Quad Core Phasing and Information : Read it here!

Printable View

Show 100 post(s) from this thread on one page

02-16-2007, 06:59 AM
esdee

Quote:

Originally Posted by n00b 0f l337

6" from Compressor,
I was thinking the same, Its doing alot better with 10 feet, around -40C with 225 watts so far (vs -22C), and I'm still considering jumping back to .031.

Quote:

Originally Posted by SoddemFX

Hi dude,

Subcooling, not superheat. I think you made a typo :)

Reason i'm asking is that 0.036" so unrestrictive that i'm not sure if you're managing to condense in any significant amount. Maybe you just need a lot more charge. What are your low and high side pressures?

Tom

Quote:

Originally Posted by n00b 0f l337

Bout 160-165 and 0 at load of 225 watts. I'm considering thats the problem as well and that .036" is too unrestrictive for these systems, it was worth a shot though.

ive used .040" cap on cascades in the past, worked just fine, but i used about 25feet of it (approx 7.5m)
10 feet of 0.036 is like 5 feet 0.031.
i would double the cap (20feet) before i give up with the 0.036" ;)

i think 10feet of 0.036 is just too short ...
02-16-2007, 09:09 AM
n00b 0f l337

More subcooling is the major improvement. More surface area to wrap.

Quote:

Thats right, too unrestrictive. But good that you're trying

Charge it harder, i think it'll burst into life about 15PSIg low

Tom

Too much sacrificing in temperature I think at that point.
Working on theoretical numbers I think I simply need to charge more, but that will sacrfice temperature. So I'm going to add more capillary length. Basing myself off your 25 feet of .04, 15.5 feet of .036 would be the equivalent. That should lower the suction but still with more refrigerant give me better load at temperatures I'm aiming for!
02-16-2007, 10:51 AM
esdee

Quote:

Originally Posted by SoddemFX

But what's the point? In a single stage i can't think of any benifit of this over 0.031" which is cheaper and you need half as much of :confused:

With 25' did you experience any "wierd" system opperation?

Tom

absolutely no wierd system opperation, been using this cascade for straight hours and many weeks (smth like 8-10 hours a day for 4 days a week)
i used 0.040 simply beacause i couldnt find 0.036 and i had some issues with 0.031 on my cascade with subcooling (probably oil freezing) and after a while temps started climbing...

you are absolutely right about cost on 0.031 vs 0.040, (and even 0.036) but i think the purpose of this thread is testing and 'trial and error' :)

i have a single stage here, which is build for my post graduate project about overclocking using single stage phase change unit, ill see if i can use a kentsfield on it.
it has simple design, 3m of 0.031 elevens 40mm evap and NEK2134GK,
02-16-2007, 10:54 AM
boshuter

Quote:

Originally Posted by n00b 0f l337

Results coming soon from the build, its looking really good, right now its just idle testing till I get home, the accumulator looks sorta like Ron's at the moment, half ice. Lets see how 12 feet of .036" works :)

Got pics?

I don't think anyone has come up with any official wattage for the quads, but I'm doubting 250 will cover a heavily oc'd one. My 950 presler pulls more than that, at least in all the calculators I've seen to date. I'm betting 300watts is going to be closer to what is needed, if not more.
02-16-2007, 10:56 AM
n00b 0f l337

275 Is what I'd say is probable for heavy load and oc.
I'll post up some pics of the unit later tonight, for the moment I have an odd question that I dont thinks been tried before. Splicing capillary lengths of different sizes?
Lets take my example, I've got 10' of .036 running, I want to add 5.5 feet more. 5.5 feet more of .036 is the same as 2.75 feet of .031, could I theoretically splice in the .031?
02-16-2007, 08:48 PM
esdee

Quote:

Originally Posted by SoddemFX

Cool, thanks for the info. I've had systems become "wierd" with 5' of capillary tube but i've never pushed it to the other extreme of length to find out what happens with very long lengths. I thought you might be using that wide length because of oil in a cascade - i know it's off topic, but did it help?

after the change to 0.040 absolutely no problem, i hope this can be pretty usefull in cascades with more than 2 stages!

