Quad Core Phasing and Information : Read it here!

[n00b 0f l337]

Its overcharged only at idle, at load the bottom of the accumulator ices (no insulation on it). I'm working to tune it a bit more to alleviate unloaded problems and I think its pretty much done. Working with a blend now in the unit, not just straight r507a.

[n00b 0f l337]

I'm guessing he's using the wax to make a mold?

[Unknown_road]

special "plaster" is poured around the wax model, then in a furnace the wax is melted out of the plaster model. then you poor in the molten silver. the molten silver will fill up the room where the wax used to be. this way you create a perfect replica of the wax model. http://en.wikipedia.org/wiki/Lost-wax_casting

[kayl]

I decided to do the vcore while i was doing the vdrop read for my cascade as well
Now i wonder what vcore this single stage can handle

[K404]

BUMP. sorry guys but this needs brought to 1st page.

A fair few guys need to read this entire thread. (Including me)

[[XC] 2long4u]

What is your idle evap temperature (or preferably idle low side pressure)?

Under idle if the evap is at ~-35C (or less) the evaporating pressure temperature will be lower than that, if you have -28C 6" from the compressor you're not overcharged, you're pretty well charged

With r507 if you have -28C under idle 6" before the compressor then you are only overcharged if the idle low side pressure is roughly 1 bar(gauge) or higher. Maybe you know this but i'm saying for other people who don't

Tom

//Hot metal melts the wax and the molten metal takes up the space of the displaced wax in the foundry/green sand. For large structures your can use polystyrene instead of wax, i've done this with alu. and it works fairly well also.

Ha Saturn cylinder heads look like styrofoam.

[Freddie123]

Sorry, havn't read all the posts (so maybe repeating someone elses views) but this:

I'm making the move to .036 for these high heat loads and using around 11 feet

Jumped out at me.

I think that maybe having it that long could cause restrictions and actually affect mass flow. Take a look at the captube sizing thread, Jinu and someone else (I forget who) were saying that around 3m of 0.031in actually gave worse temps because of the restriction, granted you're talking about 0.036 but 3.3m seems a lot. For a 250w load I think 210cm would be about right using 0.031in captube, you don't think 7ft is adequate?

[HoriWanderFuLL]

I have a Vapo Xe2 left sold my Ls and Xe...Bought a new Q6600 G0
these are my prefered quad and phase setup dosnt seem to bother the system with this Vcore...any higher and it was a differant story after just couple of minutes...the Xe2 is only capable of removing upto 180w@-18c
I have done lots of benchmarks at 4300Mhz SuperPi and such..But this Prime 95 really test your Phase units out with a good test of capable Vcore settings....Cheers Hori

[n00b 0f l337]

I tested .036" vs .031", the increase in wrapping the additional cap tube, well, not really helpful. No actual increase in performance. Also that calculation was off. I now use 8' 2-4" for a nf11fx, .036" gets the exact performance with 16' 7" of .036".

[Ic3man]

Total phase n00b here so please go easy. I've been watercooling now for a few years and the phase bug has gotten the better of me so I wouldn't mind giving it a try. I have seen a unit thats a bit of a bargain but might need some modding so heres a couple of Q's I hope the experts can help me with :

Would a Prommie Mach 2 regassed with R507 cope with a clocked Q6600 (G0)?

If not what would need to be done to get it to reliably keep the quad cool?

...and how much would it cost inc(UK &#163.

Could the unit be modded to fit a U2-UFO?



Cheers

[Stigma]

I can't quite figure this out. Is TDP the total wasted (heat) power of the CPU or the power used by the CPU in total, including its secondary function i.e. processing (lol)?

http://download.intel.com/design/pro...s/31559202.pdf

Minimum Tc 5C lololol oh rly?

The answer is yes to both.

Absolutely all eletrical apparatus dissapate as much energy as they use. (numbers for illustration purposes only, not necessarily accurate).

60watt lightbulb: 60watt in = 59watts out in heat-energy, and 1watt out in light-energy

100watt processor: 100watt inn = 100watt out in heat energy (because it can not produce kinetic energy, or light energy, it MUST dissipate all used energy to heat). And no, before you ask, there no such thing as "processed data" energy... it takes no energy to do math. The energy requirement is purely a consequence of using non-perfect materials and technology to make the CPU. A perfectly constructed CPU made from superconducting materials could likley run on no energy apart from the initial charge, or (depending on if switching still had to be used, meaning actual moving parts), atleast an incredibly small amount of energy over time.

