R9 5950x and RTX 3080 - Is my loop too small? Too little radiator? Too little pump?

Zarathustra[H] said:
There really should not be bubbles moving around inside your loop after the system hits stability after being filled.

It sounds to me like your issue may be that you still have air trapped in the system.

A few thoughts:

1.) Did you do "case gymnastics" after your fill? Varying pump speed up and down and leaning the case on its sides and front and back up again, while the pump is running to try to get all the air out. You can keep tilting and banging the system around for hours and still not get all the air out. This is usually necessary to get all the air out of the system. With some systems it clears out on it's own over time, but with others depending on the loop and how easy it is for freed bubbles to get sucked back in, it doesn't and needs lots of case movement

2.) What do you use as your coolant? Air bubble problems are a lot less rare if you use a coolant that contains a good surfactant. If you just use straight distilled water you are going to get much more air caught in the system.
Click to expand...
I turned the case every which way but loose ;-). I shook out a lot of bubbles. I did the Mayhem's part 2 I think before I filled it up with Koolance fluid. With the Mayhems, I ran it overnight, then flushed it with distilled water 4 times per the instructions. However, I don't think I got all of it out due to how the components and the tubing in the case is situated. I certainly diluted the heck out of it though. I think there is just enough left to allow it to bubble where it enters the reservoir. If you look at the first picture I posted, you can see the bubbles up top. Some of the very tiny bubbles somehow get sucked all the way to the bottom when the pump is at 100%. I need to put a little more Koolance in through my fill port to try an cut down on the air space in the reservoir some more and I guess I need to shake it out. I know air can cause inefficiency, but this seems extreme.
 
Last edited:
JSHamlet234 said:
You did go from 220W into 3x120mm rad space to 600W into 5x120 rad space. In doing so, you went from 73.3 watts per 120mm to 120 watts per 120mm, so your air to water delta under combined load will be 63% higher than before based on that factor alone. So if your delta is 12.5C now, that means it was 7.6 before. So that would make you run 5 degrees hotter under a combined CPU GPU load than you did before you brought the GPU and addl' 240 rad into the mix. Hopefully you already knew that going in.

My overall impression is that you're just pushing your chip a little too hard for the current state of your cooling. You can't necessarily run the chip just as hard as you did before if it's 5C hotter. You're also running 4 dual-rank sticks at 3600, which is a heavy load for the IOD. My guess is that your temps are getting too high under combined loads, and it's pushing the IOD off the brink of stability. Since it looks like you can't easily add more rad space, I would dial down the power limits on the CPU to get those 5 degrees back and then call it a day. There's math for that too. If your water is at 37.5 and your CPU is at 90, then the delta is 52.5. 5 less is 47.5. So 47.5/52.5*220W = 199W. So if you lower your PPT to 199, you'll reduce your package temp to 85.
Click to expand...
Thanks for your numbers. I like your analytical thought on this. I knew I'd be asking more of each mm of radiator at load, but I didn't run the numbers like you did and it is a bigger difference than I assumed. I saw others cooling similar setups to mine with a 360 and a 240 and getting much better numbers than I am before I built this out, so I'm trying to figure out what I can do. For example: https://hardforum.com/threads/rtx-3080-ftw3-ek-vector-waterblock-install-results.2009894/ What I didn't think of was how my RAM is stressing the CPU more. Also, it seems to me that gpu cooling is more efficient than cpu cooling, especially with the way Ryzen is designed. It's so small and throws so much heat into such a tiny area, it's hard to pull all of that heat out. In my head, GPUs seem to cool "better", if that makes sense? Maybe I need to pull two sticks of RAM and reconfigure my whole loop and get rid of the distroplate. I have lots of crazy 90 degree fittings and every in and out of distroplate is basically a 90. I don't think I could add any radiator to the case even without it though. I guess I could put a thin 360 on the bottom and mount the fans on the bottom of the case. But, at that point, I should just buy a different case.
 
I have a plan. I have an extra Corsair XR5 240 radiator. I'm going to add on to the feet of the case to raise it up by 1.5 inches or so and then I can add that radiator to the bottom of the case and my bottom intake fans will become pull fans. I'll be adding in the restriction of another radiator, but I'll also be able to cut out a bunch of 90 degree fittings. Think it will be worth it? I think I have all of the parts I need, it's just going to cost me time. That will give me the equivalent of two 360mm radiators plus a 120mm radiator.
 
kamikazi said:
I have a plan. I have an extra Corsair XR5 240 radiator. I'm going to add on to the feet of the case to raise it up by 1.5 inches or so and then I can add that radiator to the bottom of the case and my bottom intake fans will become pull fans. I'll be adding in the restriction of another radiator, but I'll also be able to cut out a bunch of 90 degree fittings. Think it will be worth it? I think I have all of the parts I need, it's just going to cost me time. That will give me the equivalent of two 360mm radiators plus a 120mm radiator.
Click to expand...
I would say no. The XR5 is very similar to a hw-labs GTS, if not the same, which has very high restriction (probably about the same or more than your CPU waterblock). The XR7s are HW-labs LX which has medium restriction, but the 360 edition of the XR5 has more than twice the restriction of your 360 XR7 at 1gpm. You will also be heating up the air for your other radiators. A slim radiator will probably bring the air temp by something between 1/2 and 2/3 of the delta between air and water. It will also limit your airflow into the case by quite a bit.

