Case Mod Project Hydro-Li

[Nutman]

#79: Hm, strange indeed. If there's trapped air in there, it SHOULD work it way out eventually, but with all that tossing and turning you did, I doubt there's any air left. Is the fluid hazy (because of air bubbles in it)?

[Pot Of Jam]

I have had about 6 different loops in the past, and the last one I did had the same problem.

This was using a XSPC bay top for a DDC pump and i tried for hours to get the air out. I couldnt find out the problem (removed the loop and replaced with air as I had a LAN party next day).

but the previoous loop I had the liquid gravity fed directly in to the top port of the pump from an EK spinner Res, this loop had no problem with air bubbles.

Now to be honest I think it was down to the res choice as it releases the air bubbles out easier.

Here are pics of the set up...




Hope this helps....

[Ripitup121]

Originally Posted by The boy 4rm oz

From my experience I have found that the temps do stay cooler a little bit longer. Because the res isn't collecting hot fluid and feeds relatively cool fluid direct to the blocks you should find that the fluid in general will stay a little bit cooler for longer.

I originally used this method in my lexa mod before changing to res-pump-240mm rad- cpu-gpu-120mm rad-res.

Will try this tomorrow and report back. cheers

Originally Posted by Nutman

#79: Hm, strange indeed. If there's trapped air in there, it SHOULD work it way out eventually, but with all that tossing and turning you did, I doubt there's any air left. Is the fluid hazy (because of air bubbles in it)?

I thought that too but turns out not to be the case. the fluid is fine, no hase!!!

Originally Posted by Pot Of Jam

I have had about 6 different loops in the past, and the last one I did had the same problem.

This was using a XSPC bay top for a DDC pump and i tried for hours to get the air out. I couldnt find out the problem (removed the loop and replaced with air as I had a LAN party next day).

but the previoous loop I had the liquid gravity fed directly in to the top port of the pump from an EK spinner Res, this loop had no problem with air bubbles.

Now to be honest I think it was down to the res choice as it releases the air bubbles out easier.

Here are pics of the set up...
Hope this helps....

Thanks Jam, I have the Res at the top of the loop, the odd thing was i didnt have this problem the last time i filled it 3 weeks ago... Thanks for pics


Will be working on this tomorrow and give you all an update.... fingers crossed it works with all my newfound information thanks to you guys.

Cheers
Rip

[Ripitup121]

Working on suggested solutions at the moment guys. Just wondering if the flow route through my blocks is ok. suggestions on changes are very welcome.



Have Eliminated any Pump Fault.
Have Eliminated any radiator fault.

Its gotta be the waterblocks now....

Gonna open all the blocks again to be 100% sure they are not blocked just incase.

Cheers
Rip

[The boy 4rm oz]

I know those mosphet blocks and northbridge blocks are serious flow killers, that may be contributing.

I would change the path to cpu-top mosphet-left mosphet-northbridge. The current set up may be having trouble pumping through the restrictive blocks. Going straight from the cpu to the top block will allow gravity to help feed fluid into the lower blocks and may fix your problem.

[23RO_UK]

Are you running a DDC or D5?

[mars-bar-man]

I believe he's running a DDC, so there's pleanty of head pressure there, I think it could well be what the boy from oz said. Have a look at re-arranging the loop, it may help.

[Ripitup121]

Originally Posted by 23RO_UK

Are you running a DDC or D5?

Yep DDC...

Originally Posted by mars-bar-man

I believe he's running a DDC, so there's pleanty of head pressure there, I think it could well be what the boy from oz said. Have a look at re-arranging the loop, it may help.

Did that, still the same although rad filling faster now and am using less tubing so will keep this setup, Still testing each block seperately now do.

Originally Posted by The boy 4rm oz

I know those mosphet blocks and northbridge blocks are serious flow killers, that may be contributing.

I would change the path to cpu-top mosphet-left mosphet-northbridge. The current set up may be having trouble pumping through the restrictive blocks. Going straight from the cpu to the top block will allow gravity to help feed fluid into the lower blocks and may fix your problem.

Sounds like a plan. I took your advice on Rad before res and is deffo better but didnt solve the problem. (+rep) I cant go from CPU to top Mosfet do (too tight a bend for tubing) unless i buy 2 90 degree elbows. (not gonna help flow either do )

The kicker is i have had this system running perfect 3 weeks ago as it is...

Thanks for all the help guys... Keep em coming.
Will post later on today on my finding on seperate block testing.

Cheers
Rip

[mars-bar-man]

Tbh man, I'd go grab some 90* fittings, I've used one in my loop and it's fine, the DDC has enough pressure.

