Serial loop Vs. Parallel loop

[toolmaker_03]

a detailed parts list, of the two water loops used in this test setup.
2 MCR320-QP 360mm radiators
http://www.swiftech.com/MCRx20-QP.aspx
2 EVGA GeForce GTX 580 FTW Hydro Copper 2
https://www.evga.com/articles/00592/
1 XSPC raystorm CPU block
http://www.xs-pc.com/waterblocks-cpu/raystorm-cpu-water...
1 swiftech MPC655 pump
http://www.swiftech.com/mcp655.aspx
6 corsair SP fans
http://www.corsair.com/en-us/air-series-sp120-high-perf...
6 bits power 3 way 90's
http://www.frozencpu.com/products/12084/ex-tub-744/Bits...
1 Y block
http://bitspowerwork.com/html/product/pro_show.php?prod...
1 250mm reservoir
http://bitspowerwork.com/html/product/pro_show.php?prod...
8 3/8ID 1/2OD compression fittings
http://bitspowerwork.com/html/product/pro_show.php?prod...
2 T's
http://www.xs-pc.com/hose-fittings/g14-t-fitting-black-...
5 male to male fittings of varying length
http://www.xs-pc.com/hose-fittings/g14-5mm-male-to-male...
4 female to female fittings of varying length
http://www.xs-pc.com/hose-fittings/g14-18mm-female-to-f...
2 koolance flow meters with display
http://koolance.com/ins-fm19-coolant-flow-meter
http://koolance.com/dcb-fm01-flow-meter-adapter-with-di...
1 loop filter
http://koolance.com/ins-fltr03-inline-coolant-filter
2 temp sensors
http://www.frozencpu.com/products/10373/ex-tub-620/Bits...
1 alphacool bubble down
http://www.frozencpu.com/products/17333/ex-tub-1558/Alphacool_Bubble_Down_G14_Thread_Reservoir_Flow_Regulator_-_Deep_Black.html?tl=g30c97s169
1 bitspower aqua pipe
http://www.frozencpu.com/products/11863/ex-tub-713/Bitspower_G_14_Silver_Shining_Aqua-Pipe_I_BP-WTP-C17.html?tl=g30c97s169

This system is overclocked, on both the serial and parallel test setups.
CPU overclock @ 4Ghz
GPU1 overclock @ 850Mhz
GPU2 overclock @ 850Mhz
Memory overclock @ 1600Mhz
CPU stock @ 3.2Ghz
GPU1 stock @ 772Mhz
GPU2 stock @ 772Mhz
Memory stock @ 1333Mhz
this is the serial loop setup and what it looks like, and here are the hardware temps at idol, ambient temp of the room is at 28.6C
CPU 43C
GPU1 37C
GPU2 37C
the water temps at idol are
29C-30C
the hardware temps at load are
CPU 61C
GPU1 51C
GPU2 51C
the water temps at load are
37C-38C
the delta t of the system at load is 10C
http://imgur.com/Gi4wFZO,G1Mte0g,Pb8ot4j,k8LlLK5

this is the parallel loop setup and what it looks like, and here are the hardware temps at idol, ambient temp of the room is at 28.6C
CPU 43C
GPU1 35C
GPU2 35C
the water temps at idol are
29C-30C
the hardware temps at load are
CPU 57C
GPU1 46C
GPU2 46C
the water temps at load are
34C-35C
the delta t of the system at load is 7C
http://imgur.com/pBZY5ne,yIT8rwA,CgNlfyH

this is the difference between the hardware used on the serial loop, and the hardware that needed to be added to the parallel loop.

http://imgur.com/U3yYYJl

 

[y2kmedman]

