Help with unexplained loss of performance in my loop

[dopax]

Greetings to all:
I've a loop with 4 blocks ( Maimboard , CPU, GPU and a block of aquaero XT ) ; one reservoir Aquacomputer Aqualis ; one triple and one quadruple radiator ; the loop is moved by 2 pumps MCP-35X with dual top EK , also has 2 Aquacomputer high flow sensors . All pumps at100% gave a yield of 215 l \ h ( approx ) in the loop .

The point is that my reservoir and dual top of the pumps were joined by two elbows of 30 degrees , among which there was a section of tube (the group forms 90 degrees in a relatively short space) very short (5 to 6 cm long) that , over time , to do some flexion , observe who was folding , with the consequent danger of flow restriction.
I replaced that section of the loop by a combination of fixed parts (some Nozzle adapters coupling ; a single fiiting 90 degree swiveling ; etc) forming a rigid assembly with 90 degree bend fairly smooth (the single 90 degree fitting is based on three 30 rotating bodies each) without the danger of tube section was folding by forcing the curve being so short in length.

My problem is that I found the unpleasant surprise, since I can't find explanation , that now only gives me 128 l \ h ...............This is why I ask for help from anyone who can give me an explanation that could happen . I can't find a logical explanation , since I've only replaced nothing more that a potentially dangerous short section for something better and less restrictive.
Of course , I checked the operation , separately , of both pumps, for if he had broken some by chance; is more , I haven't even been forced to completely empty the liquid loop (I've a valve drain on the lowest part thereof) ............... i don't know , i can't find explanation for this and I am blocked without knowing what to do.
Thanks in advance

 

[C12Friedman]

How old are the water blocks and when were they last cleaned? I'd venture a restriction in a block somewhere.

 

[1LiquidPC]

Can you provide any pics of what your describing? Any pics from previous setup for a good before/after comparison might be helpful too.

 

[dopax]

First of all, my apologies, because I had to have put this thread in the section of watercooling. If administrators of this forum consider moving it around, for me, perfect.
Forget comment that if all this seems strange is that the loop has been operating six months; components are new and besides, this happened suddenly, not gradually over time.
Sending photos from the section when was the problematic tube and subsequently with section replaced.

http://i44.photobucket.com/albums/f17/dopax007/El%20rincon%20del%20Peter/IMG_1553_zps119b0c44.jpg
http://i44.photobucket.com/albums/f17/dopax007/El%20rincon%20del%20Peter/IMG_1556_zps0c9f44cf.jpg

 

[1LiquidPC]

I do have one theory at the moment. Is it possible that when you were using the small section of tubing as shown in your picture that was collapsing over time due to the tight bend that it was creating somewhat of a "garden hose" affect?? I'm picturing the adjustments on a garden hose nozzle. What I mean is the pressure of the water being pulled from the reservoir to the pump was high because of restricted space in the tubing due to it collapsing. This increased pressure may have given you a higher somewhat false reading to your flow sensor since it's right after the pump. After you replaced the tubing with solid fittings(good idea based on pics), the restrictive section was gone as water had a larger opening for a flow path therefore slowing it down a bit. Again though, this is just a thought.

 

[dopax]

I wouldn't know, but I think not ........................... initially, the tube was not folded; I guess with time, forcing the curve, it was folding.
In addition, previously had another reservoir and had another type of junction with the reservoir (now use the XT Aqualis by the level sensor), and in all cases the performance was similar (214-217 L \ H)

 

[4Ryan6]

Any tube crimp like that in the future just use nylon wire ties to get it out, there was really no reason to have to disassemble.

However since you did disassemble, have you gotten all the air out of the loop?

 

[dopax]

I thought it was too short to adopt this solution you mention me
For the air bubbles , once I came back to fill it flush more than two hours with the top fillport open to bleed the loop, even shaking the pc chassis .
On the other hand, the loop had only running 3 or 4 months ............ debris problem so quickly? ........ besides, I use distilled water with additive Innovatek Pro and a spiral of silver in the loop

 

[4Ryan6]

I thought it was too short to adopt this solution you mention me
For the air bubbles , once I came back to fill it flush more than two hours with the top fillport open to bleed the loop, even shaking the pc chassis .
On the other hand, the loop had only running 3 or 4 months ............ debris problem so quickly? ........ besides, I use distilled water with additive Innovatek Pro and a spiral of silver in the loop

No it's not too short you use a nylon wire tie directly over the crimp and tighten it until the crimp comes out then it may take a couple of additional nylon wire ties on each side of the original crimp point to fully stabilize it, or even more if the crimp moves to a new location.

The wire ties will keep the tubing from collapsing and it's a much better option than going the full change out route.

Caution: I do not know if you have any Nickel plated water blocks in your loop, but if you do?, you're not supposed to use a Silver Kill Coil with Nickel Plated water blocks.

 

[dopax]

Thanks for the advice ...................I'll take into account as regards the silver kill coil .
I'll remake the whole loop and hopefully fix the problem