Heatercores In Series.....Too Restrictive???

[cartmanea]

I'm going to build watercooling for my Mobile Barton and possibly later add a block on my 6800nu. I am going to use a '77 Bonneville heatercore, Danger Den TDX or RBX block, and a Swiftech mcp-655 pump (full speed). Everything will be 1/2"-5/8".

My question is, if I start out with one heatercore and decide I want more cooling when I add the gpu block, would it be too restrictive to run two '77 Bonneville heatercores in series? I know heatercores are much less restrictive than BIX or BIP, so I don't think it would hurt my flow much, but any comments? I will have 5/8" barbs on the heatercores and everywhere else it is possible.

Thanks!

[Mr. Tinker]

I'd like to know this too.

[fareastgq]

u prob don't need 2 cores for just that, but ya know, if u can fit it in there, do it, heh. I don't know how restrictive it would be :P

[cartmanea]

I know I don't NEED two, but they will fit nicely :)

[masturdebat3rr]

i'm not too sure of the restrictions, but it shouldn't be THAT bad. however, a heatercore handleing a cpu+gpu combo should be no problem. so by adding that second heater core, you shouldn't see that much of a difference in temperature

[MaxxxRacer]

2-302 and 2-199 heatercores are only slightly less restrictive than a BIX2. the differnece isnt as big as everyone thinks.

Personally I think you dont need that much radiator, but if u were dead set on doing it, then i would run them in parallel.

[cartmanea]

That's leaning me towards an RBX then :)

pump-->gpu-->cpu-->parallel heatercores-->reservoir-->pump

I have a big case, its an old Gateway that is begging for a paint job. It has room for a dual heatercore at the front intake and another in the top. I think I'll have the front intake filtered and have the top heatercore an exhaust.

[FrozenPC4Brain]

Thats close to what I'm building! :)
A 302 in the lower front: Intake, Duel 120's shrd pulling <12v
A 302 in the Top: Exhaust, Dual 120's Shroulded pulling <12V
350 pump after lower-cpu-gpu-Duel 120V Ac fountain pumps in res*-upper-switching valves**-lower
*(only one running other wired for running when low flow)
**(external chilled water res available thru switching valves {Fuel couplers,
[bolth sides seal when unhooked]})
Other loop=
Bay mnt res/pump (dash board)-chipset splitter block-RAM DUAL blocks (two per stick)
-psu blocks-80 rad/exhaust fan***-100 rad****-res/pump-
***(above psu,{psu fans turned to blow in from rear out in front shrd'd up to upper
302 and are various volt by temp sen} rad fan at <12v exhaust)
****(100 rad tube n fin type in front of and allmost above in front facia`{tower sealed for inAIR only from front facia cover so all air crosses 100 rad then lower 302}
` 100 rad might be mnted behind lower 302 if chiller get water temp as low as planned.
Thats what 8 months of replanning/revise rethink/refabricate.*^

*^= 4 fun lol

[Mr. Tinker]

I'm going to be running a TEC setup with a venice. I have a heater core that measures 6"X9.5". The TEC is rated for 320W but I'll be running it on my meanwell s-320-12 at about 12-13V. I just wonder if I'll see a big performance increase if I add another 1x120 or 2x120 heater core or rad.

Maxx, you say parallel is the way to go?

[nikhsub1]

You wont find much of a diff going serial or parallel, each has pros and cons. Only way to tell in your setup is to try both ways. I could go into the how's and why's, but dont have time right now. Never the less, there is no one right answer and should be tried each way to find optimum results.

[Ancient_1]

In general series would be the best, the lower restriction of parralell would probably be a better choice if you have a true high flow setup (over 2~2.5 gpm) but with a moderate flow like most of us have (in the area of 1.5 gpm) I would say series would be best because the lower restriction I doubt would come close to making up for the increased efficiency of the higher water velocity in the tubes of the rad.

Also it seems like fan power has an effect, if you are using fairly weak quiet fans it seems like increased water flow has a much smaller effect and that would also seem to favor the parallel setup, so like NS1 said it really depends on each individual setup. I also believe that the differences would be that by testing on a system our temp readings aernt accurate enough to really be sure which is best performance wise.

Here is a graph BillA made from his testing on Thermal Management of the bigmomma (apears to be a chevette sized heatercore so should relate to most 2" thick dual pass heater cores)

http://webpages.charter.net/chew_toy/12big%20momma.gif

And here is a graph of the Thermochill 120.2 and as you can see it also acts quite similar.

http://webpages.charter.net/chew_toy/Thermo12.jpg

As you can see that there are only small gains in efficiency to be had with a flowrate over the 1.5~2 gpm going through the rad but there id quite a bit to gain going from .75~1.5 gpm (which would be the difference most of our systems would see between parallel and series setups)

[cartmanea]

Wow, those took forever to load, thanks for the graphs though. :)

[billb]

google laminar flow. Thats what's happening at very low flow rates (far left side) in the graphs. Dual cores in parallel will make the loss of heat exchange capacity worse. There is a threshold that you can't cross or all goes to heck.

The size of the pipe INSIDE the radiator along with the radiator face size has to be matched with the pump and blocks used. Smaller tubes inside can be used with lower flow pumps and still be efficient...if you have low restriction blocks which help keep the flow non laminar. Restrictive blocks have high reynolds numbers which increaces turbulance and efficiency...but they need high flows, hence pumps that will deliver high flows at high pressures are required....enter the Iwaki MD20RZ(T)
And airflow is even more important that water flow!