SFF.Network Streacom Launches the DA2
- Thread starter confusis
- Start date
I have an Evolv as well! The less SFF ever had!Thank you for update! Watercooling is a lot of work. I feel your pain
Is there enough room to fit in a QDC on one of the soft tubing runs? That way you can fill your loop (less painfully) with a reservoir, and disconnect the reservoir afterwards.
For example, I installed Alphacool QDCon my Evolv Shift, WIP picture:
Completed picture:
By the way, I do not recommend the Alphacool ones, as one of them cracked. Fortunately it wasn't in a loop when it broke. I have also tried Bitspower QDCs, but I would recommend Colder Products (CPC) quick disconnects / drybreaks instead.
I've chosen the Koolance QD3, I will fit in in the tube going from the top-left radiator's thread and the Inlet of the pump, it is a "long" tube and the correct one to have the reservoir in between.
So are load results better or worse than the results you saw on your horizontal orientated air build?
It's hard to compare apples with oranges as we are using different CPUs, but the best results I had on air where with the cooler oriented horizontally and fans on it configured as intake with the side fan configured as exhaust.
Small update. Flipped the back fan to intake, made a small difference to overall temps. I Like to attach a fifth fan but i don't know how without a separate fan controller. The mobo has only two fan headers... Just realized that i didn't try the pump header... don't see why wouldn't it work...
I like to say that i make small iterations with the gear I've got and hope that my small iterations give insight to you all.
I like to say that i make small iterations with the gear I've got and hope that my small iterations give insight to you all.
Could tidy up those fan cables a bit. For the 2 cooler fans, might be better just to rotate the fan clockwise 90 or 180 degrees to move the cables away from facing you.Current Setup (closed case, no side dust filter), Adjusted fan curves towards silence
So, rebuild again the system this time with a proper and good loop, helped by the Koolance QDC. What do you think about the temps? (@Nanook )
This is my "other ASUS Z390". I don't know why it is not as good as the one the water spilled on, to overclock the RAM with XMP:
I can only have with basic settings and VCCSA of 1.150 3600MHz C16 (the 3733 C17 wasn't stable)
Is this possible? Or maybe someone can help me with this! (So maybe I can use as well the ASRock). Otherwise, I will switch again to it, hoping the water damage is gone.
The temps are with MCE enabled, AVX -2, No Power limit, Thermal Limit of 100 Degrees, everything in AUTO apart the VCCIO.
I've replaced all thermal pads on VRMs and Northbridge with the Alphacool 14W/mK (Look how much is cooler now the VR temp and the PCH!!). Also, the Ram DIMMs are cooler.
Prime 95 sFFT and Heaven for more than 30 minutes.
Ambient temp 22 degrees.
I've added a Thermal sensor (the ASUS has one connector) on one thread on the pump, so I can read the water temp before to enter in the CPU block. Under load is 40 Degrees and in idle stay around 28C.
Still the problem with the GPU/Exhaust of hot air, but with the custom loop, I got higher AVX frequency and lower temps in general!
Not as good as the @Nanook one with his setup in the Ncase! Maybe is time to put the GPU under water as well?
This is my "other ASUS Z390". I don't know why it is not as good as the one the water spilled on, to overclock the RAM with XMP:
I can only have with basic settings and VCCSA of 1.150 3600MHz C16 (the 3733 C17 wasn't stable)
Is this possible? Or maybe someone can help me with this! (So maybe I can use as well the ASRock). Otherwise, I will switch again to it, hoping the water damage is gone.
The temps are with MCE enabled, AVX -2, No Power limit, Thermal Limit of 100 Degrees, everything in AUTO apart the VCCIO.
I've replaced all thermal pads on VRMs and Northbridge with the Alphacool 14W/mK (Look how much is cooler now the VR temp and the PCH!!). Also, the Ram DIMMs are cooler.
Prime 95 sFFT and Heaven for more than 30 minutes.
Ambient temp 22 degrees.
I've added a Thermal sensor (the ASUS has one connector) on one thread on the pump, so I can read the water temp before to enter in the CPU block. Under load is 40 Degrees and in idle stay around 28C.
Still the problem with the GPU/Exhaust of hot air, but with the custom loop, I got higher AVX frequency and lower temps in general!
