• 20.04.2024, 03:41
  • Registrieren
  • Anmelden
  • Sie sind nicht angemeldet.

 

Lieber Besucher, herzlich willkommen bei: Aqua Computer Forum. Falls dies Ihr erster Besuch auf dieser Seite ist, lesen Sie sich bitte die Hilfe durch. Dort wird Ihnen die Bedienung dieser Seite näher erläutert. Darüber hinaus sollten Sie sich registrieren, um alle Funktionen dieser Seite nutzen zu können. Benutzen Sie das Registrierungsformular, um sich zu registrieren oder informieren Sie sich ausführlich über den Registrierungsvorgang. Falls Sie sich bereits zu einem früheren Zeitpunkt registriert haben, können Sie sich hier anmelden.

Major issues with Leakshield

Montag, 24. Oktober 2022, 19:01

Hello,
I have some issues with leakshield. It would be great if you could help me solve them.
I didnt have some of these issues when I had the flow next sensor in the loop, but I stopped using it because it had its own issues e.g. becoming noisy when water temperature hits 35 degrees
  • Every time I shut down the PC, I get a leak alert. It's like it's not understanding that the d5 next pump has shut down
  • It'll throw an alert when the d5 next pump goes to a much lower power level.
  • Leak test is not holding negative pressure about 400, while at 375 it looks almost stable
  • When establishing pressure during shield setup, it'd stop and throw an alert that the pump was triggering too often, so it must be a leak... but the pump was triggering often because it had to do a fill up test and then establish operational negative pressure, picture below. I can fix that by manually lowering minium negative pressure to something lower than the current negative pressure



My setup :
  • Res/Pump: Aquacomputer ULTITUBE D5 NEXT 150 PRO with Leakshield
  • Tubing: EK Water Blocks EK-Tube ZMT Matte Black 9.5/15.9mm
  • Water mix: Aquacomputer Double Protect Ultra Clear



Dieser Beitrag wurde bereits 8 mal editiert, zuletzt von »piotrkun« (24. Oktober 2022, 19:16)

Montag, 24. Oktober 2022, 19:25

The first thing I notice is that you have practically no air in the reservoir, but the LEAKSHIELD needs air to work properly, the less air volume there is, the more sensitive it reacts to pressure changes.

Have you calibrated the virtual flow sensor of the D5 NEXT and given the LEAKSHIELD as a flow sensor?

Is the system completely vented?

If the LEAKSHIELD is not holding pressure I would suspect that there is an air leak somewhere, does air visibly collect somewhere in the system when you have turned off the computer overnight and the LEAKSHIELD is in shield mode?
Es gibt keinen Ausweg, den ein Mensch nicht beschreitet, um die tatsächliche Arbeit des Denkens zu vermeiden.
Thomas Alva Edison (1847-1931), amerik. Erfinder

Montag, 24. Oktober 2022, 19:45

That was as excellent tip about calibrating the pump. I have managed to calibrate it and additionally using an offset of 25% it's only off by 10% from the real flow. Before it'd show 90 l/h, and now it stays around 200 l/h which is not bad at all for a virtual sensor. The flow next sensor would show something around 225 l/h at that speed. I'll try to drain some of the water tomorrow and see how that works. Thanks a lot for some great suggestions.


Dieser Beitrag wurde bereits 5 mal editiert, zuletzt von »piotrkun« (28. Oktober 2022, 20:43)

Montag, 24. Oktober 2022, 22:05

I would give it some time to settle in and take 2-3 days to see how it behaves. For me it passed all the leak tests but there is a tube that seems to pull out at higher negative pressure so I can only run in monitor mode untill I have time to replace some tubes.

Montag, 24. Oktober 2022, 22:32

I'm curious how this goes for you. Hopefully, it is as simple as just leaving more air space inside of the reservoir.

I have entirely stopped using my LEAKSHIELD because of its false alarms. It seems like a flawed execution of a good concept.

Dienstag, 25. Oktober 2022, 17:35

I'm curious how this goes for you. Hopefully, it is as simple as just leaving more air space inside of the reservoir.

I have entirely stopped using my LEAKSHIELD because of its false alarms. It seems like a flawed execution of a good concept.
Leaving more air in the tube and calibrating the virtual flow meter didn't help. Reading this forum, I have found many mentions that small pressure leak is normal. I've posted those leak tests so either the support could tell me that it's normal or someone who has leakshiled could tell me if their system behave the same. Just notice that those small pressure leakes happaned over 2000-3000 seconds.

