H100 tubing?

Does anyone know what size the tubing is on the H100. I'm looking to replace the black tubing with clear, when I get it that is...

Its a closed system, your not supposed to replace it, mod it, open it, replace the water/liquid ever.

Thank you captain obvious, I'm not keeping my unit stock. I'm perfectly aware that, if left in its stock configuration, there is little to know maintenance required with this unit, other then blowing out the rad with compressed air occasionally. I'm going to add a reservoir and clear tubes to this system. I was simply asking if anyone here knew off hand what size the tubing was.

But it's nice to see that your vast experience on this forum has been put to good use, by posting something that doesn't have anything to do with my OP.

 

But thanks anyway!!! </sarcasm>

I have already answered this question in another thread posted before yours, stop being a smart *** and find the answer which was already given.

Tube ID is 6mm. Standard 6-ID\8-OD tube works a treat. I've done the clear-tube mod and added a reservoir.

 

 

 

 

Hello all I'm new to this forum. i went in search for answers regarding a custom set up. i built my own computer from scratch and because i wanted it to be compact I have made several mods to just about everything this is what i have for the corsair h100i tube mod. i couldn't find a lot of information so i'd thought i put this out there for our fellow modders.

 

corsair with 30 feet of clear 9mm tubing from lowes. 1 strait splice. corsair h100i 45.00$ referb model tube and connector 11.50 and i had some 50/50 in the garage for fluid.

 

heat the ends going to both rad and water block (it tight) the splice fit tight but heat was not needed.

 

after the tube is connected to everything i wrapped it around the posts and cut it in the middle i didn't figure i would have a lot of moving stuff around after i filled the system.

 

i tinker for a living so to speak so i have no problem doing it this way. less exp may want to do a dry run :!:(not on the mother board).:!: before attaching everything in place.

 

to force the 75% distilled water 25% antifreeze in to the system i placed one side of the tubing into a empty jug. then the other in the bottom of a 1 gallon jug of my CPU coolant. blowing air in to the top of the fill jug forces air free fluid into the lines.

 

i turned the computer on all sides and shook it several times and had good luck getting all the air out. on mine as long as the receiving jug was lower then the on i was blowing into it would siphon on its own. i plugged the ends with my fingers, got everything in place and stuck them together 1 little tiny air bubble got in but it settled in the rad so no worries.

 

i hope the attached pics show everything if you have any questions let me know. Happy modding!!!!

sorry i added the pic showing the splice

so,post some stress temps with your unique configuration,you've certainly peaked my curiosity ...

 

ill post more of actual use, (game call of duty) . i have ran i stress test all night and temps stay in the 60C range

 

this after 2 hours of playing cod ghost on max settings

 

 

After replacing the 25% coolant solution with 40% water wetter dosent look like much difference.

but it red!:laughing:

good temps you got going there,looks like things are playing well together...;):

What sort of flow rates were you getting with the pump unit from the h100i sir ?

I have seen a video of a very low flow rate. I will be testing my own when the new rad arrives, but I would be interested to know how you went on ?

LOW flow LOW head pressure its like 16 inches. I don't have a meter but it is very slow. I did see a thread some ware that they tested it. I will see if I can find it. basically I have noticed that I have a long sealed loop that it pushes water "good enough" to cool my CPU. I don't have it overclocked. I did increase the boost speed on the CPU to 4.0 but only under load.

 

I will say this though after I found out the flow rate was low I am still impressed with the consistent pressure. and quite operation.

 

Link http://www.overclock.net/t/1371863/corsair-h100i-max-flow-rate-test-video-result

I am looking at changing just the pump and adding a custom built res on top of the case.

 

I think its more the pump that has reduced flow because when I was filling the loop I was pushing way more water the .11 gpm. so I will see if the head has enough pressure to handle my next addition (RES) if not then ill add and after market pump.

I changed to 75% distilled water and 25% water wetter. temps didn't change much but as I said before now its red :D:

I am looking at changing just the pump and adding a custom built res on top of the case.

