16.04.2024, 22:17
Sprache ändern
Registrieren
Anmelden
Sie sind nicht angemeldet.
High Flow NEXT and conductivity
Mittwoch, 23. Juni 2021, 08:20
Greetings,
I have a situation that I don't fully understand and wanted to get some insight from the conductivity buffs.
I recently completed a custom loop with the following components (that actually touch the fluid):
Fittings: EKWB (black) Nickel
Tubes: EKWB 16mm OD Hard tube acrylic
Radiators: 2x Hardware Labs Nemesis 360GTS (copper/brass I believe)
Blocks: EKWB Nickel+Plexi Monoblock, EKWB Nickel+Plexi 3080 GPU Block, EKWB Acrylic O11D Distro Plate*
*The aluminum oxide mentioned in the materials specification for the DDC pump is not exposed to water and is perfectly safe to run in copper/nickel loops.
Monitoring: Aquacomputer Nickel-plated Brass Calitemp and Aquacomputer High Flow NEXT (unsure of construction - but my assumption is it's consistent with copper, nickel, etc.)
Steps taken (all distilled water involved was Ozarka steam distilled water):
All of the cut tubes were cleaned with tap water and dried before installation.
Fluid leak test with distilled water.
Drained most of it.
Primochill SysPrep concentrate mixed with more distilled water.
Drained most of it.
A few fill/pump/drain runs.
Filled properly with more distilled water.
At this point the High Flow NEXT was reading 12.5 µS/cm and the water quality was at 100%. I found a thread that is somewhat related to this (as I was going to put EKWB Cryofuel in until I found this thread: high flow NEXT - The next generation of flow sensors). Since the High Flow NEXT is only calibrated to DP Ultra and distilled water I decided to order some of the former which I will be putting in the loop as soon as I understand what is happening here.
It remained this way for a couple days until suddenly the conductivity started rising. This was subsequently joined by a reduction in water quality (presumably to be expected). During this time the computer was largely running passively as I have been too busy to begin overclocking. According to the flow sensor I had set it to around 150 l/h (with some Corsair QLs and Noctua NF F12s running at 50%) and was not doing anything with it - fluid temperature was around the 27-28°C.
About a week later the conductivity was far higher - somewhere in the range of 50 µS/cm - and the accompanying water quality was around 80%. This was when I decided to start recording it. As you can see from the following picture it has gotten substantially worse and I can't figure out why. In the space of about 2 weeks water quality has dumpstered from 100% to nearly sub-20% and conductivity is up to about 85 µS/cm at time of writing. While it appears the conductivity is at a reasonable level from what I gather - I'd still like to understand what is happening in case it continues to get worse. Also unsure what could be affecting the drop in water quality other than conductivity given that the loop has been closed since being at 100%.
The only theory that I have is when I did the initial leak test I noticed some bright green dots on the fins of the CPU monoblock (and to a lesser extent in the GPU block). They were still there after the SysPrep run. They are still there now but it looks like there are far less of them. Assuming that's metal oxidisation of some kind - and that having run the loop for an extended period of time has washed it off the fins and dissolved it in the water - that could possibly account for an increase in conductivity although I wouldn't know by how much.
Alternatively is it possible this has something to do with the distilled water actually being corrosive itself? I found an interesting article on the matter here: https://www.overclockers.com/pc-water-co…mistry-part-ii/
I just want to make sure I haven't done anything dumb before I put in the coolant and begin using the computer
Please advise - any insight is most welcome
Thanks
I have a situation that I don't fully understand and wanted to get some insight from the conductivity buffs.
I recently completed a custom loop with the following components (that actually touch the fluid):
Fittings: EKWB (black) Nickel
Tubes: EKWB 16mm OD Hard tube acrylic
Radiators: 2x Hardware Labs Nemesis 360GTS (copper/brass I believe)
Blocks: EKWB Nickel+Plexi Monoblock, EKWB Nickel+Plexi 3080 GPU Block, EKWB Acrylic O11D Distro Plate*
*The aluminum oxide mentioned in the materials specification for the DDC pump is not exposed to water and is perfectly safe to run in copper/nickel loops.
Monitoring: Aquacomputer Nickel-plated Brass Calitemp and Aquacomputer High Flow NEXT (unsure of construction - but my assumption is it's consistent with copper, nickel, etc.)
Steps taken (all distilled water involved was Ozarka steam distilled water):
All of the cut tubes were cleaned with tap water and dried before installation.
Fluid leak test with distilled water.
Drained most of it.
Primochill SysPrep concentrate mixed with more distilled water.
Drained most of it.
A few fill/pump/drain runs.
Filled properly with more distilled water.