Not as good as the @Nanook one with his setup in the Ncase! Maybe is time to put the GPU under water as well?
Last edited:
Great work! And yes, full watercooling is the way with the DA2 imo.So, rebuild again the system this time with a proper and good loop, helped by the Koolance QDC. What do you think about the temps? (@Nanook )
This is my "other ASUS Z390". I don't know why it is not as good as the one the water spilled on, to overclock the RAM with XMP:
I can only have with basic settings and VCCSA of 1.150 3600MHz C16 (the 3733 C17 wasn't stable)
Is this possible? Or maybe someone can help me with this! (So maybe I can use as well the ASRock). Otherwise, I will switch again to it, hoping the water damage is gone.
The temps are with MCE enabled, AVX -2, No Power limit, Thermal Limit of 100 Degrees, everything in AUTO apart the VCCIO.
I've replaced all thermal pads on VRMs and Northbridge with the Alphacool 14W/mK (Look how much is cooler now the VR temp and the PCH!!). Also, the Ram DIMMs are cooler.
Prime 95 sFFT and Heaven for more than 30 minutes.
Ambient temp 22 degrees.
I've added a Thermal sensor (the ASUS has one connector) on one thread on the pump, so I can read the water temp before to enter in the CPU block. Under load is 40 Degrees and in idle stay around 28C.
Still the problem with the GPU/Exhaust of hot air, but with the custom loop, I got higher AVX frequency and lower temps in general!
Not as good as the @Nanook one with his setup in the Ncase! Maybe is time to put the GPU under water as well?
Nice! Amazing with how fast you put together the new setup, and I think your temps are great. The Titan dumps out a tremendous amount of heat! What’s your main method of exhaust at the moment?So, rebuild again the system this time with a proper and good loop, helped by the Koolance QDC. What do you think about the temps? (@Nanook )
This is my "other ASUS Z390". I don't know why it is not as good as the one the water spilled on, to overclock the RAM with XMP:
I can only have with basic settings and VCCSA of 1.150 3600MHz C16 (the 3733 C17 wasn't stable)
Is this possible? Or maybe someone can help me with this! (So maybe I can use as well the ASRock). Otherwise, I will switch again to it, hoping the water damage is gone.
The temps are with MCE enabled, AVX -2, No Power limit, Thermal Limit of 100 Degrees, everything in AUTO apart the VCCIO.
I've replaced all thermal pads on VRMs and Northbridge with the Alphacool 14W/mK (Look how much is cooler now the VR temp and the PCH!!). Also, the Ram DIMMs are cooler.
Prime 95 sFFT and Heaven for more than 30 minutes.
Ambient temp 22 degrees.
I've added a Thermal sensor (the ASUS has one connector) on one thread on the pump, so I can read the water temp before to enter in the CPU block. Under load is 40 Degrees and in idle stay around 28C.
Still the problem with the GPU/Exhaust of hot air, but with the custom loop, I got higher AVX frequency and lower temps in general!
Not as good as the @Nanook one with his setup in the Ncase! Maybe is time to put the GPU under water as well?
Well, my method is: "let the hot air go outside somehow, somewhere!". The 140MM left fan actually helps a lot to push air a bit to the top/outside.
With the GPU in the loop I will flip the fan as exhaust, but we'll see.
Finally, I am HAPPY!
Swapped again the "damaged" ASUS MB, put the Alphacool thermal pad, reinstalled everything, and voila!!
Ok, not immediately. I had again RAM issue. It wouldn't boot with the XMP profile at 4000MHz!!! Well, I just inverted the DIMMs from AB to BA, and it booted and posted! It's so weird, didn't know that the RAM and MB are so sensitive.
BUT, the Intel BurnTest and MemTest reported immediately errors. So, I've increased the Voltage to 1.365 for the RAM and 1.165 for the VCCSA, and everything now works!!
Maybe I've placed better as well the thermal past, because the TEMPS are slightly better!
Following the final build, and after, every test did!!
Everything in BIOS is in AUTO, MCE enabled, AVX offset 2
RAM XMP I 2x16GB Kit G.Skill @4000MHz C17 17 37, 1.365V
VCCSA manual at 1.165V
Ambient temp of 22 Degrees
I've tighten a bit the inlet tube to the CPU, so it is not touching anymore the GPU backplate, I was worried for the tube+GPU high temps!