Out of curiosity, which use cases do you need PBO for outside of benchmarks? For games, plain PBO doesn't do much as few if any games reach the PB limits. Personally I would either do a full curve optimizer or leave it at plain PB. Even with oversized cooling the difference in performance with PBO vs PB will be 5-6% at best and a slight loss at worst. Curve optimizer is the way to go if you want to OC a ryzen 5000 CPU, otherwise PB will be equally good in 99% of tasks.
 
evhvis said:
I would say no. The XR5 is very similar to a hw-labs GTS, if not the same, which has very high restriction (probably about the same or more than your CPU waterblock). The XR7s are HW-labs LX which has medium restriction, but the 360 edition of the XR5 has more than twice the restriction of your 360 XR7 at 1gpm. You will also be heating up the air for your other radiators. A slim radiator will probably bring the air temp by something between 1/2 and 2/3 of the delta between air and water. It will also limit your airflow into the case by quite a bit.

Out of curiosity, which use cases do you need PBO for outside of benchmarks? For games, plain PBO doesn't do much as few if any games reach the PB limits. Personally I would either do a full curve optimizer or leave it at plain PB. Even with oversized cooling the difference in performance with PBO vs PB will be 5-6% at best and a slight loss at worst. Curve optimizer is the way to go if you want to OC a ryzen 5000 CPU, otherwise PB will be equally good in 99% of tasks.
Click to expand...

This is a very good point. You don't want to have one radiator as intake and another as exhaust, because then you're just recycling the same heat. I'm going to delete my "go ahead and do it" comment in light of that.
 
evhvis said:
I would say no. The XR5 is very similar to a hw-labs GTS, if not the same, which has very high restriction (probably about the same or more than your CPU waterblock). The XR7s are HW-labs LX which has medium restriction, but the 360 edition of the XR5 has more than twice the restriction of your 360 XR7 at 1gpm. You will also be heating up the air for your other radiators. A slim radiator will probably bring the air temp by something between 1/2 and 2/3 of the delta between air and water. It will also limit your airflow into the case by quite a bit.

Out of curiosity, which use cases do you need PBO for outside of benchmarks? For games, plain PBO doesn't do much as few if any games reach the PB limits. Personally I would either do a full curve optimizer or leave it at plain PB. Even with oversized cooling the difference in performance with PBO vs PB will be 5-6% at best and a slight loss at worst. Curve optimizer is the way to go if you want to OC a ryzen 5000 CPU, otherwise PB will be equally good in 99% of tasks.
Click to expand...
I'm actually using the curve optimizer. I looked it up and I have -5mv on all cores in CCD1 and -10mv in CCD2. I'm using motherboard limits for PBO. I may roll back to just standard auto setting though. I've been thinking of doing this as I don't really need the juice and the heat isn't worth it. At the time I got the 5950x, I did it just to squeeze all the performance I can out of the CPU as it was msrp $800 at that time and I wanted to get my money's worth, but it seems overkill now.

As for the extra radiator, do you have a recommendation for another one I could use that is slim if I could mount it as exhaust like the other two? I didn't realize the XR5 was so restrictive, thanks for pointing that out. I mean, I could hammer a thick one in there, but I don't want to shell out any more cash that I have to. Also, when I change things around, I'm going to get rid of the distroplate and get a dual DDC pump top and use a standard tubular reservoir. That is going to free up some space and hopefully some flow restriction. I have a second DDC pump and I figure I can use that. Two of those running at half capacity should outperform a single pump and should be quieter as well. I'm going to have a lot of open areas for my case to suck in outside air. The front of my case is a custom honeycomb like panel that will be open around the size of a 140mm fan as well as the bottom that will be open a similar size. There will also be some available airflow through the back there the distroplate was. The pumps and reservoir will be far smaller. Does any of that change your opinion at all? I plan to sell the distroplate and some extra RAM I have to fund this. To have all three rads as exhaust, but I would have to get a gimmicky flat reservoir and put it on the bottom of the case. If my graphics card was about 10 mm less wide, I could mount it normally and free up plenty of space for a more conventional reservoir at the bottom of the case, but it just won't fit with the side panel on.

