• 04.12.2021, 14:36
  • Register
  • Login
  • You are not logged in.

 

Dear visitor, welcome to Aqua Computer Forum. If this is your first visit here, please read the Help. It explains how this page works. You must be registered before you can use all the page's features. Please use the registration form, to register here or read more information about the registration process. If you are already registered, please login here.

AwkaBear

Junior Member

[Need Help] Leakshield cannot draw vacuum, Sees fill level as 0%, No leaks present in loop after two days

Wednesday, October 27th 2021, 7:44am

I have just built a custom loop with a with a brand new D5 Next Ultitube 150 leakshield and using the leak test feature on the leakshield shows pressure dropping rapidly as the pump fights to draw a vacuum.

The leakshield also always yields 0% fill level whenever it tries to do a fill level measurement despite the reservoir being clearly ~40% full.


Picture of Build:
https://media.discordapp.net/attachments…=732&height=549



Below is a video of the test, this was performed with and without coolant in the loop and the result is the same:

https://youtu.be/y92YjXkXxHQ




Here is what the pressure graph output looks like:

https://cdn.discordapp.com/attachments/7…44/IMG_6277.jpg




I leaktested the build for over 7 hours during a distilled water flush and another 24 hours with my coolant.

I have also unscrewed and rescrewed down the leakshield several times to ensure it is properly sealed onto the reservoir.

I figured pressure is dropping this rapidly due to a leak there would be a noticeable amount of liquid in a given area of the case but everything is bone dry.

This leads me to believe that the leakshield is defective and cannot build pressure due to a faulty vacuum pump, stuck open vent valve, etc.

I want to be completely sure that the issue is isolated to the leakshield before I RMA it but I think of anything else to rule out. Any input would be appreciated.

Has anyone else had a similar issue to this?

Attached I also have the pressure graphs for it during an attempt to activate shield mode and to do a fill measurement
AwkaBear has attached the following images:
  • Fill level measurement graph.png
  • Enabling Shield mode graph.png

Wednesday, October 27th 2021, 9:23am

To rule out a fault in the LEAKSHIELD, take the AGB completely out of the circuit and simply connect the inlet and outlet with a hose.
Then do the leak test without water.
If this works, you have a leak in the system.
Always remember that there is a huge difference between watertight and airtight.

If the leak is so big that the LEAKSHIELD cannot build up pressure, you should see air bubbles in the system.

Translated with www.DeepL.com/Translator (free version)
Es gibt keinen Ausweg, den ein Mensch nicht beschreitet, um die tatsächliche Arbeit des Denkens zu vermeiden.
Thomas Alva Edison (1847-1931), amerik. Erfinder

AwkaBear

Junior Member

Wednesday, October 27th 2021, 9:51pm

I ordered a leak tester and regular ultitube cap to test separately without the Leakshield. I will be able to pressure test in about a week.

I know there is a difference between water and airtight but cannot see any indication of non airtightness in the system.
The thing is that the leak is so huge the leakshield CANNOT build any pressure at all and there are no bubbles anywhere in the system.

This is what leads me to believe the leak is somewhere in the airvolume of the reservoir so either in the leakshield internally or the gasket to the leakshield on top of the reservoir.

I have checked the gasket and reseated the leakshield multiple times and don't see any evidence of a gap in the mounting of the leakshield

I can post more pictures and information if you would like.

I'm still skeptical that my leakshield is defective but i cannot find any other indication of a leak anywhere else

This post has been edited 1 times, last edit by "AwkaBear" (Oct 27th 2021, 9:54pm)

Remayz

Full Member

Thursday, October 28th 2021, 12:26am

i don't know if it's the camera focal, the 24 pin cable behind the reservoir throwing me off... but to my eyes the leakshield seems to be sliiiiiightly slanted towards the fan.
With no bubbles or water movement in the reservoir, it has to be a leak somewhere around or in the leakshield unit.

The pressure is still so high it has to be massive. TBH, i'd try some ghetto leak finding ^^ like.. carefully moving an incense stick or even a smoking cup of tea, like an inch from the unit, carefully going around and see if the smoke gets sucked somewhere.

AwkwaBear

Junior Member

Thursday, October 28th 2021, 6:24am

i don't know if it's the camera focal, the 24 pin cable behind the reservoir throwing me off... but to my eyes the leakshield seems to be sliiiiiightly slanted towards the fan.
With no bubbles or water movement in the reservoir, it has to be a leak somewhere around or in the leakshield unit.

The pressure is still so high it has to be massive. TBH, i'd try some ghetto leak finding ^^ like.. carefully moving an incense stick or even a smoking cup of tea, like an inch from the unit, carefully going around and see if the smoke gets sucked somewhere.

You're totally right it is slanted toward the fan, all the tubes and even the GPU is crooked in this build lol. The reservoir is actually also tilted toward the motherboard and 24 pin cable if you look at it from the side you can see the fluid level is slanted haha. I rushed to get everything done in one day so the moment something fit i just moved on.

Yeah maybe an incense stick could work but the thing is the leakshield vacuum pump literally never stops so while its running the smoke would just get blown by the exhaust port and it would be pretty hard to tell where the leak is once it stops and the vacuum drops near immediately to zero.