 

I think its more the pump that has reduced flow because when I was filling the loop I was pushing way more water the .11 gpm. so I will see if the head has enough pressure to handle my next addition (RES) if not then ill add and after market pump.

 

It is an interesting thing, the low flow rate - but it would appear logical to presume that the actual performance of the heat exchange at the CPU end is very, very good. Any problems - up to a limit - with the 100i, would be at the radiator end. I believe the design of the radiator is poor. You can see the effects of the max speed fans in the tests, and a slight increase in push pull. That would mean the block and pump are ahead of the radiator and fans. - even with the low flow rate.

I await the results of the test I am doing by adding a Koolance 140*2 (280mm copper radiator) and 2 Noctua 140 fans. I dont see any reason why the pump / block cannot handle much bigger rads as any flow restriction would appear to be in the pump itself, and as said, not causing an issue.

I have seen this cooler perform as well as a full custom loop on 6 core processors running high Vcore (150 - 170W dissipation)

A lot of people seem to think the fastest flow will give better cooling - I would tend to disagree as it all depends on the design of the block and the ability of the fluid to physically stay in contact with the metal long enough to transfer the heat.

Again, purely in theory, I cannot see any custom loop that would seriously outperform this unit on processor only. - it may be better, but not massively.

Interesting stuff.

One may also conclude that if you could force more water through the h100i block, then you may in fact degrade cooling as you may introduce turbulence into the micro-finned area in the block.

The 100i is amazingly good at what it does and a superb price. - I can only presume a better secondary heat exchanger (rad and fans) would improve it.

ok when are you doing the testing? I will be interested to see the results. hum I didn't think that the rad would be the problem.

i would think the pump would be the weak link due to the increased volume of water

ok when are you doing the testing? I will be interested to see the results. hum I didn't think that the rad would be the problem.

Hi sir,

I have just ordered the Rad from the UK, we are on our boat in Portugal at the moment so it has to be sent. I have just ordered another H100i from Amazon UK also (free delivery to Portugal) but this is only to keep as a backup (they are out of stock until Feb 1). The one I have here can be used as a test unit though.

It may all turn out badly, I dont think so but you never know.

If you look at all the actual results of testing with the H100i though, it would certainly appear the radiator heat exchange is the limiter as the more air one can push through the rad the better. In a similar way, if one could find a second hand H100i rad, then add two in series and I think the results would be good also.

If you look at the rad design, the FPI is very low, leading to minimal surface area to dissipate heat, but, allowing quite large airflow. I think this is for cost saving purposes rather than by design ;): A greater FPI with a copper / brass rad would be more efficient and probably quieter. but more expensive..

Hopefully the stuff should be here from the UK around the end of next week. SO then I will play, I have just bought the C70 case though so need to strip out the previous build and transfer it. As I will have a complete H100i and also the Hybrid Koolance 280 (2*14) / Noctua 14 / Corsair Hydro i, I should be able to swap over the units fairly easily.

If it works, I will mod the other H100i for two standard corsair 2*120 rads and fans. I can see absolutely no reason why that would not work and improve cooling providing one could fit it in the case.

At the end of the day the worse case is you invalidate the warranty, I doubt very much that you would damage the units if care was taken. I would then just have to order a D5 vairio / tacho pump and Apogee block:bigeyes: (then add the pcb from the H100i and possibly a pic micro to make the board think it is still controlling the original block and pump...

But, that is for the future lol.. one gets carried away (or should be) :)

Cap.

i would think the pump would be the weak link due to the increased volume of water

 

Hi,

Surely the limit is the head height, not the volume. The pump itself is forming a natural restriction (as it would appear to be designed to do). If you look at the volume of the water in the post with the large pipe loop, there is quite a volume there.

I think if you tried to pump to a far greater height you would have trouble, but below the pump and above to the height that the original pipes allowed I cannot see how it could make any difference, the pump just sees water at its inlet and passes it through, it has no idea of volume. Pressure is not increased etc, we are not adding restrictions in any way.

