PDA

View Full Version : aquabay/aquatube


Greeny
22nd February 2008, 15:23
Allright wizards this one has me a bit perplexed, I'm sure I've done something wrong though.

I have the 5+1/4" bay with the aquatube mounted in it, I have the outlet on the side of the aquatube (in the hole furthest from the plexi) with a 90 degree elbow joint connecting it down to my inlet on my pump which has another 90 degree elbow joint on the front of it. I have the aquajet metal tubey thing in the rez for the return flow.

No other elbow joints are used, I have a dual 80mm radiator and a dual 120mm radiator a CPU block and a GPU block. Pump is mounted at lowest point in loop, flow goes rez -> pump -> 80mm rad -> cpu block -> gpu block -> 120mm rad -> rez.

Problem is sometimes the water flow completely stops and I'm not sure whether this is a failure with the pump or whether I have inadvertently misconfigured the loop? Filling the thing was an absolute nightmare as the pump kept sucking air mostly, I did use a lot of fresh tubing though and the gpu block/dual 120mm rad are new.

I have one of those new fancy aquastream pumps with the USB controller - it may or may not be faulty.

I can take some pictures of the setup if it's hard to visualise.

Darv
22nd February 2008, 15:26
I can't imagine anything in your loop that would cause the water flow to stop. Sounds like the pump may be cutting out. :confused:

Greeny
22nd February 2008, 15:45
It's weird like, to the best of my knowledge the pump status light does not indicate an error (I'll have to recheck next time it cuts out), certainly I am continuing to receive a tacho signal from the unit to the motherboard header fan. I can see no errors from the aquasuite software. The pump appears to still be going as well judging by the vibrations from it.

I did see the flow cut a few times when I was leak testing the system, but I had thought that their might still be some air in the system at the time. I've seen the water flow cut 3 times now though, GPU temp hit over 80 last night, CPU temp doesn't seem to be a problem as it's pretty idle. I got a BSOD a couple of nights ago which I debugged and it was an error with the Nvidia driver, the thing probably overheated.

Turning the PC off then back on is enough to get the flow going again.

I am just wondering whether I should keep the loop configuration and switch back to the old pump or whether I should be thinking about reconfiguring the rez?

I can hear strange sloshing noises from the pump occassionally.

fivecheebs
22nd February 2008, 16:39
Sounds like you might have a blockage somewhere to me. If the pump tacho is still reporting a signal, then the impeller must still be spinning, unless of course it gets the tacho reading from elsewhere, but that would be daft.

Did you flush your components when you replaced the pump greenster?

Pug
22nd February 2008, 16:51
I have the outlet on the side of the aquatube
I think I see your problem...

It's been a while since I used the AquaBay AT (I'm using the CF one atm) but do you have it oriented so that the ports are lower than the inlet or level with it?
If they're level, rotate it so they're lower and that'll hopefully sort it.

Pics will probably help. Attach them if hosting's a prob. :thumb:

slame
24th February 2008, 11:19
Hi,

as soon as a water pump gets air it will stop to pump. No water pump can suck air ! The aquastream may stop because of that.

cheers

slame

Greeny
24th February 2008, 17:16
@ Pug, yes the aquatube is oriented so that the side holes are pointing diagonally downwards. I'll get some piccies.

@ Slame, you might be on to something their... I'm thinking that possibly some turbulence may be causing some air to go down the inlet or interfering with the inlet anyways. I'll have a think and see if I can find something to put in the reservoir to make sure the stream gets diffused a bit..

Pug
24th February 2008, 17:22
Do the newer AquaStreams still come with that blue gravel filter, like the Eheim it's derived from?
If you have that, try wedging it down against the base of the 'tube.
It's been known to cure turbulence issues in the past. :thumb:

Greeny
24th February 2008, 17:36
In glorious technicolour...
http://www.clansim.co.uk/downloads/pics/bayrez.jpg



http://www.clansim.co.uk/downloads/pics/bayrez_back.jpg



http://www.clansim.co.uk/downloads/pics/loops_of_fury.jpg

Greeny
24th February 2008, 17:44
No blue filter that I could see Pug, that would''ve been inside the silver filter holder at the front of the pump aye?