This is a lower restriction 20mm radiator for only $50. https://www.performance-pcs.com/wat...c-tx240-ultrathin-radiator-xspc-tx240-bk.html 0.27 psi vs. 0.37 psi on the radiators I'm already using. I'm just grasping at straws here trying to make things work. ;-). I'm definitely ditching the distroplate and turning the front radiator right side up so that the ports are at the top. I do think that the way I have it mounted, with ports at the bottom may be helping me retain some air in the case as another poster pointed out. I mounted it like I did for ease of maintenance with being able to drain water more easily as it's the lowest point in the case. However, the way the graphics card is mounted, i'll never get all the water out of it without removing it from the case anyway.
 
Last edited:
Personally I would probably just limit the CPU to 170w or run plain PB and call it a day. Your GPU temps are well below air and your CPU should have good temps at 170 or 142 watts, even with the type of water temps you are getting. Also running at 142w would mean almost 90w less than you are getting at current max power draw, which would make it easier for your radiators as well.

Getting more air into the case will help though and is most likely the main improvement you can make without changing fans or similar. Would also run a few tests at different pump speeds. If you gain less than 1 degree on the CPU when going from 70% to 100% pump speed then you have sufficient flow IMO. My CPU loop is probably slightly less restrictive than yours and it still gets around 260 l/h on a D5 so I do believe your DDC should be enough.

Another rad should give you a few degrees lower temps, but it will not be a game changer with regards to your overall temps. A 240 slim will most likely increase your total capacity by maybe 25% at high fan speeds so you would go from 12-13 degree delta to 9-10 degree delta. At lower fan speeds (sub 1000rpm) you might gain 35-40% or so with a slim 240, but it will be diminishing returns as fan speed increases.
 
evhvis said:
Personally I would probably just limit the CPU to 170w or run plain PB and call it a day. Your GPU temps are well below air and your CPU should have good temps at 170 or 142 watts, even with the type of water temps you are getting. Also running at 142w would mean almost 90w less than you are getting at current max power draw, which would make it easier for your radiators as well.

Getting more air into the case will help though and is most likely the main improvement you can make without changing fans or similar. Would also run a few tests at different pump speeds. If you gain less than 1 degree on the CPU when going from 70% to 100% pump speed then you have sufficient flow IMO. My CPU loop is probably slightly less restrictive than yours and it still gets around 260 l/h on a D5 so I do believe your DDC should be enough.

Another rad should give you a few degrees lower temps, but it will not be a game changer with regards to your overall temps. A 240 slim will most likely increase your total capacity by maybe 25% at high fan speeds so you would go from 12-13 degree delta to 9-10 degree delta. At lower fan speeds (sub 1000rpm) you might gain 35-40% or so with a slim 240, but it will be diminishing returns as fan speed increases.
Click to expand...
For giggles, I ran two tests, one at 4048 rpm on the pump and the other at 4464 rpm on the pump. Roughly 10% difference. 4464 is the max it will do whether I have it at 50% or 100%. The test was 15 minutes with Prime 95 Small FFTs and hyperthreading enabled along with Heaven looping at 1440p. At 4046 rpm, my max water temp was 37.1C. At 4464, max water temp was 37.3C. But, average ambient temp during the first test was 24.1C. For the second test, ambient was 23.9C. The water-air delta average for both tests was 11.4C. So, absolutely zero difference going from pump speed roughly 4000 to roughly 4500 rpm. CPU max on the 4000 rpm test was 90.4 and GPU was 56.7. On the 4500 rpm test, max CPU was 90.3 and GPU was 56.1. Thank you for helping me figure that out that I am not pump limited. I may test again with lower pump speed to see if it will work at an even lower rpm. I hooked up my killawatt to the computer while testing. I saw a max of 786 watts while watching it run. While sitting here typing this with HWInfo running in the background, it's 147 watts. Now, I'm thinking about my 850 watt power supply. ;-)

I'm definitely going to rebuild this thing and reconfigure the water flow as I keep getting bubbles, now I am rethinking the extra radiator and dual pump set up. Don't know if the second pump would be worth it even with an additional radiator.
 
Temps in the high 80's with averages in the 60's seems normal for zen tbh. I have a 3700x through three different loops and two high end blocks and way overkill on rads. Those temps fall right in line. 6900 stays under 40 no matter what the cpu runs hotter than seems right. I am about to upgrade to a 5900x and I expect to see the same thing.
 