I have finally read your sticky on " Tom's HW Watercooling Sticky v2.0".
Even though most of it I have been exposed to in the last 5 months of
full time research by watching You Tube and reading forums it was nice
to see it all in one place. I had even cut down my Skyrim addiction to a
crawl in order to watch all those videos.
Since I am building a PC from the ground up
with a dream budget (a dream to me) for the first time I have developed
a veracious appetite for knowledge on all things PC. I had an already
healthy knowledge base of working on desktops by upgrading the ones
I had purchased already made and enjoyed breaking the seal that
usually voided my warranty. I now have gone thru a few paradigm
shifts and have fell in love with the science and it's practical application
in consumer grade PC's. I rarely truly understand topics on first exposure
but I do keep reading and rereading things until I understand it as apposed
to just memorizing facts. So, in conclusion, I would like to thank you for
writing that sticky so I can refer back to it again and again to avoid
many pitfalls in my path as a newbie. If your 13 year path is any
indication of how long the passion for this topic can last then it is
safe to say I have something to do for a while.
Here is a diagram of how I want to set up my cooling loop completely
in parallel:

[/img]

 

[rubix_1011]

Yeah, it can be done that way. Going to be a pricey loop, but should work well.

If this were my setup, I would run dual loops - CPU and good 240 rad on one, and parallel the GPUs on either a very good 360 or 480. This is actually cheaper (assuming your radiators illustrated above are both 360 or 480.

But, this goes to show you that there are many different ways to skin a cat.

 

[y2kmedman]

After rereading the above posts I concede the fact that my plan is series/parallel. The reason why I was calling it totally parallel is I have mostly, if not only seen parallel used with GPUs only. When I first saw a parallel GPU set up I was confused. A common sign of future learning. After I realized was it was, I wondered since it improves cooling performance on GPUs why it wasn't being used on the CPUs also. I guess or heard the reason was that parallel would only work on two cooling blocks with the same flow rate or restriction. Based on my knowledge of the body, different organs need varying minimum pressure and flow rates and it mostly uses a parallel set up. Under stress or demand, flow to different parts of the body and/or organs is either restricted or improved as needed but I have yet to see a need to do this on a consumer PC. I will continue to post my findings here once I build and test my PC starting next month.

 

[y2kmedman]

Dual loops restrict the benefit of using all radiators to benefit all components equally at the same time. What do you do with a dual loop when you have room for a third radiator? You have to choose which loop to put it on...

 

[toolmaker_03]

yes two loops do lower the total cooling capacity of the rads because they are not being utilized.
in fact, I was told by rubix that my loop ran so cool because I had it was over radded (It was 3 X 360mm rads), so I took one off to prove other wise, and still he does not understand what I have done, but I know if he ever tries to get a loop to run at 9.6LPM he will then understand the hardships I went through to get my system to work at that flow rate. In a couple of years I will have my 5-way setup complete, right now I am at 12.5LPM with 2 CPU's and 3 GPU's and 3 X 360mm rads my low end target is 15LPM and best case will be 20.5LPM I am getting there but there are always restrictions that I would have never guessed, where restrictions until I get to a flow rate that I cant go past, then it is a matter of figuring out what the restriction is and removing it or modifying the configuration of the loop to accommodate for it.
thanks for the interest in the loop, and I would be glad to answer any questions you might have on the subject.

 

[toolmaker_03]

for easy understanding, my loop currently runs at 3 times the flow rate of the standard serial loop setup. with would mean nothing to most, but it is the higher flow rate through the radiators that lowers the delta T of the system.
higher component flow rate, or the flow rate through the blocks on the system does nothing to lower temps.
I have tested this theory, and have seen this to be true.

 

[rubix_1011]

in fact, I was told by rubix that my loop ran so cool because I had it was over radded (It was 3 X 360mm rads), so I took one off to prove other wise, and still he does not understand what I have done, but I know if he ever tries to get a loop to run at 9.6LPM he will then understand the hardships I went through to get my system to work at that flow rate.

I am pretty certain I actually didn't say that it ran 'so cool' due to being over-radded - but I agree that I said the reason it did cool well at lower flow rates was because it was over-radded. Remember, your issues for the longest time actually WERE due to such low flow rates and they were masked by running more radiators (3x360's). I find it entertaining that you think that I don't understand' what you are doing (of everyone on the watercooling threads here). I understand perfectly well.