My educated guess is that my issues are related to software. It almost looks like some of the leakshiled algorithms don't take into the account a change in power level of the pump and a change in temperature of the water.

e.g. pump power level alone can drop the negative pressure by 50. I've lowered my pump power from 100% to 5% and leakshiled has thrown an alert straight away. It shouldn't since it can communicate with the d5 next

Dieser Beitrag wurde bereits 2 mal editiert, zuletzt von »piotrkun« (25. Oktober 2022, 17:53)

Dienstag, 25. Oktober 2022, 17:49

The first thing I notice is that you have practically no air in the reservoir, but the LEAKSHIELD needs air to work properly, the less air volume there is, the more sensitive it reacts to pressure changes.

Have you calibrated the virtual flow sensor of the D5 NEXT and given the LEAKSHIELD as a flow sensor?

Is the system completely vented?

If the LEAKSHIELD is not holding pressure I would suspect that there is an air leak somewhere, does air visibly collect somewhere in the system when you have turned off the computer overnight and the LEAKSHIELD is in shield mode?
  • I have increased the air volume to like 30%+ (leakshiled shows 92% filled, but that's wrong)
  • I have calibrated the virtual flow meter
  • the system is completely vented
  • Just take a look at those leak tests screenshots. It's a very small air leak over 2000-3000 seconds. At 375 is almost stable. I was hoping someone could tell me if that's normal. I've found many mentions in this forum that a small pressure leak is expected because of e.g. temperature, pump speed, no system is 100% airtight
  • Air doesn't visibly collect anywhere after overnight with the shield on

I think those are software issues. I have managed to solve one of the problem when setting up shield would cause an alert. There is a dip in pressure after the shield has established pressure, which would cause an alert straight away. I have manually changed the minimum pressure from 375 to 300, so the dip can't cause an alert any more.

Dienstag, 25. Oktober 2022, 18:11

You should give us some more information.

First of all, Aquasuite has a small camera in the upper right corner, so you can take a picture of the whole window, not just a diagram.

Then it would be quite helpful if you would post the configuration of the LEAKSHIELD, without that it is also hard to say anything.

Then also a picture of the status page would be very interesting.

While writing I just saw that you just posted again.

If I interpret this correctly you regulate your pump, why?

When you start your computer you will always see the same picture, the pressure drops all by itself, this is because the water heats up and the negative pressure decreases as a result, conversely, the negative pressure increases, for example, if you are only on the desktop after a game or benchmark.

Beyond these changes, every system loses vacuum, my system lost 200 mbar overnight after the conversion to hardtubes and pumped every morning, in addition, the tube between GPU and CPU were full of air.
After I tightened the screw connections of the 90° angles, the spook was over and the LEAKSHIELD pumps again about every 5 days.

I would try to turn off the adaptive pressure control in the settings and set fixed values, beyond that I would set the pump to a fixed speed, the regulation brings you, if at all, only a marginal advantage.

On volume and pressure changes in general, the LEAKSHIELD measures extremely accurately, so there are always fluctuations up or down.

If you think that something is wrong with the LEAKSHIELD, it is best to contact support directly. support@aqua-computer.de



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

Dienstag, 25. Oktober 2022, 18:35

Zitat

If I interpret this correctly you regulate your pump, why?
I just finished building the system, so I've been testing how does the pump speed affect the cpu/gpu temperatures and where the sweet spot is between performance and noise. I also had issues with the pump noise and vibration, that's why I have been playing around with the power levels a lot. I have installed "Watercool HEATKILLER® Tube - Decoupling Kit for Stand" today which solved all my d5 noise and vibration issues. Now I can run the pump at 100%, and It's still pretty quiet.

Zitat

Then it would be quite helpful if you would post the configuration of the LEAKSHIELD, without that it is also hard to say anything.
Made a totally fresh default configuration for leakshield, using ultitube 150 defaults + adaptive pressure




Zitat

Then also a picture of the status page would be very interesting.
status page of leakshield :



status page of d5 next :



Replicating the shield setup issue again



Dieser Beitrag wurde bereits 4 mal editiert, zuletzt von »piotrkun« (28. Oktober 2022, 20:45)

Dienstag, 25. Oktober 2022, 18:48

You do realize that the LEAKSHIELD has very little headroom with these settings?

At 100% pump speed and 375 mbar pressure from the pump, the LEAKSHIELD inevitably reacts very quickly, since the system can only be protected with an overpressure above 375 mbar.

Since the flow rate values do not help you, I would rather go to a value around 100 l/h, then the pump generates a much lower pressure and the LEAKSHIELD has more headroom.

With the status page I meant the one in the Aquasuite under LEAKSHIELD.
Es gibt keinen Ausweg, den ein Mensch nicht beschreitet, um die tatsächliche Arbeit des Denkens zu vermeiden.
Thomas Alva Edison (1847-1931), amerik. Erfinder

Dienstag, 25. Oktober 2022, 18:55

You do realize that the LEAKSHIELD has very little headroom with these settings?

