D5 Not good enough?

[barfastic]

Hey guys,

Have a question.

I re did my loop to get a dedicated cpu loop going.

it consists of a Alphacool D5, Koolance Top, Feser Quad Rad, EK Supreme HF, Koolance Res Threaded onto the Top.

My question is, Shouldnt the pump be strong enough to push bubbles through? I mean, my flowmeter is barely spinning at setting 5.

Now, on my second loop, i have 2 D5, EK top, triple rad, EK FC GPU block, EK Full Cover board, EK RAM block, and the flow is ridiculous. (there is turbulance in the res)

What i noticed, before i seperated my loop, is that even with the dual d5, the original single loop had flow issues, my cpu block is clean as a whistle. But there is NO movement in the res.

check out the videos....

GPU/MOBO/RAM LOOP
http://www.youtube.com/watch?v=z_9rbtCcPKU

CPU LOOP
http://www.youtube.com/watch?v=OPZWEIGatc8

The order is RES/PUMP-> Radiator-> CPU ->RES/PUMP... i only have 2 90's, and as a reference i have the EK insert with the 2 slits.

any ideas?

[ownage]

You probably have allot of air in the loop. I had that before. My loop was to complicated and air had to move up, down, up, down and again up. Now I changed some things and my D5 pumps like mad. The water still has to go up and down 3 times but without air in the loop its so much stronger. A D5 cannot push allot of air down because of low pressure caused by the air. It can take months to get the air out. Just flip the pc all directions to get the air where it should be.
D5 is more than strong enough for cpu only, so its not the pump, its the conditions.

edit

[Church]

I suspect that in original single loop with lone HF in it there was some user error. D5 should be MORE then upto job (imho higher flow then with dual D5 in 2nd loop with motherboard and gpu full covers and ram block thrown extra). Most probable seems some kinked tubing. Over time people have experienced different problems that might cause flow blockage. Starting with gunked up WB (you said not the case with block being clean), mismatched pump inlet and outlet, not full 12V fed to pump from wrongly DIY sleeved pump wire, in few cases fittings with overly long threaded part screwed in too deep partially blocking flow, not fully bled loop with air bubbles in components and few other possibilities.

[Utnorris]

Air in your rad will cause some flow issues. I had a similar issue a while back, turned out my block was gunked up, but also had some air left over in the rad which caused some weird noise and flow issues. If you can shake it, try that or tip it back and forth. A D5 on just a CPU block should give you over 1Gpm easily. I was running cpu, two gpu's and a MB and was getting .56 GPM, so you should get 1 GPM easily. By the way, I swapped out my dual D5 for dual MCP355's and saw an increase in flow. Also, getting higher flow didn't help my temps, but it did make me feel better.

[barfastic]

You probably have allot of air in the loop. I had that before. My loop was to complicated and air had to move up, down, up, down and again up. Now I changed some things and my D5 pumps like mad. The water still has to go up and down 3 times but without air in the loop its so much stronger. A D5 cannot push allot of air down because of low pressure caused by the air. It can take months to get the air out. Just flip the pc all directions to get the air where it should be.
D5 is more than strong enough for cpu only, so its not the pump, its the conditions.

edit

I suspect that in original single loop with lone HF in it there was some user error. D5 should be MORE then upto job (imho higher flow then with dual D5 in 2nd loop with motherboard and gpu full covers and ram block thrown extra). Most probable seems some kinked tubing. Over time people have experienced different problems that might cause flow blockage. Starting with gunked up WB (you said not the case with block being clean), mismatched pump inlet and outlet, not full 12V fed to pump from wrongly DIY sleeved pump wire, in few cases fittings with overly long threaded part screwed in too deep partially blocking flow, not fully bled loop with air bubbles in components and few other possibilities.

Air in your rad will cause some flow issues. I had a similar issue a while back, turned out my block was gunked up, but also had some air left over in the rad which caused some weird noise and flow issues. If you can shake it, try that or tip it back and forth. A D5 on just a CPU block should give you over 1Gpm easily. I was running cpu, two gpu's and a MB and was getting .56 GPM, so you should get 1 GPM easily. By the way, I swapped out my dual D5 for dual MCP355's and saw an increase in flow. Also, getting higher flow didn't help my temps, but it did make me feel better.

There is definately air left in the loop, i guess ill let it bleed some more, and try shaking it about in all directions. I jsut thought that the pump would be powerful enough to push the air out eventually...

I dont think its a PSU problem (Its a Thermal Take 1200W Toughpower)... is there any way to test to see if its giving out the right voltages? (except using a multimeter thingy?)

[Church]

barfastic: under wrong voltages i meant not higher or lower output of PSU (I doubt it being more or less then +/- 1V, not affecting pump's performance by THAT much to care/measure), but rather mistakenly used pins in molex for wires, when during sleeving process by themselves wrongly connected them in connector getting kind of unwished mod of 7V (one can get from standard 4-pin molex 12V/7V/5V depending on connection scheme).