Actually, I've just looked at your loop, they're huge. It could be a case that the pump doesn't have enough power to pump the fluid all the way round the loop. Maybe.

[Ripitup121]

Originally Posted by mars-bar-man

Tbh man, I'd go grab some 90* fittings, I've used one in my loop and it's fine, the DDC has enough pressure.

Actually, I've just looked at your loop, they're huge. It could be a case that the pump doesn't have enough power to pump the fluid all the way round the loop. Maybe.

True and if it was the first time i would be looking at that too but i know it works... had it running perfect 3 weeks ago ????????

The 90 degree fittings might be the next option.
Thanks mate

Rip

[mars-bar-man]

Originally Posted by Ripitup121

True and if it was the first time i would be looking at that too but i know it works... had it running perfect 3 weeks ago ????????

The 90 degree fittings might be the next option.
Thanks mate

Rip

Np, it's what we're here for! As for it running perfect 3 weeks ago? Remember, computers are just like women, temperamental beings, it's probably hit its time of the month.

[Ripitup121]

Originally Posted by mars-bar-man

Np, it's what we're here for! As for it running perfect 3 weeks ago? Remember, computers are just like women, temperamental beings, it's probably hit its time of the month.

Lol...And a new sig is born... (+rep)

Cheers
Rip

[The boy 4rm oz]

I'd give the 90degree fittings a try anyway. Because the blocks slow down the flow rate so much the fittings will hardly make a difference.

Thanks for the +re pas well mate, glad I could help.

[Ripitup121]

Hay Guys,

First thanks to all who helped troubleshoot this with me. its very much appriciated. The vast amount of knowlage from you all makes for good company i have to say.

I have found the problem part. The CPU BLOCK. Its a Swiftech Apogee TX. only a lil bit of coolant is coming through which the pump only a foot away going full pelt. very odd really given you would think it only has a very simple job. rather than opening it up and messing around anymore i am getting an RMA set up with Scan and it replaced under warrenty. More time wasted but its the best option. Least i dont have a 3 grand system standing in my living room looking sorry for itself anymore....

Thanks Again guys,

More pics of final build as soon as the block comes back...

Woot.

Cheers
Rip

[bulldogjeff]

You've found the problem that's the main thing, I know it slows things down, but that's all part of the fun.Well that's what they say, but personally I reckon it's a load of bollocks.

[Ripitup121]

Originally Posted by bulldogjeff

You've found the problem that's the main thing, I know it slows things down, but that's all part of the fun.Well that's what they say, but personally I reckon it's a load of bollocks.

LOL, yeah it defo is Jeff... i was going mad. in the last 5 days i filled the loop 14 times in testing etc... Have to say do the reward feels nice and fuzzy.

Rip

[The boy 4rm oz]

I'm glad you found the problem mate. Don't the GTZ's have a flow plate in the upper block to disperse the water flow? That's probably what the problem is.

[Nutman]

Originally Posted by The boy 4rm oz

I'm glad you found the problem mate. Don't the GTZ's have a flow plate in the upper block to disperse the water flow? That's probably what the problem is.

Nope, doesn't look so:

http://www.swiftnets.com/assets/imag...x-exploded.png

My guess is that some gunk got stuck down there. Or perhaps something was dropped down there without him noticing before putting the loop together.

In any case, to RMA it seems a tad exaggerated, as this block comes apart pretty easy for inspection.

[Ripitup121]

Originally Posted by The boy 4rm oz

I'm glad you found the problem mate. Don't the GTZ's have a flow plate in the upper block to disperse the water flow? That's probably what the problem is.

Nope just a inlet plate that can be moved to allow for compression fittings etc. Its already off to scan anyways.

Originally Posted by Nutman

Nope, doesn't look so:

http://www.swiftnets.com/assets/imag...x-exploded.png

My guess is that some gunk got stuck down there. Or perhaps something was dropped down there without him noticing before putting the loop together.

In any case, to RMA it seems a tad exaggerated, as this block comes apart pretty easy for inspection.

Already took it apart and cleaned it, and it all looked grand, so at this stage i just wanna get a new one. if this had happened after 6 months of use then i would agree and go ok open it up again and inspect, but ofter having it running for 2 days and then this 3 weeks later.. Not a chance.

Cheers
Rip

[The boy 4rm oz]

Swiftech tend to make really good stuff and their warranty is pretty lenient. I had a Swiftech Apogee GTX and the alu top started to corrode after a few months. They sent me a Delrin replacement free of charge.