Airlock in pump

JJimmyJ

JJimmyJ

JJimmyJ

Associate
Joined
21 Jan 2010
Posts
568
Hello

Ok, I've completely rejigged my water set-up, new tubing, radiator etc etc.
Turned it all on, but no water is pumping. Took the pump out, removed the top etc and gave it a quick dry whirl. It works fine.
How do I get water into the pump itself? It seems to be constantly airlocked.
I've tried tipping etc, but it just keeps running dry.
Should I leave one tube free (over a basin) and run it so that the air has room to escape and draws water into the pump, then reconnect it all? I'm just worried that I'll be damaging my pump with all the dry running.

Thanks
 
Hello

Thanks for replys.
Ok, powering on/off is hard as it means I'm constantly turning comp on/off.
Loop is indeed Res>pump>rad>cpu.
It's at the same level as the res. I've tried tilting, but doesn't seem to make a difference. Thought the pump had died originally, but once I saw it running at full whack with the hood off I figured it had to be something else.
I may just try pouring water directly into the pump manually to see if it pushes through, then somehow try and hook it back up with the water in it??? No idea how.
 
For turning pump on and off why not just jump the psu? Unplug everything except the pump and on the 24 pin cable connect the green wire to any of the blacks. Personally i use and old piece if wire but as soon as connection is made pump will run then unplug and it stops. No needto boot
pc for that.

As for res same level as the pump, i presume the water level is higher than the pump meaning the pump is still being gravity fed? In which case pulsing the pump or leaving it running for a bit has removed all the air locks from my pump tops.
 
Greboth said:
For turning pump on and off why not just jump the psu? Unplug everything except the pump and on the 24 pin cable connect the green wire to any of the blacks. Personally i use and old piece if wire but as soon as connection is made pump will run then unplug and it stops. No needto boot
pc for that.

As for res same level as the pump, i presume the water level is higher than the pump meaning the pump is still being gravity fed? In which case pulsing the pump or leaving it running for a bit has removed all the air locks from my pump tops.
Click to expand...

Ha, a bit of hotwiring. Not sure the mrs would approve knowing my past record with electrics :)
Water level is higher than pump.
How long should it take for the pump to clear? Are we talking 10-20 mins or should it clear up within just a few mins?
 
it should take seconds if the pump is gettin primed from the water head the more head the easier there may well be an air lock elsewhere in the system stopping returns tho,

if you run a pump dry yes it can damage it but for short bursts it shouldnt be a problem

jumping the psu can be done to test as in

but if you a bit concerned then dont tbh
 
I haven't put it in my case yet. Got everything external so I could run it for a few hours to ensure no leaks.
I think it's deffo got an airlock. I'll try pumping it through to a basin so the air has somewhere to go, that should hopefully clear the block in the pump.
I did run the pump dry
 
Have you made sure you've connected the res to the inlet of the pump, and not the outlet?

edit: you should always leak test without the hardware powered, inside the case. You can buy an ocuk psu bridging kit for a few quid, just plugs into your 24 pin.

alternatively you can buy the 90w version of the phobya external 4 pin molex psu and use it for leak testing. Ocuk don't sell this.

running the pump dry knackers it out pretty quickly, btw.
 
Last edited:
Yeah it's all connected correctly.
Fortunately I still have my fan, so I can run everything outside the case as a test. I'm connecting it using a 4 pin molex. Is it safe enough to just connect and disconnect whilst the PC is running?
I need to properly sit down and have a good go at it. I was kind of rushed yesterday. Never thought it would be so hard to get some water into a pump. Usually ends up everywhere when I'm refilling my reservoir etc.
 
Why do you keep asking if connecting and disconnecting a pump to a power source from a turned on PSU with other hardware already plugged in is safe or not?

It may work out to be fine but do you really have to take that risk for the sake of simple DIY and identifying 2 wires from the 24pin ATX cable? or better yet getting a jump start plug for £1.50 or less?

And to top it off you say you rushed in to things!

If you want to keep your hardware for sure, then please take some time and go on the longer but safer road!

EDIT:
Why don't you upload an image of your current setup - specially with the RES and Pump setup - to see whats wrong!
 
Avathar77 said:
Why do you keep asking if connecting and disconnecting a pump to a power source from a turned on PSU with other hardware already plugged in is safe or not?

It may work out to be fine but do you really have to take that risk for the sake of simple DIY and identifying 2 wires from the 24pin ATX cable? or better yet getting a jump start plug for £1.50 or less?

And to top it off you say you rushed in to things!

If you want to keep your hardware for sure, then please take some time and go on the longer but safer road!

EDIT:
Why don't you upload an image of your current setup - specially with the RES and Pump setup - to see whats wrong!
Click to expand...

Blimey! Calm down :)

I should have said I was a little pressed for time once I'd put it all together so never really had time to try lots of different things. That's why I left it for the day. No way I'm rushing it and destroying my system.
I'll try and post a pic when I get home.
As far as the jump start plug, I'd never heard of one until now.
Looking at craarcs post though I should be able to have a go at that with an old psu I have lying around (if it still works). Always wondered how people tested pumps etc without powering everything on.
 
From your earlier posts I(and most probably others have as well) assumed you know the basics like leak testing with out connecting the other hardware to the same PSU, PSU jump start and flushing the RADs maybe :)
I think you have missed a very important step
RESEARCH!

Anyway, we are diverging now!
So lets get back to your problem, post a pic with your setup and try to read about PSU jump start(it is not hard to find in google)
Disconnect ALL other connectors from PSU apart from Pump power when you test the loop for leaks.
Make sure you have plenty of paper towels available, and cover the connectors and surrounding ares to see if there is a leak.
 
sounds like u done something wrong cos u shouldn't get any airlocks imo , soon as u poor liquid into the res it should flow straight down into the pump before even turning on the pump assuming u got the pump lower than the res.

if u got a 2nd pair of hands avalible to make it easier , have them top up the res slowly as u doing the psu and keep topping up as the res as it runs low then there's no reason to let the pump run dry, just use a paper clip or something to jump the psu unless u want to wait afew days and order a jumpstart block

as ava says do your research and don't rush
 
Last edited:
I bet the res side that falls into the path of the pump is the wrong way.
The pump will only pump in one direction so the water in should be connected to the res. fill the res up to the top and tilt the case so water falls towards the pump. That will work no doubt about it
 
one thing just cause you have water leaving the pump and no water dropping in the res doesnt mean to say there isnt an air lock else where

so in effect you wont be getting full flow its not just making sure you have no ait in the pump if u have an airlock else where untill you get that air to migrate you wont get full flow

your pump may well be full of water and still have a trap somewhere else that will reduce your flow rate
 
You must log in or register to reply here.
Back
Top Bottom