Smoked Brisket said:
Temps in the high 80's with averages in the 60's seems normal for zen tbh. I have a 3700x through three different loops and two high end blocks and way overkill on rads. Those temps fall right in line. 6900 stays under 40 no matter what the cpu runs hotter than seems right. I am about to upgrade to a 5900x and I expect to see the same thing.
Click to expand...
evhvis said:
Personally I would probably just limit the CPU to 170w or run plain PB and call it a day. Your GPU temps are well below air and your CPU should have good temps at 170 or 142 watts, even with the type of water temps you are getting. Also running at 142w would mean almost 90w less than you are getting at current max power draw, which would make it easier for your radiators as well.

Getting more air into the case will help though and is most likely the main improvement you can make without changing fans or similar. Would also run a few tests at different pump speeds. If you gain less than 1 degree on the CPU when going from 70% to 100% pump speed then you have sufficient flow IMO. My CPU loop is probably slightly less restrictive than yours and it still gets around 260 l/h on a D5 so I do believe your DDC should be enough.

Another rad should give you a few degrees lower temps, but it will not be a game changer with regards to your overall temps. A 240 slim will most likely increase your total capacity by maybe 25% at high fan speeds so you would go from 12-13 degree delta to 9-10 degree delta. At lower fan speeds (sub 1000rpm) you might gain 35-40% or so with a slim 240, but it will be diminishing returns as fan speed increases.
Click to expand...
I have some stuff coming for the re-re-build. I have a feeling it's not going to net me too much, but it will be fun. I will have three radiators. I'm going to add an XSPC TX240 9 (20mm thick) and I'll have the fans fixed at probably 750 or 800 rpm since the gain above that speed is minimal. The distroplate will be gone and replaced with a 100 ml aquacool reservoir mounted to that skinny rad on the front panel. That reservoir is tiny it's going to suck filling up the computer, but the way I have things mapped out, it's about all that will fit. I did buy the dual pump head for two DDCs. Hopefully, I can lower the sound by running two at lower speeds. I got an aquacomputer flow meter as well so I can tune the flow and really find the sweet spot. Not that I couldn't do that with testing, but now I'll be able to shut things down if flow drops. The new routing will be reservoir->GPU->360 rad->CPU->240 rad->240 rad->reservoir. I could easily swap the CPU and GPU in that order if needed. I'll have 3 intake fans on the bottom and the case front panel is wide open besides the 240 rad. Some air will also come in from the rear panel of the case that has some ventilation. I'll probably end up going back to PB instead of PBO as well. Hopefully, I can quiet this thing down a good bit.
 
kamikazi said:
I have some stuff coming for the re-re-build. I have a feeling it's not going to net me too much, but it will be fun. I will have three radiators. I'm going to add an XSPC TX240 9 (20mm thick) and I'll have the fans fixed at probably 750 or 800 rpm since the gain above that speed is minimal. The distroplate will be gone and replaced with a 100 ml aquacool reservoir mounted to that skinny rad on the front panel. That reservoir is tiny it's going to suck filling up the computer, but the way I have things mapped out, it's about all that will fit. I did buy the dual pump head for two DDCs. Hopefully, I can lower the sound by running two at lower speeds. I got an aquacomputer flow meter as well so I can tune the flow and really find the sweet spot. Not that I couldn't do that with testing, but now I'll be able to shut things down if flow drops. The new routing will be reservoir->GPU->360 rad->CPU->240 rad->240 rad->reservoir. I could easily swap the CPU and GPU in that order if needed. I'll have 3 intake fans on the bottom and the case front panel is wide open besides the 240 rad. Some air will also come in from the rear panel of the case that has some ventilation. I'll probably end up going back to PB instead of PBO as well. Hopefully, I can quiet this thing down a good bit.
Click to expand...
Fans at 750rpm doesn't do much, especially on thick radiators. If you want to run at 750rpm then slim radiators is the way to go and with lots of area. E.g. running my noctua nf-a12x25 pwm fans at 750 will give me water temps that are more than 3x higher than the water temp I get at 2000rpm. Those are among the best radiator fans and even they do struggle to provide good cooling at low rpms. Just going from 750rpm to 900rpm gives me significant improvements in water temp at steady state (takes about 20 minutes to reach completely steady state at those rpms). It is generally better to let water temp dictate the fan speed and ramp up at a certain level. I wouldn't be surprised if you hit 15+ degrees above ambient if you lock the fans at 800rpm.

My water temp goes to around 8 degrees above ambient at steady state at 750rpm with 210 watts while I can keep it around 2-2.5 degrees above ambient with the fans running at 2000rpm. That is water temp going into the CPU block.

