Sticking flow meter

conoflex · 25 Sep 2016 at 11:07

As above for some reason my flow meter has started sticking, only for a second. Also just before this started to happen the pump would not spin up straight away it took a couple of seconds.
I was told this might be due to under volting to the pump as I have connected to the fan pwm hub.
Any suggestions would be great,
Oh and I have flushed the whole system and the same.

MuddyPaws · 25 Sep 2016 at 13:15

conoflex said:
As above for some reason my flow meter has started sticking, only for a second. Also just before this started to happen the pump would not spin up straight away it took a couple of seconds.
I was told this might be due to under volting to the pump as I have connected to the fan pwm hub.
Any suggestions would be great,
Oh and I have flushed the whole system and the same.

take them out the loop and replace them.

conoflex · 25 Sep 2016 at 13:32

done that bought a new flow meter from overclockers, thinking might be the pump but is fairly new.

Cenedd · 25 Sep 2016 at 19:49

What pump? A DDC will start and run on 7V (at least, I can vouch for the 18W version) so it's got to be under voted by quite a bit before it'll cause problems. You can pick up a digital multimeter for around a tenner these days so you could rule it out quite cheaply. You could always plug it into a molex connector (with an adapter) to rule out the PWM hub too

Smffy · 26 Sep 2016 at 12:03

Can you post a pic of it in situ? I would think unless the pump is literally doing nothing it would register...

conoflex · 26 Sep 2016 at 12:14

It is an 18 watt DDC pump, got it all stripped down to test it.
(pump working fine)

Can't do pic in situ a mo as I say system all stripped.

Is there anything else that would cause it to stick intermittently ?
:confused:

Smffy · 26 Sep 2016 at 12:22

conoflex said:
It is an 18 watt DDC pump, got it all stripped down to test it.
(pump working fine)

Can't do pic in situ a mo as I say system all stripped.

Is there anything else that would cause it to stick intermittently ?

What model is the meter? Link?

Cenedd · 26 Sep 2016 at 13:01

A DDC is beefy enough (as is a D5 just in case anyone gets confrontational :rolleyes:

) that you can feel the vibration of it spinning if you put a hand on it. It's also got a tacho connection (like a fan) so a dedicated header will be able to report back the rpm. Both these should let you be sure whether the pump is stalling or whether the flow meter is sticking or misreporting. Smffy's right though, the model of flow meter would be useful as for example, Aqua Computer's High Flow pump is a spinning impeller that could, theoretically get jammed (especially if not flat) but their MPS flow meters are differential pressure so there isn't anything moving to get stuck - although it could get blocked potentially.

Glanza · 26 Sep 2016 at 13:12

Have you tried connecting your pump direct to power rather than through a fan controller, if yes does it do the same?

conoflex · 27 Sep 2016 at 11:25

Just running it in different combinations to what happens.

Flow meter is this one: https://www.overclockers.co.uk/primo-chill-vortex-flow-indicator-clear-red-wc-228-pm.html

Cenedd · 27 Sep 2016 at 11:38

Ok, if it's a flow indicator rather than a meter, it can't be misreporting - unless your eyes are faulty

It could be sticking but if you've already replaced it that would seem unlikely...although far from impossible. How long does it stop for? Does the pump keep buzzing during that time or can you feel it vibrating?

conoflex · 27 Sep 2016 at 11:44

Oh sorry about that listed as meter so that's what I put down

It's weird when took it to bits (pump flow indicator etc) couldn't find any bits or anything that would cause it to stick. Put all back together and it worked great for a few hours then it started hiccuping again.
It spins free then sort of stutters for a second, spins again and does it again.

Cenedd · 27 Sep 2016 at 12:20

I could be a batch with dodgy bearings. Does the pump show other evidence of working when the spinner stops? Specifically is it buzzing and vibrating the same as normal or does it go quiet?

conoflex · 27 Sep 2016 at 12:37

Not sure on that cos pump is in awkward place get at while trying to watch flow indicator

I will check it in a bit putting all back together now.

Thing is about dodgy bearings I got flow indicator from OC's and from WC uk so I think it's unlikely to be a bad batch. :mad:

Cenedd · 27 Sep 2016 at 12:40

Maybe extend the piping temporarily so the pump is more accessible?
You could re-plumb your res (if it has the ports) so that coolant returns at the top and visibly falls into the res. It's going to aerate it a bit but it'd give you definitive proof of whether there's any flow.

conoflex · 27 Sep 2016 at 12:46

Thank you very much Cenedd, I will try that first.
This is irritating as it can be. bloody computers

Smffy · 27 Sep 2016 at 18:58

conoflex said:
Thank you very much Cenedd, I will try that first.
This is irritating as it can be. bloody computers

You could also rig up some tubing and run it off a tap on low pressure for a bit and see if it sticks, another thing to check is the pump doesnt drop at any point - if you have any monitoring software that might be a good plan. Essentially its one of a few things:

1. Flow indicator is sticking, tap test should tell you that (low pressure just enough to get it to turn and rinse with distilled before replacing)

2. The pump is intermittently dropping power/pressure, possible causes:

Soft tubing with a kink in the loop
Motherboard PWM fan port is downclocking - check BIOS/software settings in case there are preset profiles which reduce pump speed based on CPU temp
Power supply loose or not delivering constantly power - check connections and try differenct socket

Thats all I can think of really mate...

Cenedd · 27 Sep 2016 at 19:36

Smffy said:
Thats all I can think of really mate...

You missed "Computer hates you!"

Good call on the kinky tubing too. And motherboard PWM headers are a mixed bag at best. Some of them are four-pin but only do voltage control (rather than PWM) on all except the CPU header - which Intel mandated by done properly. Voltage control should work on a DDC but if it drops too low it could stop - I know an 18W will go down at least to 7V. I used to have one on a toggle switch to switch it between 12V and 7V and when toggled it would stop for a few seconds - probably because it had effectively been switched off for half a second while the switch was between contacts. If the motherboard is doing horrible things with the voltage, it could do something similar....but I'm clutching at straws if I'm honest. Between tap testing the spinner (don't be on a water meter!) and waterfall testing the pump, you should be able to work out which one is giving you the problem and then work on diagnosing power/control issues to the pump if that ends up being the problem. Do check for kinky hose though....you don't want any of that in your build!

Smffy · 28 Sep 2016 at 00:08

Cenedd said:
You missed "Computer hates you!"

Haha, number 1 cause for problems with computer: No reason, just like this for a bit then fine also for no reason.

conoflex · 28 Sep 2016 at 10:53

Yer right "My Computer does hate me" :mad:

I never gave a kink in the tube a thought, will check them all whilst reassembling. Also going to rewire the pump to a molex plug and run the rpm lead to my Lamptron - FC5 controller so I can see there are ant changes.

Just like to say thanks for the input chaps, I would have still been scratching my head without it.