just got mine too i guess we in the same batch
 

Jinatsuu
Hey Guys, first time water Cooling and trying to figure out if I have to go through the pain of draining my loop and opening/tightening the water block some more or maybe change out the pads. My main components are in my signature as for what I have set up and water cooling parts below. Will mention the case I have is an open air case as well. 
So main thing is that this does seem abit too hot for the memory on the back but haven't really seen anyone else's numbers to compare against. I'm currently running stock clocks since I wanted to get a base line before I tried any OC. 

• EK - CoolStream Classic PE 360 x2
• K-Quantum Kinetic TBE 300 D5 PWM D-RGB - Plex
• Front and Active Back Plate for ftw3 with the revised standoffs
• x3 EK Vader Fans on both Radiators

 
GPU Temps - Tested with port royal
Idle: (GPU)27.8-30C (Mem)  36-38c (water) 29c 
Load: (GPU)57.8C (Mem)  74c (water) 37c-38c
Pump set to about 70% (still experimenting on where I want to adjust this)
Fans are set to 100% when water temp hits 35c 
(clockspeed) 1950 (memclock) 1219
 
 
This is at the 450 Watt Range ( 119% on power slider in msi after burner)
 
Edit: added wattage and clock speeds
 
Thanks for any info and tips as someone's first attempt at water cooling!

 

It's not horrible but can be a little better if you tighten the screws more after the pads are warmed up.

At 35C my memory junction temperature is at 60C or lower. My GPU temperature is 18-20C over my water.

bjohannes1984
Has anyone used liquid metal with this block? Worth the trouble?

I would like to see the results myself. This block and the HC has poor mounting pressure compared to other blocks.

It's possible liquid metal might even be a disaster since it requires good mounting pressure to work well.

bjohannes1984
Has anyone used liquid metal with this block? Worth the trouble?

Never tried liquid metal on my GPU's..  Been using KPx mostly these days.  

Yea, it was the right head this time around at least lol. Either way, seems like the screws are very brittle

oky want to give an update on my ek front and active backplate, just fired my 2x 3090 with ek front and active backplate and both video card is working. my memory temp for 1 card is 90c  and the other is 70c overall water temp is 45c

KingEngineRevUp

Jinatsuu
Alright, was indeed able to tighten some screws even though two got stripped. I turned up the pump speed as well to get the bubbles out but thought I'd try a port royal stress test as well and seems like memory Temps stabilized around the 60c to the 62c range. Pretty happy with that and will deal with the strip screws later if I need to get back in there lol.

Did you use the right Phillips size? Or over side too small?

Hey King,
I thought you might know the answer to this issue I’m having.
So to start off my GPU hotspot was all over the place a month ago, the temps would go anywhere from 30c to 40c above the gpu’s temps.
I took the card apart and reapplied the thermal paste. Now my hotspot temp is 10c above gpu temp but under load it tends to go 20 to 22c above gpu temps.
Do I need to redo the thermal paste? All other temps are normal.

Any help would be appreciated!
This is temps was while running Time spy:
 
This temp is under no load:
 
 
The Average is 15.1 over GPU temps.

talon951
Typical full load delta is 10-15C, so you're still pretty high. Probably doesn't hurt anything if it isn't throttling. May hurt OC'ing some.

Yes I agree, it’s for sure effecting any Ocing. I will probably neeed to drain and do the thermal past again. But I will use a different thermal paste this time. Instead of the generic EKWB.

Thank you Talon!

talon951
I'd try a thick paste like Gelid GC-Extreme or Kryonaut Extreme (not normal Kryonaut as it tends to degrade on gpus). I've had Kryonaut Extreme on my 3090 since the first of August with no degradation and good temps (11C hotspot delta, 15C block delta at 500w). It is a little difficult to spread though. I haven't personally used the Gelid but have seen reports of good results.

Primarily you want a thicker paste since the core tends to be slightly dome shaped. At least that's the explanation I've seen.

That's very helpful, The local store has these options:
- Thermal Grizzly Hydronaut Thermal Grease Paste 
- Cooler Master New Edition MasterGel Maker Diamond Particle Thermal Paste
- EVGA FROSTBITE 2 THERMAL GREASE
 
Which one do you think would work best, seems around the same W/mk.
 
And you spread it your self? I tried that once but didn't like the results. I usually do a dot method. 

talon951
I haven't used any of those or seen many examples of them being used on gpus,so really don't have an opinion.

The dot method could be your problem. That can work, but I think a thin spread works better for GPUs. Most of the hard core watercooling/OC'ers seem to do that. Use a little spreader like comes with some of the pastes. You want a thin layer, but that covers the entire core. I've done that several times now with good results. I think the combination of the size of the core, the slight curvature I mentioned, and relatively low contact pressure is why it works better for GPUs.

Thank you Talon! Your advice came in handy! I re applied thermal paste used MX-4 and now hot stop is 10C over ideal and 14-18C underload.
 
Having to repasted that card for the 3rd time, now I know that the EKWB paste that I got wasn't the best. It was detreating everywhere.
 
Again thank you man!

rjbarker
The pea size n rice grain size in some cases simple doesnt cut it anymore due to the size of the Die's.(rice grain size for sure, pea size you can still get un-covered corners of the die....I have changed over to spread....even "x" is better to ensure covering the entire die, just a bit messier....
 