Loop order doesn't matter much. You might get one degree or so better temp by running through a 360 before the CPU, but not worth it if need to do complicated tube runs to do it.
 
evhvis said:
Fans at 750rpm doesn't do much, especially on thick radiators. If you want to run at 750rpm then slim radiators is the way to go and with lots of area. E.g. running my noctua nf-a12x25 pwm fans at 750 will give me water temps that are more than 3x higher than the water temp I get at 2000rpm. Those are among the best radiator fans and even they do struggle to provide good cooling at low rpms. Just going from 750rpm to 900rpm gives me significant improvements in water temp at steady state (takes about 20 minutes to reach completely steady state at those rpms). It is generally better to let water temp dictate the fan speed and ramp up at a certain level. I wouldn't be surprised if you hit 15+ degrees above ambient if you lock the fans at 800rpm.

My water temp goes to around 8 degrees above ambient at steady state at 750rpm with 210 watts while I can keep it around 2-2.5 degrees above ambient with the fans running at 2000rpm. That is water temp going into the CPU block.

Loop order doesn't matter much. You might get one degree or so better temp by running through a 360 before the CPU, but not worth it if need to do complicated tube runs to do it.
Click to expand...
I should clarify, I'll run the fans on that slim radiator at 750 to 800. The thick 360 and 240 will get the normal fan treatment with a curve or set point. I can see how that was ambiguous in the way I wrote it now.

Believe it or not, putting the 360 rad between the cpu and gpu will make it easier for routing the tubes. Because I have to mount the video card vertically to get it to fit, the fittings for the cpu and gpu blocks are right on top of each other. It would be difficult to connect them directly to each other. I'm going to turn my top radiator around and have the fittings at the rear of the case. Going from gpu to radiator to cpu should give me just enough space for short runs. See the new routing in green.
20220209_211418.jpg
 
Teardown underway before rebuild. I don't think the cpu mounting was an issue.
20220218_130140.jpg
20220218_130147.jpg
 
evhvis said:
There is a lot more paste left than I would expect and there is also some areas that have no paste. My guess is that you still had a bit more thread before the mounting had the correct tension.
Click to expand...
If you look closely, which is hard because the lighting looks slightly different in each picture, where the block seems to be missing paste, the cpu has more and vice versa. In person, you could see it a lot better.
 
Testing the new pump setup before installing it. Even with the heatsinks, those pumps put off some heat pretty quickly with such a small loop.

 
kamikazi said:
Pump and res mounted.

View attachment 446183
Click to expand...
Did you tighten the block mounting screws until they will not move at all? The mounting is spring loaded and the plastic washers will protect the motherboards so you can tighten it until there is no more thread on the screws without damaging anything. The photo of paste that you posted leads me to believe that you didn't have the correct pressure last time. The mounting should have so high pressure that there should barely be any paste left between the CPU and the block as it gets squeezed out around the IHS (should be a big border of paste on the block and on the edge of the IHS).
 
evhvis said:
Did you tighten the block mounting screws until they will not move at all? The mounting is spring loaded and the plastic washers will protect the motherboards so you can tighten it until there is no more thread on the screws without damaging anything. The photo of paste that you posted leads me to believe that you didn't have the correct pressure last time. The mounting should have so high pressure that there should barely be any paste left between the CPU and the block as it gets squeezed out around the IHS (should be a big border of paste on the block and on the edge of the IHS).
Click to expand...
Tight as I can get by hand. I will tighten with pliers this time. I don't know why EK doesn’t put an Allen wrench slot or even a Phillips slot on the top of the thumbs screws. I don't think it would detract that much from the looks of it. Next cpu block I buy will be a heatkiller.
 
Last edited:
Your contact looks good brother. Those cpu block screws are meant to be hand tightened, that is why they are thumb screws. Please dont take a wrench to them.
 
kamikazi said:
Tight as I can get by hand. I will tighten with pliers this time. I don't know why EK doesn’t put an Allen wrench slot or even a Phillips slot on the top of the thumbs screws. I don't think it would detract that much from the looks of it. Next cpu block I buy will be a heatkiller.
Click to expand...
Wrench isn't needed, just tighten until it stops (no thread left). The springs create the pressure, but they need the standoffs to be tight and the thumbscrews to be out of thread to get the right pressure. Basically if they get difficult to turn then keep going until full stop. I only turn about 1/2 turn each time on each screw when mounting in a criss cross pattern until tight as that would give me a consistent mount. The image below is what my CPU would look like when I removed the same type of block (EK velocity plexi) from my CPU with a tiny bit more paste on the block where they made contact. The paste on the left side is from the border around the block as that side got loosened first. I did swap for a Techn AM4 when i built my new loop, but the velocity works quite good as well. The velocity definitly is prettier though.

cpu.jpg
 
Last edited:
Get yourself a good pair of mechanix type gloves to help in tightening them down. I use them to keep from tearing my fingers up on the knurling and to avoid the giant heatsinks that inevitably chop my knuckles up.
Your spread looked A-OK btw. I wouldn't worry about it if your temps were solid.
 
kamikazi said:
Plumbing is done. I think it's going to be a pain to fill.
View attachment 448383
Click to expand...
I use a fill bottle with a flexible tube. Filling is slow due to having to squeeze the liquid out, but it allows to you to get to places that you can't fill in other ways. looking forward to seeing your temps and flow. Good luck with setting up and fingers crossed that there are no leaks :)

Edit: Those ram sticks must be painful to get in with everything in that is in front of the slots....
 