What I don't understand is why you continue with the harassment when there are quite a few things that I do agree with you on and have said so. Your journey of learning and understanding your progress has shown that you've had different ideas and theories throughout the way, so I question what you are actually saying all of us are incorrect about and what you are correct about.

I checked in on your posts on those other forums - kind of the interactions I expected over there. Seems like there are quite a few similarities with my responses and theirs in regards to your testing.

 

[toolmaker_03]

well if you think that my system does not run as cool as I have submitted, than build one of your own and prove me wrong.

but I can assure you that the results that you achieve, will be similar, if not the exact same results I got.

that is the great thing about the truth, it is repeatable, by anyone.

I removed one radiator from the system, to show that I would get the same temps, even with the minimum amount of radiator space needed for the components on my system.

still you have issue with my setup, with my testing, and even my theory's, I cant help that, it is what it is, and does what I have shown, if that is not good enough for you, than I am sorry, but I am not going to do any more to prove my system works the way that I have shown it does, as it is my opinion that it would do nothing to change any of your opinions.

 

[rubix_1011]

I removed one radiator from the system, to show that I would get the same temps, even with the minimum amount of radiator space needed for the components on my system.

Right - proving that I was correct stating you were over-radding the loop.

I don't have any issues with anything you're doing any longer because I'm tired of arguing with you. I don't 'have issues' with anything you're doing - it's your hobby, go do it. My advice would be to cease the witch hunt around here for everyone who you think has crossed you. Badmouthing everyone is not going to net you support nor is it going to win you any debates.

I still don't understand what you are so mad about - I've already agreed (and shown you with tests I have done) that flow rate and lower restriction in parallel is directly related to the potential of cooling performance. This isn't news. This has been known and discussed for years. No one here is telling you that you're wrong about running anything in parallel. Sure, it can be done. However, parallel drops your flow rates (which I'm sure you know) but not everyone is going to overcome this by running 2 or 3 pumps in order to boost flow rates to overcompensate this. I think where the problem here is that you are assuming everyone should run multiple D5's and multiple cylinder reservoirs in order to have a multitude of tubes running all over their cases. Not everyone wants to do this, but your argument to our responses in your threads and other threads, is that your loop design should be king and everyone should do it, rather than being more sensible and understanding actual loop cooling and limitations and designing what works for them.

The stuff you are doing - it isn't new or ground breaking - people have already done this. It's the reason why people often parallel GPUs. We already know the benefits.

 

[toolmaker_03]

yea your so smart
by the way my test setup has only one pump (but your always so right on everything)
my issue with your understandings is that you believe that it requires a change of at least 1GPM through the loop to alter temps even though I have shown otherwise.
my other issue with your believe, is that you think that a single D5 pump can push 3GPM flow rate through a radiator, but you have yet to prove that a 3GPM flow rate through any radiator is even achievable (I know for a fact, that it is not possible, through any radiator with a single D5 pump)

but you have your opinions and the charts to show that it is.
they (meaning people like you) also had elaborate tests, and charts, to show that world was flat. it does not make them any more correct today, than they where back then.

the same can be said about the charts and tests of today,that show flow rates of 3GPM through the radiators they tested, these are made up charts and tests, that show things that are not possible, just like the world being flat, its not possible.

 

[y2kmedman]

Toolmaker_03, rubix_1011 not only agreed with you, he went a step further and pretty much said that your cooling solution is so effective that you could even remove rads because they are a waste and not even needed because it works so good. Unless I have lost my mind... I can't believe what I'm reading. You are so convinced that rubix doesn't understand you you can't see that he does. Qit playng with him....lol

 

[4Ryan6]

Here is a diagram of how I want to set up my cooling loop completely
in parallel:

[/img]

Your bottom rad is being fed with intake flow on the inlet and outlet ports,
that splitter before the rad is not going to alleviate the coolant flow stalling of that radiator,
the flow through that rad will drop seriously and almost completely stall out.