Quote:

Originally Posted by SoddemFX

I think your single stage will be very good. I wish my masters projects had been as fun as yours :D

thanks! i grabbed the chance when our teacher added some thermodynamics and refrig parts in his lesson, so i talked to him of what we are doing and that its possible to do this as a post graduate project!
02-16-2007, 08:50 PM
n00b 0f l337

I've got 6 feet more hopefully on the way soon, i'll splice up to around 16 feet and retest. I do feel that more refrigerant helps the system, but the longer captube should lower the suction. Or I'm putting in a CPEV and being done with it :P
02-20-2007, 05:30 AM
esdee

ok guys, i have a qx6700 for some tests for a couple of days, it's already on the asus striker insulated with the above single stage. mounted
first impression is what we already know ,kentsfields are HOT!

i ll try to build a quick heatload based on resistors for compare...

1.5Vcore is hot like hell!
02-20-2007, 07:50 AM
kayl

Quote:

Originally Posted by esdee

ok guys, i have a qx6700 for some tests for a couple of days, it's already on the asus striker insulated with the above single stage. mounted
first impression is what we already know ,kentsfields are HOT!

i ll try to build a quick heatload based on resistors for compare...

1.5Vcore is hot like hell!

nice work esdee, so 1.5v is hot hey, i cant wait to bench my Q6700 :D
02-20-2007, 11:26 AM
esdee

r507 seems to reach it's limits... r410 is also handy for tests, ill try to sum up in a couple of days
02-20-2007, 11:39 AM
n00b 0f l337

R507a has reached its limits? As in what sense? Of retaining a load? I doubt that.
02-22-2007, 08:43 PM
kayl

My Q6700 ES B1 arrived today, installed and booted phase change.
Not the best Q6700 but still a Q6700
3860MHZ with 1.7bios and droops to around 1.61v windows similar to my old E6600. You wouldnt belive it, but my Intel 925 is alot hotter at max oc than this Q6700 at this OC, wish i knew what 4.5ghz felt like though.
currently -30c load with r507, quad prime 95 with 42 case/drier temps at load.
Strange thing though is core temps, its been stuck at 1c at any oc idle or load using both the asus Tool and speedfan ?

http://img139.imageshack.us/img139/4...stembp4.th.jpg

http://img72.imageshack.us/img72/563...9880rg9.th.jpg

Seems that with my resistor load and variac using a 3mm copper plate its no where near 250w. more like 220-230w. I have found the older dual core cpus are a lot hotter at high oc’s as well

http://img139.imageshack.us/img139/7...9871fk3.th.jpg

Heres the HWBOT WPrime32m results, 10.7Sec, ranked 28 = 20.2 point
http://hwbot.org/hallOfFame.do?type=...plicationId=14

wprime1024M, 5mins 50sec, ranked 23rd, = 11.4points.
http://hwbot.org/hallOfFame.do?type=...plicationId=15

Water cooling previous owner results, hot little bugger

CPU is cooled by Cathars storm - Which was stripped down and checked for cleanliness - which it was
The rad is a GT Stealth 3x120
All fans in the enitre setup are Nexus fans @ 12V
The pump is an MCP655 at its fastest setting
The coolant is a 50% radiator fluid, 50% distilled water
The hose is a reinforced clear high pressure garden type hose 7/16 ID

so under phase there was a hudge improvement. I need to do the vdroop mod on this mobo and see how hot it is with 1.7vcore loaded.

http://img139.imageshack.us/img139/2...ef12en0.th.jpg

http://img139.imageshack.us/img139/9...005arf8.th.jpg

http://img139.imageshack.us/img139/4...c956uu3.th.jpg
02-22-2007, 10:50 PM
harvshark

Quote:

Originally Posted by kayl

Strange thing though is core temps, its been stuck at 1c at any oc idle or load using both the asus Tool and speedfan ?

Kayl.....I have posted a couple of times in this forum about the inability to directly read core temps at subzero. I don't think anybody has come up with an answer as to whether it is a bios problem, or if the DTS info coming from the quad is incorrect at subzero. Wish there was a workaround, but I haven't found one yet. Bios hardware monitor probably reads the same 1c reading you are getting from Probe and Speedfan.
02-23-2007, 02:31 AM
esdee

Quote:

Originally Posted by n00b 0f l337

R507a has reached its limits? As in what sense? Of retaining a load? I doubt that.

it's limits for my setup. (compressor/captube)
anyway, with a carefull setup a comrpessor with 1/2hp and r410 would be ideal for quad cores!
02-23-2007, 05:53 AM
hipro5

Quote:

Originally Posted by kayl

My Q6700 ES B1 arrived today, installed and booted phase change.