Last edited:
evhvis said:
I use a fill bottle with a flexible tube. Filling is slow due to having to squeeze the liquid out, but it allows to you to get to places that you can't fill in other ways. looking forward to seeing your temps and flow. Good luck with setting up and fingers crossed that there are no leaks :)

Edit: Those ram sticks must be painful to get in with everything in that is in front of the slots....
Click to expand...
Actually, the way everything is set up, RAM will actually not be that bad. Because of the perspective, you can't see it, but the tube exiting the cpu and going to the back radiator is a giant loop that almost reaches the side panel. I have so much tubing run it probably doubles the reservoir, which is only 100ml.
 
Last edited:
Finally got around to some more work on the beast. All the fans are mounted and it wasn't loads of fun. Probably only 3 of the fans have all 4 screws holding them in place. I'd say 10 fans is about the max you can fit in this case. Notice the 2 Noctua NF A12x25 fans kind of hiding in the back. I didn't want to spring for two more Corsair LL120s, and the Noctuas are far better, quieter fans. If I wanted to blow $250, I could see myself replacing all the Corsair fans with those in black and getting my lighting fix from a couple of strips. Hopefully, I'll have it up and running in a couple of days.

computer.jpg
 
kamikazi said:
Finally got around to some more work on the beast. All the fans are mounted and it wasn't loads of fun. Probably only 3 of the fans have all 4 screws holding them in place. I'd say 10 fans is about the max you can fit in this case. Notice the 2 Noctua NF A12x25 fans kind of hiding in the back. I didn't want to spring for two more Corsair LL120s, and the Noctuas are far better, quieter fans. If I wanted to blow $250, I could see myself replacing all the Corsair fans with those in black and getting my lighting fix from a couple of strips. Hopefully, I'll have it up and running in a couple of days.

View attachment 453602
Click to expand...
nice, looking forward to temps :)
 
I fired it up finally, but still have some work to do on the rgb ridiculousness. All I did so far was calibrate the flow sensor. Looks like at 9% on the two pumps, I get just below 200 liters per hour. I can get around 275 liters per hour at 30%. 275 liters per hours is right at 1 gallon per minute which seems like the standard to shoot for at load. From there, you can keep cranking up the pumps, but don't make as much headway. I'm going to work on curves vs. set point based on water to air delta this week I believe.
 
OK, I just ran Heaven at 2560 x 1440 along with P95 FFTs for about 45 minutes with the pump and all fans at max speed. CPU and GPU were stock. My Killawatt showed up to 707 watts being consumed. HWINFO 64 showed the CPU maxed at 124 watts while the GPU maxed at 380 watts. Max temp on the CPU was 62C which was 34 above ambient. Max temp on the GPU Core was either 58.1C (GPU [#0]: NVIDIA GeForce RTX 3080) at 30C above ambient or or 48.9C (GPU [#0]: NVIDIA GeForce RTX 3080: EVGA iCX at 20.9C above ambient. Max temp on the memory was 58.0C at 30C above ambient or 48.4C at 20.4C above ambient respectively. Max coolant temp was 36.0C and max above ambient was 7-8C. Max ambient was 30C. Max coolant flow was 322.80 liters per hour which translates to 1.42 gallons per minute. However, I believe I had the pumps limited to a max of 70%, so maybe that was max flow.

Max Temps above ambient with max cooling:
CPU: 34C
GPU: 30C or 20.9C
Coolant: 7C to 8C

computer.jpg
 
Last edited:
kamikazi said:
OK, I just ran Heaven at 2560 x 1440 along with P95 FFTs for about 45 minutes with the pump and all fans at max speed. CPU and GPU were stock. My Killawatt showed up to 707 watts being consumed. HWINFO 64 showed the CPU maxed at 124 watts while the GPU maxed at 380 watts. Max temp on the CPU was 62C which was 34 above ambient. Max temp on the GPU Core was either 58.1C (GPU [#0]: NVIDIA GeForce RTX 3080) at 30C above ambient or or 48.9C (GPU [#0]: NVIDIA GeForce RTX 3080: EVGA iCX at 20.9C above ambient. Max temp on the memory was 58.0C at 30C above ambient or 48.4C at 20.4C above ambient respectively. Max coolant temp was 36.0C and max above ambient was 7-8C. Max ambient was 30C. Max coolant flow was 322.80 liters per hour which translates to 1.42 gallons per minute. However, I believe I had the pumps limited to a max of 70%, so maybe that was max flow.

