Vortex in Reservoir

[ICPete]

I'm currently building up a system with a loop of 1/2" OD acrylic tubing, with dual D5 pumps and a PrimoChill CTR. The latter has its input via one of the upper side ports, while the output is the central bottom port. After initial filling I'm getting a substantial vortex that sends air down the center of the vertical reservoir and on to the pump.

I haven't gotten all the air out of the system yet, as it gets trapped in the pumps after collecting at the top of the reservoir and then getting drawn in to the pumps again. This video shows the action with only one of the two pumps operating (the left one):

Spoiler: Video of vortex

https://i.imgur.com/ujHL4FH.mp4

These are PWM pumps, so they are running at full 12V speed from the power supply. I don't want to power up the motherboard until I've finished leak testing and running-in the cooling loop. Much of the advice I've read on this and other forums suggests that running the pump at reduced speed should help. But I'd like to eliminate the vortex even at maximum speed of the pump.

PrimoChill sell a Vortex Killer piece, but only their pump-enabled CTRs have the tapped holes for mounting it. So I designed my own, after studying the topic online. This is about 120mm tall and fits loosely within the acrylic tube of the reservoir, except the six "finger" loops that are supposed to be springs to center it and make sure it doesn't rotate. I've placed an order to get this part 3D printed in ABS/ASA via FDM. It's more complex than it probably needs to be, but that doesn't impact the cost much, and I wanted to make sure it will do the job. BTW, that central hole is blind (a dead end) at the bottom.

Spoiler: 3d model of vortex spoiler

It should arrive within 10 days and I'll post an "after" video after I install and test it.

 

[ICPete]

Both of you exactly right. I put those thin "hairclip" looking features there to contact the bore of the reservoir. I measured it carefully and it's not perfectly round (after all it's cast acrylic). The ID varies a bit, so anything designed to contact the ID would need to be "springy". And absolutely I won't know whether there is enough tension to provide the friction required to keep it from spinning, until I try it.
Color would have been nice, but I selected ABS or ASA, whichever runs next, and whichever color runs next. Looking back, I could have selected red ABS for just $6 more, matching some of my colored bits.
It will probably arrive in black ABS, as that's probably the most popular color.

 

[ICPete]

Here is a photo of the 3D-printed vortex breaker pushed into a short piece of CTR tubing (left over when I cut 70mm off the length of my CTR):

Spoiler: Vortex Breaker in piece of CTR tube

Here are two more views of it:

Spoiler: Bottom view of vortex breaker

Spoiler: Top view of vortex breaker

Finally here are a couple of videos of the water & bubbles running through the system. In the second video the air has finally dissipated from the right-hand pump. and all the gurgling noise has stopped:
https://www.dropbox.com/s/e1jxu9bv8kavgwm/Vortex Breaker in Operation.mp4?dl=0
https://www.dropbox.com/s/ouw42b3vu5ybtaw/Dual D5 Pumps Running Silently.mp4?dl=0
[I hope those links work...]

You'll notice the other change I made was running the input line to the reservoir into one of the bottom fittings, instead of the top. That may have also helped eliminate the vortex. By rights I should do an experiment now without the vortex breaker in place, and see if it was needed now or not. However, 1) I don't want to spend the time to do that; and 2) I don't want to find out that I wasted the cost of getting this part made!

 

[ICPete]

Thanks mcnumpty.