Cap.

 

Ok looking at the pic this is what I'm thinking. hot water leaves the block(RED) to the radiator, leaving the rad to the reservoir I build on top of the case(ORANGE) one side of the top to the other 9 3/4" away. travils down to the lower loop (BLUE) after that I would guess close to room temp. then travels up to a heat exchanger attached to the GPU stock fan coolers heat pipes* (LIGHT GREEN), then back to the pump block.

 

* heat exchanger copper on copper tied together with copper thread. maybe a little thermal paste in there to.

 

thoughts: water above and below the pump to prevent pump strain, thermal dynamics hot goes up and cold goes down. natural flow lessen pump strain.

 

I don't feel that the heat exchanger will pickup to much heat to affect the CPU temp much and if it does ill change set up. Any little bit to cool the GPU in my opinion will help.

 

I'm putting this out like this because I want opinions. and I like finding the lowest cost for the budget DIYers

 

cost so far for just the cooling loop" 65.00

I think that would be worth a shot, absolutely. Just ensure you feed your pump directly from the res. (that applies to all loops really)

 

As to the effect of volume, it can only be an issue in 'weight' (Pressure) at the input to the pump, (unlikely to cause a problem - or, as a restriction to the input to the radiator.

If you had a 1 gallon bucket with a pipe in the bottom and had an inch of water in it, you could easily pump fluid in, if there was a foot of water in the bucket, then it would be more difficult. One can get around this easily by placing the inlet to the bucket higher up, and the outlet at the bottom or lower down. Same with higher volume radiators such as this one for example. - the options for fittings give you a way of negating the effects of pressure on the pump oulet.

http://www.watercoolinguk.co.uk/p/Alphacool-NexXxoS-Monsta-240mm-Radiator_25156.html

 

The small images gives a good idea, it is more down to the choice of rad. A longer / wider radiator of the same thickness mounted in the top of the case would make no difference to pressures at all. A deeper radiator would possibly cause an issue if bottom fed, but as can be seen , it doesnt need to be.

As you say, choice of route and loading is key. (IMHO)

I think that would be a good mod.

btw, you can get some super radiators from caravan and boat 12v fridge units where they use them as the heat exchanger, superb little beggars.

Small fridges and freezers can also supply some interesting finned pipe exchangers from the back that can be fitted inside a case panel..... my first cpu cooler used parts from a bar fridge and a home made copper block. Pump was a 240V Eheim fishtank pump submerged in a tupperware container - Piping was silicone surgical tubing (still good) worked great !.

Model shops (rc models that is) sell brass tube in rectangular shape. these can be cut and soldered easily into ram coolers and mosfet coolers.... Just add some 10mm Microbore copper for the pipe, you can make a barb by soldering a microbore compression fitting ring to the pipe stub.... but, I digress lol.

I took a second look at the pic and i made a mistake hot from the CPU>top of the rad. exit the rad bottom then to the res on top of the case and so on.

I took a second look at the pic and i made a mistake hot from the CPU>top of the rad. exit the rad bottom then to the res on top of the case and so on.

Done that way you are reducing any real loading on the outlet of the pump. I think that would be a real good system, just dont add any restrictions. The res may help a lot in the actual day to day running and setup, I actually have ordered a small 100mm one with my Koolance rad, but would like to leave it a sealed unit if I can, if I have issues with air and filling, I will use the res.

It is always there if I ever do go custom loop.

I run 2 gtx 660 TF MSI gpus and they kick out about 130W each, but on the c70 case the fans in the side panel will pull air out directly over them (2 corsair 140 on the plexi panel) the 140 Noctuas are going on the rad as Corsair didnt do a higher pressure fan in 140 that I could find (and I like the Noc's !!)

Also, I dont find a need to OC my gpu's. but I do want the processor OC'ed as I am into flight sims and ya need all the grunt you can get on the cpu.

On the MSI Twin Frozen 660's the fans are pretty quiet anyway so no worries.