[Martinm210]

The Koolance top/res setups won't see much moveing in the reservoir because the reservoir technically acts more like a large Tline than a reservoir where water is dumping into it and leaving. The top itself and elbow inside serves as the baffle system. This works great in preventing vortex issues, but it does eat up some pumping power and the reservoir water is more still than your typical reservoir.

As long as the system is bleeding, I wouldn't worry about it.

[barfastic]

my system wasnt bleeding... im currently draining.... this build has been a nightmare and i SWEAR its been jinxed... im gonna put it aside for a while. i could of finished 2 other builds i have in the making by now.... :/

thanks for your advice, and thanks for the tips. Freshly Squeezed, is going on pause.

[NaeKuh]

your D5 cant purge!!!

u need to tilt your case from the back so the barbs point down this way \

Right now the D5 is cavitating hence why u hear it make a super loud noise.. and cant purge.

Turn off the pump... tilt case.. and then turn back on, and you'll see a BIG air bubble get pushed.
Then once u hear it cavitating again, turn it off.... repeat.

Make sure your D5 is at the highest setting also until u have fully bleed.



Barf... dont quit now.. your at the VERY END, and just have bleeding problems.
Dont worry, 75% of the people i know and have helped, which is a LOT always have problems at the bleeding stage, when they mount there rads exactly like you did, so your in a tough spot from the getgo.

Just follow my directions on tilting, and cycling the pump on and off.
Cycling the pump causes the pump to shoot more head pressure at start due to pump start voltage being higher.

You just need to cycle the pump while the case is tilt.. ie.. something big in the front of your case.
Your case should be like / while your barbs are pointed \


Give bleeding 1 more shot, and then retire... but do it after u had a beer, or eatten a pizza.
But as i said your at the VERY END and u just got a NASTY air pocket stuck in the highest part of your loop... ie. your radiator.

[Martinm210]

I didn't notice the air bubble the first time skimming.

Not sure about the pump purge thought (I couldn't quite make sense of the background noise), but inverted rads...especially ones with large boxy plenum chambers are notorious for being impossible to bleed with anything but extreme flow rates. Small tapered type plenum chambers work best in maintaining adequate velocity to self purge.

Most rad designs out there ignore this velocity problem at the plenum and simply make a square box to serve in collecting the tube flows thinking bigger is better. Unfortunately, that leaves you with very low velocities coming out of the radiator tubes and a resulting air trap. About the only way to get around that is to either tip the case, or allow room in tubing to tip the radiator around until it's all free of air.

[Dreddy71]

a lot of times when im bleeding my loops i will spend about 10-30 minutes running the pump till it sounds like its at full speed, shutting it off and repeating till i dont see bubbles in the lines then i switch it on till i hear the pump taking n bubbles and repeat till it goes about an 15min with no sound. then i get too lazy to top of the rez when it gets the rest of the air out a week later.

[barfastic]

ok, so i didnt take it apart... i did drain it though...

when i refilled the res, and started it up, the flow meter was spinning like crazy, after getting it almst full, it slowed down. ive tilted it, cycled it, tilted in ALL directions, shook it, shook it violently, and still its not performing.

after its been tilted and cycled, no more bubbles come out... heress the vid to show you.
http://www.youtube.com/watch?v=8gtpYaQuEqU

[Waterlogged]

The loop won't bleed the way you have it set up. You need one of these (link) to replace the black ring that mounts the res to the top and plumb your return into it. Make sure you get a 60mm version so it works with the top and res body because they also have 80mm version as well.

[barfastic]

The loop won't bleed the way you have it set up. You need one of these (link) to replace the black ring that mounts the res to the top and plumb your return into it. Make sure you get a 60mm version so it works with the top and res body because they also have 80mm version as well.

this is the exact way they have it setup aswell... :/

[Waterlogged]

this is the exact way they have it setup aswell... :/

The DDC tops are designed the same way. . .having 2 of those, I can tell you from first hand testing. . .they just don't bleed very well, if at all, set up like that. Get that collar, it'll solve your problem in a heartbeat.

[barfastic]

bleh, arent available anywhere in europe... time for storage

[Waterlogged]

Coolbits.dk shows in stock. (link) The "scumbag" shop has a shipment coming in about 2 weeks and Highflow lists it but it looks to be OOS ATM.

[Utnorris]

The DDC tops are designed the same way. . .having 2 of those, I can tell you from first hand testing. . .they just don't bleed very well, if at all, set up like that. Get that collar, it'll solve your problem in a heartbeat.

Can confirm this, got to have that collar.

[barfastic]

lol... the scumbag shop? which one is that?

coolbits.dk i wasnt aware of. nice, thanks.

The whole system is drained now. I changed my focus a bit to my muffin build. so this gonna be on the slow steamer for a bit now. but thanks!