At 100% pump speed and 375 mbar pressure from the pump, the LEAKSHIELD inevitably reacts very quickly, since the system can only be protected with an overpressure above 375 mbar.

Since the flow rate values do not help you, I would rather go to a value around 100 l/h, then the pump generates a much lower pressure and the LEAKSHIELD has more headroom.

With the status page I meant the one in the Aquasuite under LEAKSHIELD.
I've fixed that already before by lowering the minimum pressure to 325 in the configuration. I've done a fresh default configuration to showcase the problem to you and removed my configuration changes from affecting the results.


Here's the picture :






Dieser Beitrag wurde bereits 2 mal editiert, zuletzt von »piotrkun« (25. Oktober 2022, 18:58)

Dienstag, 25. Oktober 2022, 19:03

Removed to lower the text spam in this thread.

Dieser Beitrag wurde bereits 1 mal editiert, zuletzt von »piotrkun« (25. Oktober 2022, 19:31)

Dienstag, 25. Oktober 2022, 19:03

You don't understand me, the pump generates a pressure of 375 mbar, as soon as the vacuum drops below this value your system is not protected, I can't say exactly how much vacuum is necessary, but if you are only protected up to let's say 385 or 390 mbar vacuum there is not much left.

With these settings, the LEAKSHIELD makes little sense and will certainly give you no pleasure.
Es gibt keinen Ausweg, den ein Mensch nicht beschreitet, um die tatsächliche Arbeit des Denkens zu vermeiden.
Thomas Alva Edison (1847-1931), amerik. Erfinder

Dienstag, 25. Oktober 2022, 19:09

You don't understand me, the pump generates a pressure of 375 mbar, as soon as the vacuum drops below this value your system is not protected, I can't say exactly how much vacuum is necessary, but if you are only protected up to let's say 385 or 390 mbar vacuum there is not much left.

With these settings, the LEAKSHIELD makes little sense and will certainly give you no pleasure.
Ah I understand, but the real problem here is that leakshield is not reaching the set pressure because of this dip during setup. This is the real issue here. It can easily pump up to 440 during the leak testing and there are no drops there. Set point pressure is set to 425 and that should have worked perfectly fine.

Dieser Beitrag wurde bereits 4 mal editiert, zuletzt von »piotrkun« (25. Oktober 2022, 19:14)

Dienstag, 25. Oktober 2022, 19:24

You are right, the drop is a problem, because 30 mbar already brings you very close to the limit, if your pump produces significantly less pressure, because you do not set it to 100% there is no Problem.
Also my pressure drops immediately after pumping, but since my pump at 65% and 3500 rpm produces only 253 mbar pressure it is completely not important!
Don't get hung up on one behavior, but set up the system so the components can work, then see how long it takes for the LEAKSHIELD to pump up and if you still get alarms, it can't work cleanly in the current configuration.
Es gibt keinen Ausweg, den ein Mensch nicht beschreitet, um die tatsächliche Arbeit des Denkens zu vermeiden.
Thomas Alva Edison (1847-1931), amerik. Erfinder

Dienstag, 25. Oktober 2022, 19:25

Thank you for trying to help me. I have lowered the pump speed to 70% to at least have the protection working. But this shouldn't be the case at all, leakshield specification says it can handle a single D5 pump at max pressure.

Dieser Beitrag wurde bereits 1 mal editiert, zuletzt von »piotrkun« (25. Oktober 2022, 19:26)

Donnerstag, 27. Oktober 2022, 19:19

Your LEAKSHIELD membrane is clogged which causes the pressure drop after establishing vaccum.
Do the clogged membrane there is a big difference between the vacuum in your loop and the LEAKSHIELD itself. The solution ist to clean the membrane or exchange it.
Cleaning can be done by adding some drops of isopropanol to the membrance surface, wait a few minutes and remove it with a tissue. Don't touch the membrane with the tissue, just remove the isoprpanol.
Do this a few times (up to 10 times) to clean it.

Additional you should exchange your coolant. We noticed that this often occurs with the EK ZMT tubing but we have currently no idea why. If this tube is just EPDM it should be clean and shouldn't issue any softener issues.
Less than 1% of our LEAKSHIELD customers have issues with clogged membranes. But after exchaning the cooland it should be mostly solved.

Donnerstag, 27. Oktober 2022, 23:43

If it's like mine, the clogging is due to condensation. There is no fix for this other than to stop using the product. ;(

Montag, 31. Oktober 2022, 17:05

Mine looks really damaged after about 6 months. There is black particles and after cleaning with IPA it has strange 2 tone texture.

Dienstag, 1. November 2022, 03:13

Wow. That membrane does not look good at all. What are you using for coolant?