Here you can see the threaded sensor in place
And the test with temps!
Intel BurnTest Very High settings.
Prime95 sFFT for more than 30 minutes
Prime95 sFFT + Unigine Heaven for more than 30 minutes
AIDA64 for more than 30 minutes (the fans here never went 100% BTW!)
And the system's temp in idle after 2 minutes that AIDA was stopped
Impressive effort taming the 9900K. Great component and coolant temps. If you put the Titan under water, the coolant temps would be raise... but it looks like you’ve got room for coolant temperature to increase.Well, my method is: "let the hot air go outside somehow, somewhere!". The 140MM left fan actually helps a lot to push air a bit to the top/outside.
With the GPU in the loop I will flip the fan as exhaust, but we'll see.
Finally, I am HAPPY!
Swapped again the "damaged" ASUS MB, put the Alphacool thermal pad, reinstalled everything, and voila!!
Ok, not immediately. I had again RAM issue. It wouldn't boot with the XMP profile at 4000MHz!!! Well, I just inverted the DIMMs from AB to BA, and it booted and posted! It's so weird, didn't know that the RAM and MB are so sensitive.
BUT, the Intel BurnTest and MemTest reported immediately errors. So, I've increased the Voltage to 1.365 for the RAM and 1.165 for the VCCSA, and everything now works!!
Maybe I've placed better as well the thermal past, because the TEMPS are slightly better!
Following the final build, and after, every test did!!
Everything in BIOS is in AUTO, MCE enabled, AVX offset 2
RAM XMP I 2x16GB Kit G.Skill @4000MHz C17 17 37, 1.365V
VCCSA manual at 1.165V
Ambient temp of 22 Degrees
I've tighten a bit the inlet tube to the CPU, so it is not touching anymore the GPU backplate, I was worried for the tube+GPU high temps!
Here you can see the threaded sensor in place
And the test with temps!
Intel BurnTest Very High settings.
Prime95 sFFT for more than 30 minutes
Prime95 sFFT + Unigine Heaven for more than 30 minutes
AIDA64 for more than 30 minutes (the fans here never went 100% BTW!)
And the system's temp in idle after 2 minutes that AIDA was stopped
Actually, I really don't know! And I think is just for the left fan. Wait, I have the bottom and top filter removed! So the air comes out a bit from there.
Small update.
After all those tests, I've just open Chrome, 1 minute, and the PC froze. So I restart, and again. Then, I've lowered the RAM speed, and again, freeze. While I was checking the BIOS, the screen has started glitching. SO, I run a proper MemTest86, and after 3 minutes, the error has started to appear. So, I did the test at JEDEC speed, and it passes. I've started to overclock the RAM manually, and after 3200MHZ, the test gave me errors.
So, I've put my 32gb Kit 3200Mhz C14, and the test passed. I've started to overclock this RAM, and it passes the MemTest, 2 full passes with ZERO errors at 3866MHz C18 18 38, 1.38V, VCCIO 1.15V, VCCSA 1.2V.
NOW is stable. I've RMA the 4000MHz Kit, after all these days I'm pretty sure it is faulty!
I redid a Prime sFFT+Heaven for more than 30 minutes, and I confirm the previous temp! (some degrees of difference wherever, but is normal.)
Fan Profile "standard" in BIOS.
(@ShaMana the good motherboard, SSD and VRM temp I'm pretty sure are related to the fan blowing air on them, and It's a Vardar EVO ER, so till 2000RPM under stress. In your config, the FAN is on the right side and an exhausting, maybe this is the reason!)
After all those tests, I've just open Chrome, 1 minute, and the PC froze. So I restart, and again. Then, I've lowered the RAM speed, and again, freeze. While I was checking the BIOS, the screen has started glitching. SO, I run a proper MemTest86, and after 3 minutes, the error has started to appear. So, I did the test at JEDEC speed, and it passes. I've started to overclock the RAM manually, and after 3200MHZ, the test gave me errors.
So, I've put my 32gb Kit 3200Mhz C14, and the test passed. I've started to overclock this RAM, and it passes the MemTest, 2 full passes with ZERO errors at 3866MHz C18 18 38, 1.38V, VCCIO 1.15V, VCCSA 1.2V.
