• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

New watercooling (my first one :D)

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.


New Member
Jun 30, 2005
Hmmm where to start, I've been working on and off on this project for
ages now mainly because i was wating to buy stuff which meant i had to
delay the build. Initially I was going to get a Hydrocool Hydro-Stream
HS5 waterblock, but it never seemed to come out so I gave up wating
and settled on the Danger Den RBX. (sorry about the poor pics btw, my
cameras rubbish and my systems tuff to get to).

I made up a spreadsheet for some calcs of pressure loss and flowrates
for an Eheim 1250 pump, and using the dimensions of all the tubes in a
Ford Fiesta 2 heater matrix (UK btw for any US reading thinking WTF?).
The calcs showed there was plenty of pressure head for the system i
wanted to build so I went on and ordered everything.

I knew I ws going to upgrade my pc soon so I bought the Athlon 64 block
with lucite top, could have got the brass top but I thought it might be
handy to be able to see the o-ring seal and any blockages that might
occur. The tubing I used was 1/2" dangerden stuff and it doesnt kink,
it just ovalises a bit when you bend it, its pretty nice stuff. As I
said Ehaim 1250 pump, along with Jubilee hose clamps (the metal screw
type) some plumbers PTFE tape (turned out to be essential) a nozzle kit
for the water block and some 1/2" barbs...I think thats it!

Some may be wondering what about a reservioir? I say, already made in
my spare time out of Stella cans.--> I cut holes where needed for water
to flow between the cans (3 in total) and used Araldite to stick em
together. I then cut a hiole for a barb and glued that in the bottom
of the can tower. Later on in the build this leaked a little, I tried
to seal it up with silicone sealant but it didnt work. Also in this
leak plugging stage i managed to destroy the tower by leaning on it by
mistake-none too happy. So I remade it with alot of araldite to join
them, then in an effort to take the load of the joins I stuck on some
small bits of strip metal to bridge the cans. I then removed the barb
completely, along with the old araldite which had peeled from the
smooth surface of the cans. I then refitted the barb using silicone on
the inside of the can (using my finger) and some bits of flat rubber
washers and slicone on the outside. This didnt leak once after I left
it full over night so I was satisfied it would be fine. Pic below



The exit of the reservoir at the bottom of the cans lead straight into
the suction side of the pump, which I had fitted with a 1/2" barb for
easier connection. The standard suction fitting on the 1250 is 3/4",
and I couldnt find tubing/ barbs for that size easily. I used a short
piece of tubing between the two because the inside diameter of the two
barbs was slightly different, so the flow would get messy as it would
have to negotiate a step, the tubing gives the flow a little time to
settle and allows for some misalignment too.


The barb on the suction side came with my pump but only because the
shop I got it from does that, normally thay dont come with an adapter
barb. I forgot to put PTFE tape on the thread before filling the
system so I ended up with a small leak from that area. I drained the
system and put loads of tape round the thread, I also stole the O-ring
off the original suction side nozzle of the pump. I then reinserted
the barb and then wrapped more tape around it just in case. That fixed
the problem nicely. On all hose clamping areas, I wrapped PTFE tape on
the barbs and electrical tape around the hose where the clamps might
cut the tube.

I used the original connection barb that came with the pump for the
pressure side, I could have used one that I bought but the thread on
the one I bought was longer than the othe other and I didnt want it to
foul the impeller of the pump. The pressure side hose then connected
to the heater matrix. I stretched the hose over the bulbus ends of
the heater matrix tubes, it was pretty hard to do that but got there in
the end, after doing that I then found out on here that heating the
tube up a bit might help (in some warm water or similar). I then
clamped the hose etc. Oh yeah I mounted the pump on some rubber mounts
to try and isolate any vibration.


The exit of the heater core connected to the inlet of the RBX
waterblock which I fitted with the narrowest slot nozzle. The
numbering of the nozzles has changed, I think in the old numbers it
was number 4. I considered using the number 5 (old number which is now
number 2) with the multi jet configuration but others have said that
they block easily so I avoided it and went for the narrow slot to
increase the inlet velocity as much as possible to impinge over the cpu

The two outlets of the RBX were joined using the Y adapter supplied
with the block which loops back out of the case....


....and into the top of the reservoir.


I used a bent pesice of stip metal and wrapped electrical tape around
the hose and metal to make a hook so it would stay in the cans drinking
hole easily without cutting the tubing. I then filled up with
distilled water and zerex racing super coolant (1 115ml bottle) and ran
it over night with no leaks. I then had to wait a few days to build my
new PC.

After various problems backing up files and building my new system I
was ready to attach the block onto the CPU....er no! What the **** is
that in my tubing! There were a cple of large lumps of brown
something, which I guess was dirt and rust etc left in the old heater
matrix even thought I had flushed it :bang head . So I had to drain down again, back flush the system with a hose (not my preference to use tap water for micro biological reasons but I wasnt going to take it all apart
again) and refilled the system. I then attached the block to the cpu not forgetting to put some artic silver over the heat spreader and off I went. I also tried to make sure the cables were pretty tidy:


When it was 29 C in my room on the first day I ran Prime95 torture and
it got up to 46 C. I ran it yesterday and only got to 38C, the lowest I've seen it was 30 C at idle. I pretty sure I could get even better temps but I have a very low flow 120mm fan (acoustifan dustproof) which doesnt shift much air but u can barely hear it, even if you listen carefully. I aslo used a silicone gasket and rubber washers to mount the fan, along with some packing foam between any contacting areas to prevent rattling. You can see the gasket in the next pic.


On large areas of the box I placed some more foam to try and stop the
box droning slightly as the air moved through it and placed the box on
some more foam as it was resonating throught the floor boards.


Putting the box on some foam made quite a difference even though its
super quiet anyway, but I cant help trying to make things better :D
Its not even anodized, so it will be eaten quickly. Just a warning, It will corrode if you don’t take precautions. However, I forgot to say that it will start around where you cut holes. When I do my electrolysis experiments I often use a cut up soda can, and they have some sort of plastic coating inside. This protects it from the acidic soda, and until scratched its anti-corrosive(used the soda bottle as the ground, was making sharp things. VERY sharp) It then began to bubble and peal off into the then brown water.
er yeah, i put anti corrsion stuff in-zerex, although I could do with putting some more in as the mix is a bit weak cos i had to flush it and lost some of the additive in the process.

Yeah I did think about the metal round where i cut the holes in the cans, but fingers crossed it will be fine for a while. At least the cans are outside the case :rolleyes:

cheers peeps