Dual Pass vs. Single Pass Heatercore

[greenman100]

For the following system, what would be the difference in CPU die temp between a dual pass and single pass heatercore, assuming fin density was constant?

D-Tek Customs WW w/Logic Cooling copper top lapped to 1200 grit
Via Aqua 1300
1/2"ID and 5/8"ID throughout
10"x6" of cooling area
2x 85 CFM torin blowers

Thanks....

[rogerdugans]

Build the system and try both, then let us know.

Sorry amigo, but it just ain't that easy! There are a huge number of variables that can affect things in a water cooled system, most of which are ignored even by those who water cool.

The closest I can come to an answer is that I'd be willing to put $100 on the single pass being better than the dual pass rad- as long as I get to be there for the build and testing. ;)

[greenman100]

Originally posted by rogerdugans
Build the system and try both, then let us know.

Sorry amigo, but it just ain't that easy! There are a huge number of variables that can affect things in a water cooled system, most of which are ignored even by those who water cool.

The closest I can come to an answer is that I'd be willing to put $100 on the single pass being better than the dual pass rad- as long as I get to be there for the build and testing. ;)

I was just looking for an idea of the difference....like less than 1C, or 3-4C. I'm thinking less than 1C.

and I might take you up on the $100, if you can find a way to south florida. ;)

[UberBlue]

You'd have better luck asking what the air speed velocity of an unladen swallow is.

[pelikan]

An African or Europeon swallow?

[ls7corvete]

LOL, you guys crack me up

[9mmCensor]

Originally posted by pelikan
An African or Europeon swallow?

Both... The British (Euro) swallow winters in South Africa.

[reaper79]

a grevious edit: Well of course, the African Swallow is nonmigratory...

Moding the rad to be a single pass basically just divides the resistance by...4??? Thus allowing higher flow rate. As seen int he waterblock reviews an increase of a gpm might give you 2-3 degrees at best, and that is if your lucky enough to get another gpm or 60gph out of it. I bet at least 1-2C though.

Probably would help more on more restrictive setups before diminishing returns sets in.

Cheers! :D

[9mmCensor]

Originally posted by reaper79
Now remember, European Swallow are nonmigratory...


But the internet says they are
http://www.garden-birds.co.uk/swallow.htm

[reaper79]

But supposing two swallows carried it together?

[MoreGooder]

They could grip it by the husk.

[greenman100]

Originally posted by reaper79
a grevious edit: Well of course, the African Swallow is nonmigratory...

Moding the rad to be a single pass basically just divides the resistance by...4??? Thus allowing higher flow rate. As seen int he waterblock reviews an increase of a gpm might give you 2-3 degrees at best, and that is if your lucky enough to get another gpm or 60gph out of it. I bet at least 1-2C though.

Probably would help more on more restrictive setups before diminishing returns sets in.

Cheers! :D

thanks for one contributing post out of 8....

wow.....thanks guys. :rolleyes:

[greenman100]

Originally posted by UberBlue
You'd have better luck asking what the air speed velocity of an unladen swallow is.

And why is that?

At least people could list the pros and cons of single vs. dual, and some theories.

this swallow crap is a waste of bandwidth

[[AK]Zip]

Originally posted by greenman100


And why is that?

At least people could list the pros and cons of single vs. dual, and some theories.

this swallow crap is a waste of bandwidth

Look in my post below about the radiator question. I mathimatically worked it out and it came out that the 1 is much better then the 4 and a little better then the. I will be making a identicle radiator 3 times. A single, double, and quad pass and prove once and for all which is better for a slow, med, and fast pump. This will be tested on a P4 and T-Bird chip with different water blocks and pumps.

-Alex-

[woods]

calm down greenman, it is called a joke. well, more of a humorous reference to an old monty python movie. I personally thought it was funny.

And besides the difference between double and single pass hc is going to vary wildly between systems that any test done on one system will not always be relavent to others.things such as air density and altitude along with humidity and temp along with a million different variabls would make it pointless to do a general comparison.

and wether or not a thread is becoming a waste of bandwidth is up to the mods, not you to decide. besides, you just added to the "wasted bandwidth" with your comment.


sorry about that little rant, it has been a tough day.