I'm thinking the only way to rule it out for sure is to replace it with a regular cap and pressure test the loop without the leakshield installed. If that holds pressure then the leak has to be the leakshield

This post has been edited 1 times, last edit by "AwkwaBear" (Oct 28th 2021, 6:30am)

Remayz

Full Member

Thursday, October 28th 2021, 9:37am

you'd have to let the smoke rise from the base of the reservoir, see if it gets pulled under the leakshield, towards the seal. whatever the pump does above, blowing it doesn't matter.
But the cap test can let you know at least if the loop is leak tight.
If nothing is dripping you can use the computer anyway. That's basically how it's done usually hehe

AwkwaBear

Junior Member

Thursday, October 28th 2021, 8:13pm

you'd have to let the smoke rise from the base of the reservoir, see if it gets pulled under the leakshield, towards the seal. whatever the pump does above, blowing it doesn't matter.
But the cap test can let you know at least if the loop is leak tight.
If nothing is dripping you can use the computer anyway. That's basically how it's done usually hehe
Ah yeah thats a good idea ill have to try it out.

But yeah so far the computer hasnt leaked a drop in 4 days with it running most of the day everyday.

It would be nice to have my leakshield working though. I spoke with the RMA team and they want to wait for me to leak test with the regular cap before they dispatch a replacement

Thursday, October 28th 2021, 8:31pm


*snip*

I'm thinking the only way to rule it out for sure is to replace it with a regular cap and pressure test the loop without the leakshield installed. If that holds pressure then the leak has to be the leakshield

For comparison, pressure test with the leakshield installed also.

AwkwaBear

Junior Member

Thursday, October 28th 2021, 10:26pm


*snip*

I'm thinking the only way to rule it out for sure is to replace it with a regular cap and pressure test the loop without the leakshield installed. If that holds pressure then the leak has to be the leakshield

For comparison, pressure test with the leakshield installed also.
Doesn't the leakshield have an auto vent valve that relieves any overpressure though? Cause the leakshield does its leak test by using vacuum.

I thought a pressure tester would just relieve to atmosphere.

I could try it though

Edit:

Yeah i just checked you can't pressure test a leakshield as it will just relieve through the vent valve according to the Leakshield FAQ.
https://aquacomputer.de/tl_files/aquacom…FAQ_english.pdf

Source code

1
2
"Can I test a system with LEAKSHIELD for leakage with Dr. Drop? No. LEAKSHIELD is designed so that overpressure always discharges. This happens even when the
system is switched off. The leakage test should be performed accordingly using LEAKSHIELD itself."

This post has been edited 2 times, last edit by "AwkwaBear" (Oct 28th 2021, 11:18pm)

AwkwaBear

Junior Member

Tuesday, November 2nd 2021, 6:49am

I pressure tested and found the issue is not the Leakshield but rather the ultitube leaking from the bottom of the glass where it meets the part where the inlet and outlets connect.


Below I have a picture of me wiping some coolant out of the leaking section. There was much more before I took the pictures.




I have removed the whole reservoir and tried to tighten down the reservoir but the only place to tighten is the ring that connects the D5 pump to the section BELOW the inlet and outlet nozzles. This does not actually tighten the section that is leaking.


I believe the reservoir that I got was defective and will have to RMA at this point.

Tuesday, November 2nd 2021, 9:24pm

That sucks but at least the Leak Shield sort of did its job. How much coolant was down there? Did you just not see it earlier?
Sorry you have to go through an RMA. Good luck!

AwkwaBear

Junior Member

Monday, November 8th 2021, 6:20am

That sucks but at least the Leak Shield sort of did its job. How much coolant was down there? Did you just not see it earlier?
Sorry you have to go through an RMA. Good luck!

Okay so I forgot to update with the most recent developments.

Turns out the glass was not COMPLETELY tightened as I tried really hard and was able to get about another quarter turn. I think that accounts for the drops I was able to find in the pictures I posted. I swear I tried to tighten the glass multiple times before but I think its just really hard to turn.

The leaking from the bottom of the glass was not present until I switched out the leakshield for generic Ultitube cap with one port having a leak tester and the other having a stopper to plug it. I started trying to use a pressure tester to maintain 0.5 Bar pressure but it was still rapidly dropping. I think pressurizing the glass started pushing water through the tiny gap as I kept trying to pressure test it for a while.

The tiny leak at the bottom of the glass would not be large enough to account for the fact that pressure completely reduces to zero in a couple seconds when pressurizing the loop/leakshield unable to even build pressure. Also, if the leak was from the bottom of the glass there would be insane bubbles coming in through there when drawing vacuum with the leakshield.

I realized after going back the next day that I am an idiot and did not see that there is actually a pressure equalization hole on the regular ultitube cap in between the ports. I duct taped this over and pressed down on it with my finger and leak tested the reservoir and lo and behold it holds pressure perfectly. This was while the reservoir was installed in the loop so at this point I believe I have ruled out every component but the leakshield.

I ran the leakshield again to verify it still cannot build pressure and its the exact same issue as before.

I have corrected my information I sent to the RMA team although they did already send me a new glass portion. (They have been extremely helpful and responsive, A+ customer service!)

The replacement leakshield will take another week to ship but I will replace it and retest and update this post with my results.

TLDR;
leak in glass was from not being tightened all the way, loop successfully leak tested with regular ultitube cap however, leakshield still doesnt work meaning that the leakshield must be defective, I will update this post with results after replacing the leakshield

This post has been edited 1 times, last edit by "AwkwaBear" (Nov 8th 2021, 6:23am)