Try to pick the 2k6 CPU bench and loop it for some time......;)
02-23-2007, 06:08 AM
epion2985

Quote:

Originally Posted by harvshark

Kayl.....I have posted a couple of times in this forum about the inability to directly read core temps at subzero. I don't think anybody has come up with an answer as to whether it is a bios problem, or if the DTS info coming from the quad is incorrect at subzero. Wish there was a workaround, but I haven't found one yet. Bios hardware monitor probably reads the same 1c reading you are getting from Probe and Speedfan.

here's a "work around" of a sort :)

http://www.xtremesystems.org/forums/...d.php?t=133859
02-23-2007, 06:09 AM
esdee

Quote:

Originally Posted by kayl

I need to do the vdroop mod on this mobo and see how hot it is with 1.7vcore loaded.

things are gonna get really hot with 1.7 on C2Q ...
waiting for results kayl :toast:
02-23-2007, 09:01 AM
harvshark

Quote:

Originally Posted by epion2985

here's a "work around" of a sort :)

http://www.xtremesystems.org/forums/...d.php?t=133859

'Of a sort' indeed. Still, it does not address the issue of why things do not work correctly in the first place. Is the DTS info corrupted or can bios not interpret it correctly?
02-23-2007, 01:17 PM
pumbertot

nice work Kayl, SLI_dog has managed 4.2GHz Wprime on a vapo LS so seems they arent as hot as we all thought afterall.
02-23-2007, 01:20 PM
n00b 0f l337

Just spliced in 6 more feet of .036 to harvsharks huge quad core cooler. Results after vacuum.
I think they are colder then we though, but hotter then some still think. Once you start pushing excessive voltage ( and many of us do! ) it does get even hotter.
02-23-2007, 01:25 PM
pumbertot

Quote:

Originally Posted by n00b 0f l337

Just spliced in 6 more feet of .036 to harvsharks huge quad core cooler. Results after vacuum.
I think they are colder then we though, but hotter then some still think. Once you start pushing excessive voltage ( and many of us do! ) it does get even hotter.

yeah so true. I think those with vcore vmods will certainly be able to get close to the 300W mark. unfortunately(or fortunately for my phase unit), I cant solder worth a :banana::banana::banana::banana: so not much chance of me going above 1.60-1.63V on the P5B-D.
will have results tomorrow of a Berkut modded MACHIIGT(R507a+extra condensor) with QX6700. Hoping it can withstand 1.6Vcore.
02-23-2007, 06:09 PM
kayl

Quote:

Originally Posted by harvshark

Kayl.....I have posted a couple of times in this forum about the inability to directly read core temps at subzero. I don't think anybody has come up with an answer as to whether it is a bios problem, or if the DTS info coming from the quad is incorrect at subzero. Wish there was a workaround, but I haven't found one yet. Bios hardware monitor probably reads the same 1c reading you are getting from Probe and Speedfan.

thanxs for that. Yes is bios it still reads 1c. Im trying to get the latest speed fan see if that makes a difference.

Quote:

Originally Posted by hipro5

Try to pick the 2k6 CPU bench and loop it for some time......;)

hipro can you pm me a link to that program and ill download it and have a bench.

Quote:

Originally Posted by esdee

things are gonna get really hot with 1.7 on C2Q ...
waiting for results kayl :toast:

esdee, if its like the other C2 chips in vmoded mobos things get hot quick.
Real hot. Guess theres a reason why ASUS and intel are good friends by have the crap ASUS vdroop. they just dont want ppl running high volts :stick:
02-23-2007, 06:11 PM
s7e9h3n

Quote:

Originally Posted by kayl

hipro can you pm me a link to that program and ill download it and have a bench.

I may be wrong, but I think he's referring to 3dMark 2k6 ;)
02-23-2007, 06:12 PM
n00b 0f l337

Running 15' 8" of .036 right now, refrigerant charge puts high side at 14.2 bars, and low side at .4 bars. -34C evaporator temperature @ 250 watts of heat right now. Bit overcharged I beleive. Very good ressults so far.
02-23-2007, 06:16 PM
s7e9h3n

Quote:

Originally Posted by n00b 0f l337

Running 15' 8" of .036 right now, refrigerant charge puts high side at 14.2 bars, and low side at .4 bars. -34C evaporator temperature @ 250 watts of heat right now. Bit overcharged I beleive. Very good ressults so far.

LOLNOL...I think it's "bar"-singular regardless of the number preceding it :D Nice temp...Now let's see this beast!

Show 100 post(s) from this thread on one page