NOW is stable. I've RMA the 4000MHz Kit, after all these days I'm pretty sure it is faulty!
I redid a Prime sFFT+Heaven for more than 30 minutes, and I confirm the previous temp! (some degrees of difference wherever, but is normal.)
Fan Profile "standard" in BIOS.
(@ShaMana the good motherboard, SSD and VRM temp I'm pretty sure are related to the fan blowing air on them, and It's a Vardar EVO ER, so till 2000RPM under stress. In your config, the FAN is on the right side and an exhausting, maybe this is the reason!)
@fabio The fan I use blows hot air directly at the back heating the back panel and the remaining air remains trapped inside. The CPU fan blows upwards directly at the top curved panel heating that. At some point the whole case becomes heating in top and back side panels.
It's entirely possible more air escapes through the top and bottom. I've got the filters on and they do block a fair amount of flow. We've got the same mobo and m.2 drive and our cards provide a fair bit of heat so the filter do seem like a logical difference.
I'm thinking to reorient the CPU cooler horizontal and force the air out through the back panel. I'll avoid heating the top area at least. Maybe put a 14 on top as intake if possible.
But @fabio you've got a great setup. 9900k in that case is a feat on it's own, i didn't dare do it. Well done.
It's entirely possible more air escapes through the top and bottom. I've got the filters on and they do block a fair amount of flow. We've got the same mobo and m.2 drive and our cards provide a fair bit of heat so the filter do seem like a logical difference.
I'm thinking to reorient the CPU cooler horizontal and force the air out through the back panel. I'll avoid heating the top area at least. Maybe put a 14 on top as intake if possible.
But @fabio you've got a great setup. 9900k in that case is a feat on it's own, i didn't dare do it. Well done.
Those filters are too much restrictive.
For the fan on top, one 120mm fits for sure! At least is what the manual say.
About the 9900K, actually is a hot CPU without limits, but not so much compared to the AMD 2700X. I have it, and it's really not easy to keep it cool and stable at 4.1GHz! Of course the power consumption is higher, but for example, to me happen only during long sessions of rendering. Otherwise is way cooler than the AMD counterpart.
I am trying to use it in a DAN case with only a 120mm rad, but without limits, it goes immediately to 100degrees. I think limiting at 160W and some AVX 4 offset it stay "cool" as well! Is a great chip, expensive, but great!
Try what are you proposing, but I remember, actually, that to me the rear fan wasn't so helpful! The top exhausting/intaking will be so much better IMO!
I haven’t touched the ring clock on my build. Pretty happy with where it’s at right now. I did some cpu rendering/processing last night, and was amazed at how fast the 9900K is.
Cool! Thanks man!I haven’t touched the ring clock on my build. Pretty happy with where it’s at right now. I did some cpu rendering/processing last night, and was amazed at how fast the 9900K is.
BTW, in software single threaded is even better!
After some tweaking I was able to get my temps to a more reasonable level.
2700x on Asus x470-I
3.9GHz @ 1.175v - 72.8˚ CPU (tdie) after 20 min (Prime 95 sFFT) with GPU idle
3.9GHz @ 1.175v - 81.5˚ CPU (tdie) after 20 min (Prime 95 sFFT) with GPU max load (73˚)
Streacom support reported the tempered glass side panel shown off last year at Computex may be coming out as soon as April. In anticipation of that release I have modified my setup. Granted, it is not necessarily the most ideal layout for cooling but aesthetics are a big component of my build. Exact position of the PSU, cable management, and a GPU sag remedy are all outstanding items.
2700x on Asus x470-I
3.9GHz @ 1.175v - 72.8˚ CPU (tdie) after 20 min (Prime 95 sFFT) with GPU idle
3.9GHz @ 1.175v - 81.5˚ CPU (tdie) after 20 min (Prime 95 sFFT) with GPU max load (73˚)
Streacom support reported the tempered glass side panel shown off last year at Computex may be coming out as soon as April. In anticipation of that release I have modified my setup. Granted, it is not necessarily the most ideal layout for cooling but aesthetics are a big component of my build. Exact position of the PSU, cable management, and a GPU sag remedy are all outstanding items.
Similar threads
