-
Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
90 Degree right before pump?
- Thread starter eightballrj
- Start date
- Joined
- Jun 11, 2004
Removing the elbow at the inlet of the DDC is responsible for most of the aftermarket top performance boost. I'm not sure if that's because the pump cavitates in its stock config or if it's because the elbow causes the impeller to fill unevenly, but the proof is in the flow rates.speed bump said:
All the blocks in a loop have the same flow rate, and you'll loose the same amount of flow no matter where the elbow is in the loop. As long as it's not right at the inlet, it's all the same to the pump.Mycobacteria said:
- Joined
- May 15, 2006
Yes, while blocks usually have a 180 degree turn, they are not made the same way the bends are made. The 90 degree bends are just that, bent, in which you lose flow rate (just like cheap car exhausts). Block are made to be lower restriction and therefore don't tax your pump as much.Otter said:
I can't remember where the article is that I read a while back regarding this. I just remember that having a 90 before the pump seriously hurt the pump's performance. I may have been on procooling, but I can't for the life of me remember.
Ha, found it. I though I had it bookmarked. http://www.procooling.com/index.php?func=articles&disp=37&pg=1
I also read where this was tested, but can't remember where that one was...
Ha, found it. I though I had it bookmarked. http://www.procooling.com/index.php?func=articles&disp=37&pg=1
I also read where this was tested, but can't remember where that one was...
Last edited:
OP
- Joined
- Jun 11, 2005
- Location
- Missiskippi
- Thread Starter
- #27
The key to the story is:
Eliminate as many hard bends as possible. Water is incompressible and an elbow before or after the pump will not affect overall flow of the system with everything else being constant. However, directly before the pump could cause enough turbulence to cause cavitation in the pump. If no cavitation then no problem.
Correct?
However, I will no be using any elbows as I will be changing the loop to avoid any hard bends.
Eliminate as many hard bends as possible. Water is incompressible and an elbow before or after the pump will not affect overall flow of the system with everything else being constant. However, directly before the pump could cause enough turbulence to cause cavitation in the pump. If no cavitation then no problem.
Correct?
However, I will no be using any elbows as I will be changing the loop to avoid any hard bends.
- Joined
- May 15, 2006
Yes, because of the way they make the bends. They actually bend straight tubing, cause it to cave in a little on the inside of the bend. This decreases flow. If they make it the proper way where a very hard piece of metal is inside, and then they bend it, no.Otter said:
- Joined
- May 11, 2004
- Location
- Montana tech Butte mt.
Otter said:I never have been able to make heads or tails out of the equations in that article.
Number of velocity heads?
i'll try to explain this to you.
Head loss in the energy equation is defined by
((k1+k2+k3+kn...)+f(L/d))*v^2/2g
Friction factor f which is determined using Ks/d, reynolds number, and a moody diagram. You can probably search and find a good tutorial on this somewhere.
K is a factor of restriction for a 90° Miter bend it is 1.1 without vanes or .2 with internal vanes.
For swept bends it bcomes a factor of the bend radius divided by the diameter of the pipe.
For fittings they are considered constants for each general type of fittings.
L/d is the length of piping divided by the diameter of the piping.
V^2/2g is called your velocity head. It is your velocity(ussualy dtm from flow rate) divided by 2*your gravitional accleration constant (32.2 Ft/s)
That V^2 /2g is almost nothing at 2gpm though so I would be much more worried about WB restrictions rather than 90° bends personally.
Also I repeat my question whats so bad about having a 90° fitting right before you inlet. It should only be as bad as a 90° fitting anywhere else.
I don't think the DDC with a modded top is a good reason though becuase under restriction you have barely any higher flow rate with a modded top VS the stock top(about .1 GPM) and that could be attributed to the modded tops being 1/2" rather than anything I can think of.
- Joined
- Jun 11, 2004
Thanks for the clearer explanation of the equation. I'll go over it later when I've got more time. Hazen-Williams looks useful too.
http://www.kmisystemsinc.com/files/Technical References/PressureLoss.pdf
http://www.systemcooling.com/images/reviews/LiquidCooling/Swiftech_MCP350_Mod/image10big.gif
(full article: http://systemcooling.com/mcp350_mod-01.html )
If the elbow problem is turbulence causing cavitation, then do pumps in series need some tubing between outlet and inlet to keep turbulence from the first impeller from causing cavitation in the eye of the second one?
http://www.kmisystemsinc.com/files/Technical References/PressureLoss.pdf
It makes more of a difference than that. It's true that the difference is less with the DDC-2, but with the 10W pump, it's quite dramatic. Check out the PQ curves.speed bump said:
http://www.systemcooling.com/images/reviews/LiquidCooling/Swiftech_MCP350_Mod/image10big.gif
(full article: http://systemcooling.com/mcp350_mod-01.html )
If the elbow problem is turbulence causing cavitation, then do pumps in series need some tubing between outlet and inlet to keep turbulence from the first impeller from causing cavitation in the eye of the second one?
- Joined
- Jun 11, 2004
This is what an elbow looks like. It's made to be this way and does not crimp.thideras said:
http://www.petrastechshop.com/1odbaco90el.html
- Joined
- May 11, 2004
- Location
- Montana tech Butte mt.
Otter said:
The hazen williams eq is simply a more refined version of the Head loss equation as applied to that application.
As far as difference I wasn't talking about free-flow difference, becuase it should make a fair amount of difference there do to that being the largest restriction placed upon the pump in a free flow situation. However as restrictions in other places(such as water blocks) are added it matters less and less.
to see what I am saying check out petras tests here.
http://www.xtremesystems.org/forums/showthread.php?t=117215
http://www.xtremesystems.org/forums/showthread.php?t=137484
- Joined
- Jun 11, 2004
Yes, I've seen those. Rather obviously, I wasn't talking about free-flow either. Of course the performance difference will narrow the closer you get to where the PQ curves cross, and if you're going to use the two most restrictive blocks on the market, you might as well stay with the stock top.speed bump said:
My point, though, was that the performance improvement with the modified top on the original DDC demonstrates that having an elbow at the inlet of a pump is not a good idea. I suppose you'd have to test with a similar narrow elbow elsewhere in the loop to really prove the point, but I doubt that would produce as large a difference as the inlet elbow.