I'm thinking about bending those tabs on the sides of the 5+1/4" bay to lift the rez up a couple of bays, I think theirs enough slack and it should help the downpipe sit down a bit more. I planned it near the top anyway.

Pug
24th February 2008, 18:12
Unless you can re-route that hose slightly, it might be worth shortening it by an inch or so. It seems to be angling up slightly which might be working against you.

Otherwise, what you have in mind may well sort it (together with adding a little more coolant if you can manage it).

Greeny
24th February 2008, 18:33
Funnilly enough I was thinking that hose could do with being an inch shorter myself after taking the pictures. I think Ill go with plan A to start with and see how I get on.

Cheers to all for the input.:toast:

Pug
24th February 2008, 18:49
Actually, before you do, just try losing that ziptie that's holding it to the other tube and see if that helps, to start with.


Nice rig, btw. :thumb:

Greeny
25th February 2008, 12:50
The ziptie is actually just securing the power wire from LED on the GPU waterblock to the tube, those tubes are both freemoving. ;)

I went for minimizing the turbulence after I had bashed the stupid 5+1/4" bay drive tabs for about 5 minutes with a hammer and got nowhere, I'll have to dissassemble the whole thing and cut them off.

Got a plastic scouring pad from the kitchen which had loose cell sponge on one side and chopped that up and inserted it in the rez. I think that *might* have solved the problem, the pump sloshing noises appear to have disappeared and I haven't seen the flow stop yet (cross fingers).

Cheers for the compliment. The rig is a wee bit messy like, I had all the wiring installed like a total pro and then I went and broke the PSU by switching it on and off rather rapidly a couple of times while I was priming the loop (£££ new PSU). The IDE cable was a single device jobber, but of course the motherboard threw a hissy fit and wouldn't pick the drive up properly so I threw in the ribbon cable quickly in disgust. Oh and the internal USB card didn't work either so I had to poke the cable out the back. The motherboard took a fair bit of tuning as well. Overall this machine has been a pain in the arse so far. lol.

Greeny
14th April 2008, 14:00
Last weekend i gutted the PC and chopped the 5+1/4" bay tabs so I can lift the rez. Re-assembled it all and the pump was still struggling.

This weekend I replaced the pump with the old pump and it looks to be going well so far (priming it was still a bitch though), removed the sponge as well. I think the new pump is faulty.

Replaced the evil new motherboard from hell (Asus P-5E WS) with the same one I had before (Asus P-5WDG2 WS Pro) so at least I have a usable system now.

4 days work in total with the flu was not fun... Back to internet poker bankroll accumulation.

Pug
14th April 2008, 14:56
Have you tried testing the pump independent of the loop now that you have it removed?

Greeny
14th April 2008, 17:02
I'm not quite sure what to test for? I can set it pumping and see if it goes off, but if the conditions of the problem are related to the loop setup then I might not see anything. I'll try the thing out once I get some time.

Pug
14th April 2008, 17:14
Well, I guess the optimal test would be to fashion a loop that includes a couple of elbows on the inlet, siphon prime it, set it running and see if you can observe anything abnormal about the resultant flow from that. :shrug:
If it is the pump at fault, I would hope it would become immediately apparent. :thumb:

Greeny
15th April 2008, 00:26
I'll ave to rake oot some fittings first.

Here is a picture of the offending item, I took the top off. Impeller looks intact. I had the other one apart as well, but I did not notice if the impellar moved freely, this one has some play in it. The rubber button was on top of the impellar of the old one, its in the top cap on this pump. No significant detritus inside it. The screwmarks for the outlet barb are quite deep into the unit I wonder if the impellar might have been catching on the outlet.

http://www.clansim.co.uk/downloads/pics/pump.jpg