Your design is creating a coolant flow head on collision at the lower radiator,
which is going to force your water pumps to be stressed much more than they should and shorten their lives.

It is also going to compromise the cooling you could be getting from the lower radiator, by stalling the coolant flow through it.

Many say the flow order doesn't matter but I always run my setups with the coolest coolant hitting the water blocks first and the coolest comes out of the radiator.

To all participants in this thread;

This Serial Loop Vs. Parallel Loop argument can get quite entertaining,
but when it provides information that can confuse or mislead one completely new to water cooling,
it is more detrimental to Toms Hardware IMO.

I am all about the additional 1c ~ 3c a loop arrangement can make when it comes to overclocking,
but when it can be a solution that shortens component life, I am not on board with that, and will be against it 100%.

As a Retired Moderator having zero authority but fully aware of Toms Hardware presentation and responsibility, IMO,
this argument has gone on long enough, and should be in some way either sand boxed or quarantined.

 

[y2kmedman]

Blue lines are leaving the radiators. Not feeding them. I didn't write arrows on every turn on every line because I didn't want the design to look cluttered. As you know the water follows the path of least resistance and will go toward the reservoir on the right.
The color represents the relative temperature of the water. Red for warm, blue for cool.
Funny thing is after messaging back and forth with toolmaker_03 on his set up that he is experiencing a low pressure or vacuum effect before one or both his pumps and he is only using one reservoir. I anticipated a low pressure hence me adding a second pump, but not a vacuum effect. Given that a vacuum has formed, I am betting that there is high pressure on the other side of the blocks on their inlets.

 

[4Ryan6]

Blue lines are leaving the radiators. Not feeding them.

That is true for the top rad in the picture but not the bottom rad, you don't need arrows drawn to understand what's going to happen there, at least I don't,
not all of the flow will take the path of least resistance, some of it will be neutralized and create flow restriction by the out flow of the radiator colliding at the T junction.

You need to rethink that design.

Dealing with flow restriction is water cooling basics, but why would you actually create additional flow restriction?

Please explain how that could possibly be beneficial?

 

[y2kmedman]

As far as causing premature ware and tear I am in full agreement. It's not worth it. As you all can read in my posts, I am talking about theory not application. It will be a month until escrow closes and I can afford to build this monster pc that I have in mind. I fully intend to test and learn as much as I can so I can understand why a parallel set up acts the way it does. If what I plan to do shows any indication of premature ware, I intend to find the cause and avoid it or eliminate it all together. I intend to find the right way to do it by discovering all the wrong ways and document them.

 

[y2kmedman]

Water can do some weird things in a tube. Did you know that a 3cc syringe causes more risk for causing pressure damage in a central line than a larger 5cc or 10cc syringe. I understand how you can picture it happening in your mind but water coming together at a T does not stop or slow down as much to make a visible difference. We can disagree till the cows come home but until we test our assumptions with experimentation it's just that, a disagreement.

 

[y2kmedman]

If you are worried I will use Y connectors and not T connectors. That way the collisions of the water molecules are side swipes instead of head ons...

 

[4Ryan6]

As far as causing premature ware and tear I am in full agreement. It's not worth it. As you all can read in my posts, I am talking about theory not application. It will be a month until escrow closes and I can afford to build this monster pc that I have in mind. I fully intend to test and learn as much as I can so I can understand why a parallel set up acts the way it does. If what I plan to do shows any indication of premature ware, I intend to find the cause and avoid it or eliminate it all together. I intend to find the right way to do it by discovering all the wrong ways and document them.

Theory is all well and good, if not for theory I would not have the Chilled Water Cooling I am running today, but from numerous lessons learned the hard way, theory does not always work, as a matter of fact it can sometimes end in disaster.

If you are worried I will use Y connectors and not T connectors. That way the collisions of the water molecules are side swipes instead of head ons...

I am not worried about your proposed setup, I've given the advice I have to offer from my years of water cooling experience, what you do with that advice is totally up to you.