[UberBlue]

Originally posted by greenman100


And why is that?

At least people could list the pros and cons of single vs. dual, and some theories.

this swallow crap is a waste of bandwidth

I'm sorry about that. I meant to edit in more info but got side tracked.

To get a hard number would require lots of testing with very expensive equipment. Not alot of people have access to those kinds of resources, so it's not likely your ever going to find a hard number. Just lots or educated guesses.

[greenman100]

It just bothered me that I posted a serious question and got tons of joke answers...


my comments about wasting bandwidth was not a waste of bandwidth, because it was encouraging you guys to conribute.

UberBlue, I did not come in expecting a hard number. I did expect some order of magnituide. It has been claculates that order of flow results in less than one degree differences. A hard number was never found, but it was narrowed down to less than a degree. Would you at least say the difference is less than 5 degrees? 4 degrees?

Thanks.

[feyd83]

Originally posted by greenman100
It just bothered me that I posted a serious question and got tons of joke answers...


my comments about wasting bandwidth was not a waste of bandwidth, because it was encouraging you guys to conribute.

UberBlue, I did not come in expecting a hard number. I did expect some order of magnituide. It has been claculates that order of flow results in less than one degree differences. A hard number was never found, but it was narrowed down to less than a degree. Would you at least say the difference is less than 5 degrees? 4 degrees?

Thanks.

Even a guess amount would depend on how restrictive your system is. I bet I would see a noticable difference in my current restrictive-as-crap system. Once I hook my Iwaki up though, I doubt that I would see much of a difference between the two.

EDIT: Duh, you did list your system. What's the specs on a VIA?

[Nandro]

I think he was looking for something more along the lines of a whitewater is known to be better than an innovatek. Or in general, what is better? What has given people the best results from what we do know? I would imagine 2 would be better if the pump was powerfulll enough to make resistance a non issue as is the case with my lytrom modular cooling system. That thing could probably go through 10 rads with power to spare.

[Xymurgy]

I'd say that single-pass HC's would outperform dual-pass HC's with more restrictive blocks. The increased flow would give more than a C with say a WW block. If it was some sort of maze block, I'd say the difference would be negligable.

[AK]Zip, you need to tone down that sig, btw.

And I always enjoy Monty Python references.

[Cyrix_2k]

Ok, as I said im my thread, a Single pass HC has 16 times less resistance than a dual pass HC. I found this info on two different websites ~a week ago. If you google "Dual pass to single pass" you may find the same info I did a few pages in to the search.

Note: I plan on doing this mod in a few weeks, and I will let everyone know the results.

[slater3333uk]

Anyone care to explain why it is 16times better to me? surely it depends on how many veins went each way to start with?

I was thinking closer to 4times less myself.

[infinity9]

my comments about wasting bandwidth was not a waste of bandwidth, because it was encouraging you guys to conribute.

their comments made people laugh, whats wrong with that? esp a bunch of overclocking freaks jacked on caffeine prob sitting at work stressing out over some dumb work crap. laughter is good ;)

but seriously you asked your question in such a way that you got a smartass answer. if you had simply asked what might other people have experienced with the scenario or any info on the differences/advantages/disadvantages you would have gotten better answers.. sometimes the answer is in the question...

[reaper79]

When it comes down to resistance, it is 4 times less, equation is:

1/finalres = 1/r1 * 1/r2

r1 being the first pass, and r2 being the second pass. This is basically the difference between series vs parallel, add the res. of the two radiators together for dual pass (r1+r2), or use the above equation for the same two radiators in parallel, ie single pass.

To get the 16X as mentioned above I believe this is a factor of increased flow rate^squared, that is:

Res X Flow^2, but I don't think this is applicable to our setups.

Cheers! :D

[Cyrix_2k]

I personally think it would be 4x, but i did read 16x... They even went on to say that a 3 pass core would have 64x the resistance of a single pass... That would mean your second equation would work. Well, it's less resistance no matter what, right?

[reaper79]

Right you are good sir! I believe I read the same article as you did and I think they assume that the water velocity doubles, which thanks to other restrictions in our system, it doesn't.

