Overclock.net › Forums › Cooling › Water Cooling › Noob on 1st water-cooling build - need your advise please.
New Posts  All Forums:Forum Nav:

Noob on 1st water-cooling build - need your advise please. - Page 2

post #11 of 14
Here's my views on dual pumps, and it's based on one of Martin's tests which I can no longer find:

There is very little performance noticed in increased water flow above 1 GPM. If you can get 1 GPM, good. 1.5 GPM is better but there is not much difference performance wise.
Of course, two pumps should make a little more noise, but I don't think that is much on an issue if you properly decouple your pumps.
My issue is with the increased heat introduced into the loop by having dual pumps, and yes pumps do put out some heat.
Martin's tests showed that the extra heat introduced into the loop pretty much negates the added performance of the increased flow.

If you can get 1 GPM with a single pump, than I would leave it at that.
The only real advantage (and it's debatable) of a second pump is the added redundancy in case one of them should fail. But today's mobos and VGAs have enough fail-safes installed to prevent serious damage from overheating resulting from a pump failure.

In my next build, I am planning 4 x 120.3 rads with 3 way SLI blocks, CPU block and mobo blocks. I was thinking of getting a second pump because mobo blocks tend to be very restrictive, thus impeding the flow. But I will just try to run my CPU and mobo blocks in parallel to counter the mobo block's restriction and see if that can work.

Anyway, these are just my views, your mileage may vary.

Cheers cheers.gif
Dark Vader
(6 items)
 
  
CoolingCoolingCoolingCooling
39 X AP-45 2150 rpm Gentle Typhoon fans Lamptron CW611 FrozenQ Dual Bay Reservoir 2x XT45 480. 1x UT60 480 ~ 1x UT60 360 ~ 1x UT... 
OSCase
Linux Ubuntu CaseLabs STH10 
  hide details  
Reply
Dark Vader
(6 items)
 
  
CoolingCoolingCoolingCooling
39 X AP-45 2150 rpm Gentle Typhoon fans Lamptron CW611 FrozenQ Dual Bay Reservoir 2x XT45 480. 1x UT60 480 ~ 1x UT60 360 ~ 1x UT... 
OSCase
Linux Ubuntu CaseLabs STH10 
  hide details  
Reply
post #12 of 14
Thread Starter 
Quote:
Originally Posted by Addiction View Post

As for dyes, it is a personal choice. A lot of people will swear by solely running distilled water and biocide. Others will swear they have no issues with the dye based coolants. As far as I know, some people get 2-3 years between flushings of their systems running Mayhems dye based coolants and pastels. So go with what you want in that regards. I would, however, recommend you read some further reviews and comparisons on running both of them, and make an educated decision based on what you want in terms of maintenance. I would recommend though, that if you did go with a dye based coolant, to use Mayhems. His products track records speak volumes to the quality of his coolants comparatively speaking to any other dye based coolant IMO.

Continue from above regarding Addiction's feedback on coolants with dyes.. This is like my 10th time to tell myself to re-think about tubing and fluid colors.

I know it will be smarter and have less trouble to use dye-less coolants... I guess I am making the same mistake that most WC-PC owners would too to use colored coolants on the first WC build. My 2cents tells me showing off with colored flow is important on the first run.

Knowing the facts that WC component manufacturers (EKWB in my practice) would not stand on their warranty if using coolants from others. At the same time I wonder how they can tell if I am not using their coolant without a forensic test. More about this coolant topic, EKWB can denied warranty claims because "not using their tubings, fittings, radiators, pumps...." and resulting their coolants somehow chemically reacts with other non-EK products.

Sigh~ Everything gets complicated when greedy business minds are involved, and in general I know there is no warranty when dealing with WC-PCs. I don't mind a bit of dye residue stained the blocks and clouding up the tubes, but I just pray for no clogging issues and frying components. Replacing tubes and flushing out the fluids should be on routine bases and not to be too hard. Dye stains should not be much visible when using the same pre-mixed fluids in my opinion. I can also imaging there are not many tubes need to be replaced since most of the tubes are not in the motherboard chamber in my setup.... (as my plan with each individual chambers within TH10 and also 2x pedestals)

Thanks for bring up the "Mayhems" here because it was strongly recommended by a guy who runs the only shop supports WC components around my town. (his name is Mike and he should be a member here too)

I live at the center of Silicon Valley where it takes only minutes to drive to headquarters of Intel, AMD, Nvidia, Corsair, Asus... and it is just too rare to see WCPC guys out here...