I am curious though, are you presently water cooling?

Water can do some weird things in a tube. Did you know that a 3cc syringe causes more risk for causing pressure damage in a central line than a larger 5cc or 10cc syringe. I understand how you can picture it happening in your mind but water coming together at a T does not stop or slow down as much to make a visible difference. We can disagree till the cows come home but until we test our assumptions with experimentation it's just that, a disagreement.

In all your posting you've made numerous references to the human body but if there are references to actual water cooling you've directly had experience with I don't see that?

Do you have any pictures of your present cooling?

 

[y2kmedman]

I have absolutely no experience watercooling a pc. But I do have a more than passing knowledge from many sources on the subject and I seek knowledge from more people. Since I have to wait a month to start my first watercooling project. I thought i could use that time to learn more. I have a basic knowledge of physics and I maintained iv lines professionally for 13 years. You have to know pumps and flow rates troubleshoot so my learning curve will be different for your average novice. I also run the risk of knowing enough to do real damage but engaging discussion will help me along 5he path. I also acknowledge that later on down the line I will probably look back at this moment and know how important knowing the real world implications of a T connection means when you are trying to squeeze that last drop of performance out of your system or when the law of diminishing returns should be listened to. But for now I let me personally out and poked a little fun.

 

[4Ryan6]

I have absolutely no experience watercooling a pc. But I do have a more than passing knowledge from many sources on the subject and I seek knowledge from more people. Since I have to wait a month to start my first watercooling project. I thought i could use that time to learn more. I have a basic knowledge of physics and I maintained iv lines professionally for 13 years. You have to know pumps and flow rates troubleshoot so my learning curve will be different for your average novice. I also run the risk of knowing enough to do real damage but engaging discussion will help me along 5he path. I also acknowledge that later on down the line I will probably look back at this moment and know how important knowing the real world implications of a T connection means when you are trying to squeeze that last drop of performance out of your system or when the law of diminishing returns should be listened to. But for now I let me personally out and poked a little fun.

Let me say this to you hopefully with compassion and understanding, on theory alone you came into this thread with no way what so ever to test and backup what you say.

You join Toms Hardware and your first post is to confront one of our long term members rubix_1011 in his dealings with toolmaker_03, totally unaware this is not the beginning,
of confrontations with toolmaker_03 and many of us here at Toms Hardware.

Try to imagine a comrodery between those of us that have been members here a long time as we're kinda like a family,
that we can argue with each other even in serious head butting sessions but still respect each other,
but God help the outsider that calls out one of our own.

Taking that into consideration, I am saying you have been very respectfully treated!

I would like to suggest having a lot more hands on experience before jumping into a thread like this because,
even though toolmaker_03 can come way out of left field sometimes with his ideas, he does actually have the equipment to test.

Whether he realizes it or not we do care about him and my suggestions to him have been to improve his presentation, not to tear him down,
because I believe he has valuable input and so does rubix_1011, he just needs to learn to present it better and who is actually interested in it.

Take care Sam.

PS: When you actually do get into water cooling I'll be glad to help you all I can! Ryan

 

[rubix_1011]

yea your so smart

Now you're just trolling.

my issue with your understandings is that you believe that it requires a change of at least 1GPM through the loop to alter temps even though I have shown otherwise.

When did I ever say this? And a change in 1GPM how - from existing? From 1-2 GPM? From .5 to 1.5 GPM? You are overly generalizing a scenario including flow that I've don't ever recall saying - because what you are saying doesn't make sense. Flow rate directly impacts DT; this was already known.

Listen - the reason you're seeing what you are seeing is because you have halved restriction so much as well as running more than one D5 in your loop - you've reduced restriction AND added flow and pressure - this is the entire principle of how DT is evaluated.

my other issue with your believe, is that you think that a single D5 pump can push 3GPM flow rate through a radiator, but you have yet to prove that a 3GPM flow rate through any radiator is even achievable (I know for a fact, that it is not possible, through any radiator with a single D5 pump)

I've never said this. It continues to be an argument of yours - a D5 can push 3GPM unrestricted - show me where I said 3GPM through radiators?

by the way my test setup has only one pump (but your always so right on everything)

Many, many of your photos show 2 D5's being used in your testing. Sometimes you mention you are running 1, sometimes you are running 1 at 24v. Which is it?