Max Temps above ambient with max cooling:
CPU: 34C
GPU: 30C or 20.9C
Coolant: 7C to 8C

View attachment 456114
Click to expand...
Nice that you have gotten the temps to a good place :)
 
Man, Microsoft Flight Simulator 2020 is hard on a system. I set up aquasuite to have the fans max out when you hit 11 degrees water to air delta. My 2500 max speed fans hit 2000 and my 2000 max speed fans hit 1500. Max water to air delta was 9.6 degrees C. CPU (Tctl/Tdie) topped out at 75.5C and GPU topped out at 58.3 or 50.7 depending on which sensor I trust. Ambient was around 26 or 27.
 
One strange thing I've noticed is that my flow meter is now showing I get less flow at the same pump rpms as before. When I first got it going, I was getting 322 l/h at full tilt at around 4500 rpms. Recently, I ran it wide open on the pumps and had the same rpms, close to 4500, and I was only getting in the 275 l/h range, which is right at one gallon per minute. I then pressed the spring loaded air fitting I have on top of the reservoir to let the pressure equalize and I could can see the fluid level drop about a centimeter in the reservoir. Note that it will do this under load. When not running hard, the opposite happens, the fluid level will rise about a centimeter. Anyway, after pressing it with the pumps at full tilt, the flow meter showed me getting right at 300 l/h. It makes sense that pressure changes can change the flow rate reported because the flow meter uses pressure as opposed to an impeller to measure flow. However, it's still 22 l/h less than what I was getting. I'm guessing it's not a big thing in the grand scheme of things. Obviously, the pumps are strong enough to compress the water a bit when allowing more air into the reservoir. I guess it's possible that temperature of the air and the coolant will also effect the pressure in the loop, changing what the flow meter shows. Seems like it can swing pretty far in either direction, from 275 l/h up to 320 l/h at the same settings depending on air pressure and temperature. I didn't consider that when choosing my flow meter.

On another topic, I am going to go ahead and change fans in the system. I was all set on just replacing everything with Noctua NF A25x12 fans, but on paper, they are lower static pressure than the LL120s I have that max out at 3.00mm, albeit with much less noise. However, I'm now looking at NF-F12 Industrial 2000 or 3000 rpm models. I know that the 3000 will be much louder at full tilt, but I would limit the top end to avoid that. Thoughts? Other radiator fans I should look at? Research I've done so far is all over the place, but the NF-F12 Industrials always seem to be near the top. I may start a new topic on this.
 
Phanteks T30 should be better than the NF-A12x25 PWM once you get to somewhere between 1300 and 1600 RPM (depending in the setup). The T30 fans have a 3000RPM mode as well. The NF-A12x25 PWM is probably the best radiator fan for low noise (sub 1200RPM). Disadvantage of the T30 is that it is 30mm thick instead of 25mm and the noise type is slightly higher pitched than the NF-A12x25 but as a case fan they are superior and they are also better at higher RPMs on radiators. EK-vardar would be a step down from the NF-A12x25 PWMs.
 
evhvis said:
Phanteks T30 should be better than the NF-A12x25 PWM once you get to somewhere between 1300 and 1600 RPM (depending in the setup). The T30 fans have a 3000RPM mode as well. The NF-A12x25 PWM is probably the best radiator fan for low noise (sub 1200RPM). Disadvantage of the T30 is that it is 30mm thick instead of 25mm and the noise type is slightly higher pitched than the NF-A12x25 but as a case fan they are superior and they are also better at higher RPMs on radiators. EK-vardar would be a step down from the NF-A12x25 PWMs.
Click to expand...
I started a new thread on fans here https://hardforum.com/threads/which...ghfan-12-or-turbo-12.2018716/#post-1045321111 Someone else pointed out the T30s that I didn't know about. I don't think they'll fit on the 360 in my computer though. I did find some cool stuff. I didn't know the Gentle Typhoons were still available.
 