At this point the High Flow NEXT was reading 12.5 µS/cm and the water quality was at 100%. I found a thread that is somewhat related to this (as I was going to put EKWB Cryofuel in until I found this thread: high flow NEXT - The next generation of flow sensors). Since the High Flow NEXT is only calibrated to DP Ultra and distilled water I decided to order some of the former which I will be putting in the loop as soon as I understand what is happening here.
It remained this way for a couple days until suddenly the conductivity started rising. This was subsequently joined by a reduction in water quality (presumably to be expected). During this time the computer was largely running passively as I have been too busy to begin overclocking. According to the flow sensor I had set it to around 150 l/h (with some Corsair QLs and Noctua NF F12s running at 50%) and was not doing anything with it - fluid temperature was around the 27-28°C.
About a week later the conductivity was far higher - somewhere in the range of 50 µS/cm - and the accompanying water quality was around 80%. This was when I decided to start recording it. As you can see from the following picture it has gotten substantially worse and I can't figure out why. In the space of about 2 weeks water quality has dumpstered from 100% to nearly sub-20% and conductivity is up to about 85 µS/cm at time of writing. While it appears the conductivity is at a reasonable level from what I gather - I'd still like to understand what is happening in case it continues to get worse. Also unsure what could be affecting the drop in water quality other than conductivity given that the loop has been closed since being at 100%.
The only theory that I have is when I did the initial leak test I noticed some bright green dots on the fins of the CPU monoblock (and to a lesser extent in the GPU block). They were still there after the SysPrep run. They are still there now but it looks like there are far less of them. Assuming that's metal oxidisation of some kind - and that having run the loop for an extended period of time has washed it off the fins and dissolved it in the water - that could possibly account for an increase in conductivity although I wouldn't know by how much.
Alternatively is it possible this has something to do with the distilled water actually being corrosive itself? I found an interesting article on the matter here: https://www.overclockers.com/pc-water-co…mistry-part-ii/
I just want to make sure I haven't done anything dumb before I put in the coolant and begin using the computer
Please advise - any insight is most welcome
Thanks
Mittwoch, 23. Juni 2021, 09:55
Distilled water is definitely very corrosive. Given time it will do short work of the nickel plating on your blocks. It will also darken brass and copper parts 
I believe that's what you're seeing, water slowly incorporating nickel and copper ions.
You need to use a corrosion inhibitor of some sort to avoid having that happen. The problem with Cryofuel is it uses another method of corrosion prevention that basically gives the water higher conductivity. It's totally fine for the loop, but just don't work with the AC sensor.
So, DP ultra it is, and maybe don't wait too much.
If i were you i'd drain the loop of contaminated water, and refill with the premix to start fresh.
At work (semiconductor fab), all the cooling circuit is purified water (not DI but close). Only old machines still have brass quickconnects and copper circuits. They clog and are flat out black inside, because of corrosion. All the rest has to be stainless steel because of its corrosive nature.
I believe that's what you're seeing, water slowly incorporating nickel and copper ions.You need to use a corrosion inhibitor of some sort to avoid having that happen. The problem with Cryofuel is it uses another method of corrosion prevention that basically gives the water higher conductivity. It's totally fine for the loop, but just don't work with the AC sensor.
So, DP ultra it is, and maybe don't wait too much.
If i were you i'd drain the loop of contaminated water, and refill with the premix to start fresh.
At work (semiconductor fab), all the cooling circuit is purified water (not DI but close). Only old machines still have brass quickconnects and copper circuits. They clog and are flat out black inside, because of corrosion. All the rest has to be stainless steel because of its corrosive nature.
Mittwoch, 23. Juni 2021, 21:15
Thanks for the quick response
After draining I can see the browning of the nickel plates (and the interior of the fittings) - I have now filled it with DP Ultra and will report back results once I've hooked everything up (on pump by-pass at the moment)
I did have to leave a bit of the distilled water in (about ~50-100ml in a loop of ~800-850ml) as a full drain would have required significant disassembly. I'm hoping it will dilute without issue.
Hopefully this will mitigate further damage. Turns out I was doing the dumb thing
After draining I can see the browning of the nickel plates (and the interior of the fittings) - I have now filled it with DP Ultra and will report back results once I've hooked everything up (on pump by-pass at the moment)
I did have to leave a bit of the distilled water in (about ~50-100ml in a loop of ~800-850ml) as a full drain would have required significant disassembly. I'm hoping it will dilute without issue.
Hopefully this will mitigate further damage. Turns out I was doing the dumb thing
Ähnliche Themen
-
English forum »-
high flow NEXT - The next generation of flow sensors
(26. August 2020, 11:17)
-
- Wasserkühlung »
-
2 Fragen zu meinem neuen high flow NEXT (Wasserqualität & Alarm bei ausgeschaltetem PC)
(21. März 2021, 03:13)