Just noticed that my Leakshield is adding just a small amount of pressure more often, to bring the pressure just above the minimum-pressure (197mbar) instead of the set point (around 240mbar). So it hovers around 200mbar, then when it ducks below the minimum, it adds just a little. I thought it should add enough to bring the pressure up to the set-point, rather than the minimum.

Is there something I am doing wrong, or am I thinking about this incorrectly?

Thanks!

I should add that my D5 pump is running at 50% static speed (not changing). I have Leakshield settings to use default pressure values, and adaptive pressure is ON.


Thanks!

Sorry.

So it generally looks something like this.

Now also, though the water level is not decreasing, I notice the reported fill level is lowering substantially (is now at 66% when was 88% yesterday), but there is no leak... the pressure alarm is going off frequently tonight.

This is since I turned the computer back on from a few hours ago.

I do think there might be an air leak somewhere, but I don't know why this would change the behaviour of the Shield mode going to just above minimum pressure, instead of Set-Point

I have re-ran "measure fill level" multiple times.

I get between 72 and 60% (again, the level of liquid in the res does not appear to have to changed from yesterday when it was over 80).

Back to back measurements are giving different results, and I feel like there is a dip in pressure when shield mode is turned on and set point pressure is reached (per the screenshot above... shield mode activated.. pressure builds to around set point, then falls to near the "min" level... sits there and slowly decays until the pump activates a small amount keeping it right at the minimum level. I do not believe this is the same behaviour it was showing yesterday.

oh no

Just distilled water at the moment (switching to a coolant soon). No dyes.

Tubing is EK ZMT, 16/10.


I have been keeping it clean, I can't imagine it would be dirt ...

What do you recommend to do?

Also, the system has only been built for 10 days...

Is it possible the LeakShield is faulty?

Thank you for your reply!

I did flush the system: radiators were thoroughly rinsed until water was clear: tubes were rinsed, and a flush of distilled water was made as well. I guess it's possible things still come out, this makes sense.

Isopropyl alcohol is safe to use on the membrane? I thought it was very delicate. Is there a procedure I can follow to ensure I do not damage it?

Thanks for your assistance Sebastian.

Confirming that this did seem to somewhat fix the problem... it started recording pressure closer to the set-point, but as of today, I noticed it was repeating the behaviour of last time. I opened the LeakShield after switching everything off, and noticed there was a lot of condensation on the bottom of the leakshield and membrane.

Not sure if I am doing something wrong to be causing this. The water temp is only set to 35degC but I can't imagine anything can be done about this.. but if it is going to be an ongoing problem, I wonder if there is a way that people handle it?

Thanks for the feedback, but I don't feel like I should have to buy a new membrane yet.

It certainly cannot have been damaged by plasticizers.. this is a brand new (2 weeks old now) system that was built using ZMT tubing. The system was flushed wish water before.

I have since changed the fluid to Corsair Hydro X clear (thank you for the suggestions both @Remayz and @Stephan ) as I cannot cheaply get DP Ultra in my country (quotes are 88Euro for shipping through AquaComputer shop).

I don't think it's plasticizer or contaminant related... the system is brand new and using brand new parts that do not contain plasticizer.

Also when I look on the underside of the leakshield, I can see loads and loads of condensation: the bottom of the lid is dripping wet, and the membrane is partly covered. I know from here, and also a post on Reddit that an AquaComputer staff person said the membrane won't work if the membrane is wet, and it seems to indeed be wet with condensation (can confirm that since cleaning it, the liquid did not directly touch the membrane).

I guess I have a situation where the condensation is forming on the membrane.

The previous comment on this forum I can't find so I might be mistaken, confusing it with the reddit post (or another one). But here is the Reddit link . The quoted text is

"Not sure about the setup that you have in mind now, but if you can make sure that LEAKSHIELD will not suck in coolant through this T-connector, then it will work.

A membrane protects the device from coolant but when this membrane becomes wet, LEAKSHIELD can not operate correctly until it is dry again."

No worries, I'm happy for the input.

I have 122mm from the top of the LED ring (positioned at the bottom of the reservoir), to the lid of the reservoir (bottom of the leakshield). The water level is around 98mm. This gives around the right amount of.. I don't think it's overfilled but I might get told I am wrong. Visibly, I would estimate it at around 3/4 or 4/5 filled, keeping in mind this is me saying visually how full it is to my eye.

So just now this happened and I found it it interesting (attached a screenshot): I went from gaming, to mining, and switching between. I heard the leakshield turn on really quickly which I knew meant that the minimum level had been triggered and it added just enough to get over the min level (see between 12:56:15 an 12:56:40 of the picture). I went in, released the pressure, and immediately activated shield-mode. it did the level test, and activated pressure back up to the set point.

I think the condensation is not fully inhibiting it, but rather there is a software setting that, when minimum level is triggered, does not add pressure back to set point, but just tops it back over minimum. I don't understand why this would be the case but it does seem to be given I did not change any setting, and you can clearly see the pump activation taking the pressure back over minimum but not to the set point in this photo...

Regarding the somewhat contradictory statement.. I agree.. but I think it might be translation. Also I note Sebastian did say "(water / dirt / somthing other)." in one of the earlier messages of this thread.

It makes sense to me that a large water droplet over the membrane could inhibit function. I do have a possible future engineering solution to this (if it turns out to be a problem, which it doesn't seem to be for others so far) but it wouldn't work on this round of leakshields: condensation bumps... e.g. the bottom of some cast-iron pot lids have little dimples... these allow condensation to form on them and drip down off of each dimple... This should mean that condensation won't form into larger droplets that might make their way to centre sensor/membrane area but instead drop back into the reservoir, except for the condensation that actually forms on the centre piece (which might negate the whole process anyway). Anyway, that's an aside. I am curious about the way the software works in adding pressure when the minimum level is triggered.

And I agree.. this is an outstanding piece of equipment and technology.. it makes me so much more comfortable to have such a system. I love it and don't want to be without it.