Thanks to OCN and members again,
Jinx
post #13 of 14
Thread Starter 
Quote:
Originally Posted by PepeLapiu View Post

Here's my views on dual pumps, and it's based on one of Martin's tests which I can no longer find:
There is very little performance noticed in increased water flow above 1 GPM. If you can get 1 GPM, good. 1.5 GPM is better but there is not much difference performance wise.
Of course, two pumps should make a little more noise, but I don't think that is much on an issue if you properly decouple your pumps.
My issue is with the increased heat introduced into the loop by having dual pumps, and yes pumps do put out some heat.
Martin's tests showed that the extra heat introduced into the loop pretty much negates the added performance of the increased flow.
If you can get 1 GPM with a single pump, than I would leave it at that.
The only real advantage (and it's debatable) of a second pump is the added redundancy in case one of them should fail. But today's mobos and VGAs have enough fail-safes installed to prevent serious damage from overheating resulting from a pump failure.
In my next build, I am planning 4 x 120.3 rads with 3 way SLI blocks, CPU block and mobo blocks. I was thinking of getting a second pump because mobo blocks tend to be very restrictive, thus impeding the flow. But I will just try to run my CPU and mobo blocks in parallel to counter the mobo block's restriction and see if that can work....

I did see that Martin's test result somewhere and I remember some charts (jpg file?) were missing already in that post. The numbers did not mean too much to me at that time but I pick up the idea of getting redundant pumps can double safety insurance just like you said.

And... yes, I saw and thought about parallel flows between CPU and ChipSet blocks too. You can call me a chicken but I won't do it unless I know the flow resistance in comparison between CPU_BLK and MB_BLK(s), and I have monitoring over the flow rates of each path. Just from my understanding and common sense from my EECS background, that there is a chance of small (minimum to none) flow can occur to one of the paths since there are difference in flow resistances in each path. Just like putting electrical resistors in parallel in a circuit, the electrical current (or flow) can be dramatically changed due to changing in resistance (or clog build up). In our case, the debris/residuals would build up faster in the path with smaller flows, and this build up can grow exponentially fast....

Haha.. more WC builds for you? This will be my first build and hopefully my last build too. My typical computer would last 7-10 years, and I only do dust-out, redo the thermal paste, and adding HDDs when necessary. From this build I just pray to keep the same chassis, reservoir, radiators, pumps, and fittings for next build. I don't leave task computer on idle for long, and WCPC is not cheap at all~~~~~

I actually has a lot of questions and doubts about why people do WCPCs... WCPCs cost more money, have higher risks, and require more maintenance overall. I don't want to vent it out now.... not until I finish my first build.

Thanks again,
Jinx
post #14 of 14
Definately, running the mobo and cpu in parrallel is not for everyone. I will need to test around and fiddle to ensure both sides are getting sufficient flow.
What I would like to find is some kind of valve I can ajust to do just this. For example if the mobo side is not getting enough coolant, I just dial down the cpu side's valve to ensure more flow to the mobo side.
This will need to fiddle around a bit.

Cheers cheers.gif
Dark Vader
(6 items)
 
  
CoolingCoolingCoolingCooling
39 X AP-45 2150 rpm Gentle Typhoon fans Lamptron CW611 FrozenQ Dual Bay Reservoir 2x XT45 480. 1x UT60 480 ~ 1x UT60 360 ~ 1x UT... 
OSCase
Linux Ubuntu CaseLabs STH10 
  hide details  
Reply
Dark Vader
(6 items)
 
  
CoolingCoolingCoolingCooling
39 X AP-45 2150 rpm Gentle Typhoon fans Lamptron CW611 FrozenQ Dual Bay Reservoir 2x XT45 480. 1x UT60 480 ~ 1x UT60 360 ~ 1x UT... 
OSCase
Linux Ubuntu CaseLabs STH10 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Water Cooling
Overclock.net › Forums › Cooling › Water Cooling › Noob on 1st water-cooling build - need your advise please.