I haven't read the whole thread, but I don't think that you're pushing too loop hard. I have Corsair 7 series rads, a 360 and a 240, with a TechN cpu block and EK gpu block on my 300w bios 2080Ti FE. Using all white 120LL fans as well. Pump should always be at 100% all the time. Very warm air coming out of my case, but gpu stays at 2000/8000, cpu stays all core 4.6 with mem at 3600. 3900XT stayed 4.4 all core.
PBO is garbage, stop using it. Keep Ryzen Master installed. But disable PBO. Start using Ryzen ClockTuner. I literally dropped 18C on my previous 3900XT and 20C on my current 5800X. It's from 1usmus, same guy that made the the ryzen dram tool we all know and love.
20220425_071337.jpg
 
I had a 5950x with a 3080FE and then a 3080ti FTW3 on 2 x Corsair 360mm XR7, with my water temp around 31-35, my 5950x (pulling up to 250W) would hit around 85c.
 
alxlwson said:
I haven't read the whole thread, but I don't think that you're pushing too loop hard. I have Corsair 7 series rads, a 360 and a 240, with a TechN cpu block and EK gpu block on my 300w bios 2080Ti FE. Using all white 120LL fans as well. Pump should always be at 100% all the time. Very warm air coming out of my case, but gpu stays at 2000/8000, cpu stays all core 4.6 with mem at 3600. 3900XT stayed 4.4 all core.
PBO is garbage, stop using it. Keep Ryzen Master installed. But disable PBO. Start using Ryzen ClockTuner. I literally dropped 18C on my previous 3900XT and 20C on my current 5800X. It's from 1usmus, same guy that made the the ryzen dram tool we all know and love.
View attachment 467124
Click to expand...
Man, that's bright.

I actually got it all together last night except for cable management. I have 3 radiators and 15 fans in my case now. I'd guess I'm probably one of the only people out there with 15 fans in a PC-O11 Dynamic. I'm glad I didn't take a picture yet as it just hit me in the middle of the night that I mounted my bottom exhaust fans upside down. So, every single fan in the case is blowing out at this moment. It's sucking air in through all the cracks and the front panel which is like honeycomb around the radiator that is mounted on it. It did an impressive job cooling when I was testing with Heaven benchmark considering.
 
This is a re-post from my thread asking for help picking out new fans. 13 fans in the front:

View attachment 469357

But wait, there's more....

View attachment 469358

Two on the back of the 240 rad in the back.

The bottom of the case has 3 Arctic P12 intakes, properly installed this time. The XSPC TX240 in the front has 2 Arctic P12 push fans. The Corsair R7 240 has 2 Phanteks T30 push fans and 2 Arctic P12 pull fans. The Corsair R7 360 has 3 Arctic P12 push fans and 3 Phanteks T30 pull fans.

I'm pleased with the cooling and sound profile. I set the max fan speed for all fans to 1500 rpm in aquasuite. Here's my fan curve, it maxes the fans at 11 degrees water temp to ambient air temp delta:

View attachment 469368

I ran P95 and Heaven together and let it go for 30 minutes. Water-air delta maxed out at 9.6 degrees Celsius. Fan speed maxed at 1450 rpm which was perfectly acceptable sound-wise, much better than the LL120s that were replaced. GPU maxed at 54 degrees and cpu (Tctl/Tdie) maxed at at 68.1 degrees. That was with an ambient of 22 degrees C, so GPU 32 over ambient and cpu 46 over ambient. Now, I'm sure when I run MS Flight Simulator, it will get hotter and a little louder, but I haven't done it yet. I may up my pump speed. It barely got to one gallon per minute under the test I did.
 
kamikazi said:
This is a re-post from my thread asking for help picking out new fans. 13 fans in the front:

View attachment 469357

But wait, there's more....

View attachment 469358

Two on the back of the 240 rad in the back.

The bottom of the case has 3 Arctic P12 intakes, properly installed this time. The XSPC TX240 in the front has 2 Arctic P12 push fans. The Corsair R7 240 has 2 Phanteks T30 push fans and 2 Arctic P12 pull fans. The Corsair R7 360 has 3 Arctic P12 push fans and 3 Phanteks T30 pull fans.

I'm pleased with the cooling and sound profile. I set the max fan speed for all fans to 1500 rpm in aquasuite. Here's my fan curve, it maxes the fans at 11 degrees water temp to ambient air temp delta:

View attachment 469368

I ran P95 and Heaven together and let it go for 30 minutes. Water-air delta maxed out at 9.6 degrees Celsius. Fan speed maxed at 1450 rpm which was perfectly acceptable sound-wise, much better than the LL120s that were replaced. GPU maxed at 54 degrees and cpu (Tctl/Tdie) maxed at at 68.1 degrees. That was with an ambient of 22 degrees C, so GPU 32 over ambient and cpu 46 over ambient. Now, I'm sure when I run MS Flight Simulator, it will get hotter and a little louder, but I haven't done it yet. I may up my pump speed. It barely got to one gallon per minute under the test I did.
Click to expand...

Pumps should also run max speed.
 
You must log in or register to reply here.
Back
Top