Cheers! :D

[pauldenton]

Originally posted by Xymurgy
I'd say that single-pass HC's would outperform dual-pass HC's with more restrictive blocks. The increased flow would give more than a C with say a WW block. If it was some sort of maze block, I'd say the difference would be negligable.

[AK]Zip, you need to tone down that sig, btw.

And I always enjoy Monty Python references.

actually i suspect that the reverse would be true, for 2 reasons:

1) the reduction in resistance is more marked with the maze.
a maze4 is 3.3" H20 at 1gpm (from joe's test) a whitewater 18" (from BillA's test). a (double pass) heatercore about 4" - say 1" for a single-pass. say 5" for tubing/res/pump etc.
thus the maze loop is 12.3"/9.3" H20 at 1gpm (double pass/single pass)
and the whitewater loop 27"/24"...

so the whitewater gains a 3/27 reduction in resistance (approx 11.1%), and the maze 3/12.3 (approx 24.4%)

2) the maze gains more from extra flow than the whitewater (see phaestus' test at procooling)
http://www.procooling.com/reviews/html/dangerden_maze_4_review.php

in either case the gain in flow would be less than theose percentages...... partly because resistance is proportional to the square of the flow, so the point at which the resistance would be the same would be 5.7% and 17% higher respectively, and partly because no pump of the type we use can deliver higher flow at the same head (positive displacement pumps would, but that's another story...)

any gain in the rads efficiency from the flow is likely to be negligible (assuming the original flow wasn't very low)

no way these kind of increases would give rise to big temp drops from the block... even +17% on a maze4 wouldn't be worth a degree on any realistic heatload...

of course temps aren't the whole story, and it's possible that it might raise the maximum stable OC ....

[reaper79]

Let's put it this way: If your waterblock has a steep slope(c/w vs flow rate), that is it's better with more gpm, then a single pass rad couldn't hurt. :D

[Cyrix_2k]

Originally posted by reaper79
Right you are good sir! I believe I read the same article as you did and I think they assume that the water velocity doubles, which thanks to other restrictions in our system, it doesn't.

Cheers! :D

That's cool. So now the resistance figure for our systems is more liek 4x? Someoen over at pro-cooling said they did the math and it came out to be 3.5x, but when they measured it, it came out to be 4x... Something to think about.

[reaper79]

???? Sounds like the opposite might be true. 4x less resistance from a scientific standpoint, and probably more like 3.5x actual, especially once you take into acount extra fittings and how the Y's are implemented.

[Cyrix_2k]

He said the 1/2" barbs added the extra resistance.

::EDIT::

Ohh, you said LESS resistance. I meant 4x MORE measured (3.5x math) resistance going from single pass to double pass. We're both right :D

[UberBlue]

READ ME SEYMOUR! (http://forums.procooling.com/vbb/showthread.php?t=5936)

It starts of slow, but has a ton of good info twords the end with hard numbers courtesy of BillA.

[felinusz]

Greenman100 - The waterblock you intend to use, the Whitewater, will benefit a good deal from a less restrictive circuit, with a dual-pass radiator being a big cause of restriction, a single-pass radiator being much less of one.

The pump you intend to use is small enough that diminishing returns aren't really an issue when it comes to flow rate - making the rest of your circuit less restrictive will be very much justified by the benefits, even if it takes some work. Making the rest of your circuit less restrictive will allow your waterblock to perform closer to its potential, and will improve your results. As such, making your setup as unrestrictive as possible to maximize your waterblock's potential should be a priority when you design your system.

So, you would definetely benefit from using a single pass radiator over a dual pass one.

Hard numbers for your specific circuit plan are a different issue altogether. An educated guess, and a little bit of logical thinking when you design your circuit should suffice, and the results will speak for themselves :).

[SysCrusher]

Single pass rad would be the way to go if you can do it. They provide less restriction, and a better (higher) deltaT. More surface area. On a dual pass rad, by the time the water passes through the second pass it's already cooled that there is a Lower deltaT, the second pass is not efficient. The second pass is wasted surface area. Single pass rads do not have that problem.