-Stigma

[tim-]

Ic3man:
with regas and captube adjustment the mach2 will work with a q6600, no problems.

we arn't allowed to talk prices here.. but it isn't any crazy money we are talking about.

[Stigma]

Correct me if I'm wrong, but a reciprocating compressor (that basically just having a decent sized empty tank at the end of the suction line as far as I can understand) will totally remove any chance of floodback no?

I don't quite understand why anyone would NOT buld a reciprocating tank into their systems then... unless its an issue of size, and your trying to make a real tiny phase system or something...

I say it again and again:

The TDP is not the power consumption of the chip. And usually not even by far.

The sooner people like in this thread and whoever makes that "xtreme PSU calculator" absorb this very easy to understand fact, the better for all of us.

Basically AMD's way of measuring TDP (atleast up until now) is to rate it by how hot it could theoretically get in the absolutely worst case scenario.

Intels way is to rate a CPU by the average output it would be likely to have if running at max load in a real-life scenario.

Thus Intels estimate is the more realistic and "useful" number, while AMDs number is more technically accurate and conservative. In Intels case, while the CPU might in (extremely short) short spurts consume more power than the cooling solution can dissipate, it will be under such a short period that the materials involved won't even have time to heat up, and thus its not a problem.

Its a it the same as when your trying to pick a PSU. If you base the power of the PSU by making it powerful enough to cope with the power needed to run a high-end 3D game for example (which is realistically the most power you will be drawing al at once), then it will not be anywhere near the same as the number for theoretical maximum power consumption if all components in the PC simultaneously spiked. And... even if you did encounter such a sitation for a split second, most PSUs can cope with outputting way way more power than they are rated for, provided that is it over a very short period of time. Its the average over time that is the "problem" because of heat primarily.

This is generally why people should base their calculations on averaged maximum numbers even if they aren't technically the maximum.

-Stigma

possible to liquid cool the compressor? there's a new mod for someone aha

Of course it is... I'm working on it

I'm doing a direct and relatively simple approach by having some cooling blocks out of copper milled in a curve so that they fit the surface of the compressor perfectly (then using some large thermal pads to fill the remaining micro-gaps).

it was mostly an idea for making it possible to completely isolate the box and ths reduce noise by not having fans (or room for the noise to escape the padded walls), but I suppose it will also help the longevity of the compressor as well as possibly giving very slightly better temps...

-Stigma

I always thought that the power consumed was only one of many things that are a function of the heat output. Doesn't efficiency play in to it?

No. Power in = power out. Allways allways allways!

Energy usually changes its form though. usually we see it change between electrical, chemical, kinetic (movement inertia), thermal and light.

A CPU can however only consume electrical energy and only dissapate energy in form of heat, so for a CPU (and pretty much all other electronics): electrical energy in = heat out.

energy/mass (which is the same, as Einstein tells us) can not be created or destroyed. it is one of the fundamental laws of the universe, like gravity.

The only thing efficiency does is decide how many mathematical calculations we can do pr joule of energy. Obviously more is better, but power consumed is still the exact same as power dissipated no matter the efficiency level.

EDIT: Sorry for the multi-posting, but I got late to the party, and just read through 14 pages

-Stigma

[Duh]

stygma, it would be nice If instead of using 4 posts you edit and place all in one.. it is not needed to use a post each time you quote

[Stigma]

Sorry. as I explained I was reading through the thread from the start, so it wasnt orignally my intention to have this many responses

I will edit the previous ones together for the readers convenience.

-Stigma

[gelatinousfury]

Glad I found this thread because Jinu is sending out my repaired NF11FX single stage unit on Monday. We were talking on the phone yesterday and we got into the capacity of the unit...I was worried that my B3 Q6600 might be too much for the SS unit but he said that he was testing the unit with a 250-watt heat load and temps were staying around -32C. My unit is tuned a bit more for capacity as opposed to lower temps. I hope this gets me up and running nicely once I receive the unit.

Yesterday on the phone I wish I would've knew about this thread because neither of us knew the heat load of an overclocked B3 quad core.

[AliG]

I think you'll be fine, Jinu knows his stuff and has built phases with the nf11fx for quadcore before as have many others. But if you ever start noticing problems with floodback or rising temps after a while I give him a call