 

[toolmaker_03]

I do have a 24V step up, but I have never showed a pic of a setup with a D5 running at 200%, or at 24V, I use that for when I build a test loop that has a potentially higher flow rate possible, than what a standard D5 can produce.
keep in mind that a 5-way parallel setup with two CPU blocks and 3 video card blocks has a potential flow rate of 20.5LPM with is above the flow rate of even two standard D5 pumps at 12V in series with each other with nothing else on the loop but a flow meter.(this is only 14LPM)so I have to over volt the pumps to 24V see what restrictions I am running into, so that I can account for them, or remove them from the loop.
my only other option is to buy a Iwaki RD-30, 24V water pump, that can produce the flow rate I need for such a large build, with at this point I do intend on doing, as I do not like to over volt my D5 pumps.

 

[mrface]

Lets try to keep this thread civil and avoid the Urinary Olympics. It seems this thread is on the precipice of that.

Cheers,
Cam

 

[4Ryan6]

Urinary Olympics :lol:

That was Priceless!

 

[toolmaker_03]

to get back to why I have done this was to be able to cool my system way back 3 years ago when I started my build I had 4 X 360mm radiators in series, with the CPU and both video card blocks where in series as well and the system ran hot my flow rate at the time was only 1.2LPM this was unacceptable to me. so I started testing each of my blocks and the components on my loop to get some kind of understanding of what was happening and if there was a way to fix it as I knew I had enough hardware to cool my components. as I played around with different configurations for my loop I started to notice certain changes in my load temps, they started to get lower as I started to get my loop to run faster and faster.
in the end I had enough information to come to the theory, that I presented earlier on in this thread,

ok so this is my theory on how I lowered the delta t of my system with each new configuration of my water loop.
first thing is that, I did need to have enough radiator space to remove most of the heat being generated, by the components on the water loop.
than it is my opinion that the delta t of my water loop could be directly related to the flow rate through the radiators on my system.

this is how I came to that opinion.
on my serial loop setup with one pump, and a lower flow rate through the radiators of 1.2LPM it will have a delta t of 13C
on my serial loop test setup with one pump, the total flow rate through the radiators was 2.4LPM and I had a delta t of 10C
on my parallel loop test setup with one pump, the total flow rate was 6.8LPM so half that is 3.4LPM and that is the flow rate through one of the radiators on that loop and I had a delta t of 7C
on my parallel loop final build with two pumps, the total flow rate was 9.6LPM so half that is 4.8LPM and that is the flow rate through one of the radiators on that loop and I had a delta t of 4C
well that is my theory anyway, it seems to work.

so for some practical application of this theory it seems that for about every 1LPM increase of flow rate through the radiators, the delta t of the loop goes down by 3C.
I would like to test this theory further by seeing if a flow rate of 5.8LPM through one of the radiators will continue to lower the delta t of my loop by 3C if so, it would bring the total delta t of my water loop down to 1C :lol: I don't think that is going to happen, but I would still like to see what does happen to the delta t of my system at the 5.8LPM flow rate through one of the radiators.

I have tested this further and at the 5.9LPM flow rate through one of the radiators, with this setup, it only dropped my delta T by 1C, so now I have a delta T of 3C for this system, I increased the flow rate on the system by changing the position of the second pump to right under the outlet of the second GPU, I had to drill a hole in the bottom of the case and attach the outlet of GPU 2 to the inlet of the pump with a 50mm male to male fitting, by doing this I was able to bring the total flow rate of the loop to 11.8LPM, if i cut that in half i get 5.9LPM flow rate through one radiator, so now we start to see diminishing returns for the high flow rates, but still good information to know.