: Dec 2000 Location: Posts: 2,371.493,106 Posts: 4,440	not being a self-respecting physist, I take many such liberties and was at a loss as to how to preclude the nanospheres assessment in air, rather than water, yea sg = 1 will look for the papers (3-4 mos back, difficult)

05-11-2004, 05:07 PM	#146
jlrii Cooling Savant Join Date: Feb 2004 Location: Massachusetts, USA Posts: 158	"As I understand it, this technology is more suited to providing lesser coolants with thermal characteristics closer to that of water without many of water's inherant problems, corrosion for example." I believe this refers to the thermal conductivity of straight water with nanoparticles. It references paticals of both Al and Cu. Quote: "For example, the use of A12O3particles ≈13nm in diameter at 4.3% volume fraction increased the thermal conductivity of water under stationary conditions by 30% [6].Use of somewhat larger particles (≈40 nm in diameter) only led to an increase of less than ≈10% at the same particle volume fraction [5]; more in accord with theoretical predictions[7]. An even greater enhancement was recently reported for Cu nanofluids, where just a 0.3% volume fraction of 10 nm Cu nanopar-ticles led to an increase of up to 40% in thermal conductivity [8], a result that is more than an order of magnitude above the increase predicted by macroscopic theory. Currently, the origin of such remarkable increases in the thermal conductivity of nanofluids eludes theoretical understanding." It doesnt mention water in motion

05-11-2004, 09:57 PM	#147
AngryAlpaca Cooling Savant Join Date: Jan 2004 Location: Alberta Posts: 631	Yes spoon fed is awesome. I didn't realize what the copper actually did for cooling.

05-11-2004, 10:15 PM	#148
Cathar Thermophile Join Date: Sep 2002 Location: Melbourne, Australia Posts: 2,538	Well the plans for the refined prototype are complete. Unlikely that I'll make too many changes after this next prototype. My goal with the refined prototype is to exceed the SS across the entire test flow-rate range (2-10LPM), and exceed the XS at anything above 4LPM. The XS has quite phenomenal low-flow performance so I won't set my aim quite that high, but if I do beat it, I'll be a happy person.

05-11-2004, 11:12 PM	#149
AngryAlpaca Cooling Savant Join Date: Jan 2004 Location: Alberta Posts: 631	What kind of pressure drop does the XS have? All this talk of flows is kind of useless without it. For all we know, the X-flow could have the best performance at 10LPM (we just can't get to that flow... 37.5 feet... Hehe)

05-12-2004, 08:50 AM	#152
AngryAlpaca Cooling Savant Join Date: Jan 2004 Location: Alberta Posts: 631	Why doesn't anyone ever try overvolting the MCP 600? jlrii: The approximate difference between the WW and the Cascade is 1C... Why not just grab the WW? Is that 1C really important enough to make you keep that whiny (and much warmer) HSF?

05-12-2004, 07:07 PM	#156
AngryAlpaca Cooling Savant Join Date: Jan 2004 Location: Alberta Posts: 631	Yeah, 60% more flow in an actual system isn't that great. How much improvement does the D4 allow?

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