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

Question about barbs and elbows.

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

Bon3thugz43v3r

Member
Joined
Oct 6, 2002
Location
The Sunshine State
Ok, so the way my watercooling is set up, the tubes crimp because they are like an L. Well my crimps are right where the barbs are on the pum/res. So I ordered some elbows. My question is, is there anything I should put between the elbow and the barb? Like some kind of rubber or something? I dont know, but maybe somone has experience with this. Thanks.
:thup:
 
stay away from elbows!
i recommend doing anything in your power to not use elbows!
they will kill flow...

if you want to get rid of the crimp, try moving stuff around, use more tubing, anything except elbows
 
Why would elbows kill the flow in a WC system? Any data to back up that claim?

I had to use several of them to get things to fit in my case (antec 835II series). I would be very interested to hear what performance drawback each elbow contributes? (I got five right now).
 
I think in one post someone said one elbow is equivalent to about 10 feet of tubing or something. So 5 elbows would like running the water through 50 feet of tubing alone, which is very bad.

Also each elbow is just 2 more places where it can leak :D

What kind of tubing are you using? If you are using cheap vinyl tubing, get some clearflex 60, or tygon, they probably won't kink.

Do you have any pics?
 
iceblock said:
Why would elbows kill the flow in a WC system? Any data to back up that claim?

I had to use several of them to get things to fit in my case (antec 835II series). I would be very interested to hear what performance drawback each elbow contributes? (I got five right now).
Elbows are the WORST flow killer of them all, especially at the INTAKE of the pump. Do yourself a favor and read THIS
 
nikhsub1's link is a great article- I am lousy at math so all the equations don't help me, but the explanations and descriptions do.

I, however, have always been one to test theories in practice as much as possible (which has cost me hardware at times....;)) so I have done my own testing with and without 90s and 45s.

Invariably, temps have improved with the systems that do NOT have any fittings. Not always a huge difference, but at least 1c and as much as 5c. In 3 separate systems, tested to different degrees.

My hardest test was all copper tube with 9 90s and 2 45s, an Eheim 1250 and (slightly modified) Swiftech MCW462u waterblock.
Temps were about 2 or 3c average lower when I had no fittings at all. Tubing was longer, same ID, same system.

Fittings are bad.
 
Well.... After spending all day Saturday trying to "fit" my new WC setup in my case, I gave up and went and bought the 90' elbows to get it to fit.

The main reason is the tubing I ended up with: Tygon 1/2"ID 3/4"OD w/ reinforced braiding. It seemed flexible enough in the store, but trying to flex to make it fit inside my case seemed very difficult. What I've learned, and what people seem to overlook, is that Tygon is a general tubing brand that covers lots of different types of tubing, some are less flexible than others....

So, I'll keep looking for the ideal (flexible) type of tubing... Until then I am stuck w/ my 90' elbows :(
 
iceblock said:


So, I'll keep looking for the ideal (flexible) type of tubing... Until then I am stuck w/ my 90' elbows :(
What you need is Tygon Formula R-3603. I use 1/2" ID 3/4" OD and you can bend the heck out of it before it kinks. It is also VERY flexable.
 
iceblock said:
The main reason is the tubing I ended up with: Tygon 1/2"ID 3/4"OD w/ reinforced braiding. It seemed flexible enough in the store, but trying to flex to make it fit inside my case seemed very difficult.

What store did you go to get that?

I think that braiding is meant for high pressure, not necessarily flexibility.
 
I got it from a local plumbing/hardware store (Home depot/lowes/OSH only had vinyl types), and they only carried the braided one.

Where can I find the R-3603 Tygon fromula? Anyone got it locally, or mainly from an online order? If I can find it locally, I'll try and upgrade my set up soon. (I'll try and post some pics soon).
 
I know www.dangerden.com sells it for watercooling. Mine is being sent right now:)

I think the reason the elbows slow the flow has to do with the water having to change directions suddenly, and energy is lost to friction, correct me if this is wrong (just finished 9th grade physics).

-Noah
 
oh to simplify your questions answer without reading the article ( which you should do anyways) elbows change the direction of water very abruptly and the water uses energy to travel through the tubes but imagin throwing a ball at a brick wall and the force will rebound the ball when hit straight. so you just reduced the energy of forward movement of water which reduces flow.
now throw the ball at a gradual angle in the gutter or at a curved wall and the force isnt rebounded but only slightly reduced but the ball continues to have forward energy

as i understand it thats pretty much how sharp turns reduce flow but i could be wrong =)
 
i dont know, but maybe at a hardware store you could find somting in between, like a 45 degree angle, or an elbow with a slow curve insted of a sharp 90.
 
Well what about to the original question. Can I just slide an elbow onto a barb and be done, or is there something I should use to connect the elbow to the barb?
Thanks:D
 
Bon3thugz43v3r said:
Well what about to the original question. Can I just slide an elbow onto a barb and be done, or is there something I should use to connect the elbow to the barb?
Thanks:D

When I did my experiment with 90 degree elbows I connected things by using tubing between fittings or soldering elbows together.

With barbs, I would suggest using some tube as a joint- you only need a small amount and the fittings can be touching so it can work in a tight spot.
 
Bon3thugz43v3r said:
Well, what stops flow more, kinds or elbows :(
You decide...
I think that elbows would work better than a kink.
Anyone agree?
Point is you dont need to have either... If the tubing needs to be a bit longer, so be it but it would be better than a kink or a MITRE elbow. A 90 degree SWEEP elbow (copper) is 1/2 as restrictive as a MITRE elbow (sharp 90 degree, usually plastic or brass). 45 degree sweeping elbows are 6x better than a 90 degree MITRE elbow. Here is the coefficient chart on fittings from the article i linked to above by pHaestus:

FittingK coefficient
-----------------------
90 degree Mitre 60f
90 degree Sweep 30f
45 degree sweep 10f
T straight through 20f
T through branch 60f
 
actualy because of the mounting of the beast im building and the size of the clearflex60 i may also run into a restrictive flow problem with my design in 2 spots so i was considering getting a brass threaded coupling to copper tubing because you can bend it in a curve more sharply without restricting flow as much nd the pipe is few millimeters thiner due to it being copper so i might be able to route it easyer . wait till you guys see the pictures once i finaly get my parts and ill show you what im talking about :)


anyways my question nik is did you test various curves that wher not joints too ? i would imagine the coeficient would be much less for them but again i could be wrong
I want a semi tight 90degree turn using copper tubing (possibly less of an angle i hope) and i hope that metal tubing will be the answer
 
Back