- idle temp of GPU (3080Ti FTW3) is always within 1-2c of ambient room temp (generally 21c)
- Maximum load temp of GPU with 99% load over a few hours is 46c
- As soon as I close my GPU intensive Program it takes about 5 min for the GPU to settle from load temp 46c to room temp idle.
9*120mm Rads
D5 Vario "set n forget" at 2900 RPM
Fans "set n forget" 1450 RPM

I used a credit card to spread a thin layer of thermal paste with the last card I blocked, which is a 3080 Ti FTW3. Mine too is within 1-2° of ambient at idle, but it takes a bit longer to come down from a heavy load probably because I have less cooling space than you (two 360 radiators). I’ve got my D5 pump set to max, and fans around 1250 RPM with a ramp up to max starting at 65° and ending at 85°. This is the same setup I had with my 2080 Ti, but that card ran a bit hotter. I’m wondering if it’s because I used the “pea size” method of applying thermal paste for that, or if the 3080 Ti is more power-efficient (less heat waste).

ObscureEmpyre
Question for you guys: Do you experience what sounds like coil whine using the EK Quantum Vector FTW 3080/3090 water block and backplate? I had it with my 2080 Ti and even worse with my 3080 Ti. However, I came across a Reddit thread about how it’s caused by a couple of thermal pads on the back of the card. Since I have soft tubing, I was able to carefully remove the backplate without disassembling my loop to remove those pads as an experiment. Well, it worked. I can still hear it, but only if I put my ear right up to the case and only just barely. Before, I was able to hear it outside of the room my PC is in. Below is the link for reference and there’s an image URL in one of the comments that illustrates which pads are the culprit, although it was a 3080 XC3 in question.

https://www.reddit.com/r/...il_whine_resolved_see/

When I waterblocked my first 3090 FTW3 with the EKWB waterblock (with non active backplate, and then later with the active backplate), I heard a fair amount of coil whine. When that card died, I put its RMA replacement in my system for a while with the stock cooler. Knowing what to listen for, I could hear coil whine with the card in its stock configuration. I was having other problems, someone suggested the possibility of a power supply capacity issue, and I'd also read someplace that coil whine could be caused by a power supply, so I went ahead and replaced my Seasonic 850w PSU with a Seasonic 1300w PSU and... my coil whine stopped. When I re-waterblocked the card (with the EKWB active backplate) I have just the tiniest amount of coil whine, which could have been present but masked by the stock cooler (I don't know anyone who wasn't as intimately familiar with the system as me would hear it).

So my experience lends credence to the PSU notion, and against the bacckplate notion, as I heard the whine both in a stock configuration, and with different backplates. FWIW.

ObscureEmpyre
Question for you guys: Do you experience what sounds like coil whine using the EK Quantum Vector FTW 3080/3090 water block and backplate? I had it with my 2080 Ti and even worse with my 3080 Ti. However, I came across a Reddit thread about how it’s caused by a couple of thermal pads on the back of the card. Since I have soft tubing, I was able to carefully remove the backplate without disassembling my loop to remove those pads as an experiment. Well, it worked. I can still hear it, but only if I put my ear right up to the case and only just barely. Before, I was able to hear it outside of the room my PC is in. Below is the link for reference and there’s an image URL in one of the comments that illustrates which pads are the culprit, although it was a 3080 XC3 in question.

https://www.reddit.com/r/...il_whine_resolved_see/

Hey Obscure.....I had a couple of 780Ti's and 1080Ti's (SLi) also w EK Blocks and I think we are just more "sensitive" to the high pitched / frequency "squeal / whine" (white noise) we hear as we dont have any racket from GPU fans to drown it out....you know what I mean?
Sometimes the 3080Ti I notice it (if I really listen and focus on it) other times dont really notice it......completely bearable vs the GPU Fans imho....
Interesting though removing the pads....good find my friend!
 
I too am old school and my loop is 3/8 - 5/8 flex tubes...dont have the patience for hard tubes at this stage ;) My Loop hasnt changed (other than components and replaced the D5 Vario pump once) since 2013/14...still using the same BP fittings amd Rads ;)
Replaced the "O-Rings" in the BP fittings as well...and of course replace the tubing on average may once ever 18 months when I get off my lazy butt and flush the system or when I replace a component...I use EK clear concentrate coolant w distilled.
 
And yes I can confirm the 3080Ti FTW3 with its power draw really increases temp. For example the 3080Ti FTW3 under full load in my loop maxes at 46c...while waiting on my RMA I threw the Evga 1080Ti SC Black w EK Block back in the same loop and the highest temp it would reach was around 34c ....playing either SOTR or CP77.....same CPU (9900K) OC....so I would think your 2080 Card likely running cooler than these power beasts ;)
 
 

Jinatsuu
Hey Guys, first time water Cooling and trying to figure out if I have to go through the pain of draining my loop and opening/tightening the water block some more or maybe change out the pads. My main components are in my signature as for what I have set up and water cooling parts below. Will mention the case I have is an open air case as well. 
So main thing is that this does seem abit too hot for the memory on the back but haven't really seen anyone else's numbers to compare against. I'm currently running stock clocks since I wanted to get a base line before I tried any OC. 

• EK - CoolStream Classic PE 360 x2
• K-Quantum Kinetic TBE 300 D5 PWM D-RGB - Plex
• Front and Active Back Plate for ftw3 with the revised standoffs
• x3 EK Vader Fans on both Radiators

 
GPU Temps - Tested with port royal
Idle: (GPU)27.8-30C (Mem)  36-38c (water) 29c 
Load: (GPU)57.8C (Mem)  74c (water) 37c-38c
Pump set to about 70% (still experimenting on where I want to adjust this)
Fans are set to 100% when water temp hits 35c 
(clockspeed) 1950 (memclock) 1219
 
This is at the 450 Watt Range ( 119% on power slider in msi after burner)
 
Edit: added wattage and clock speeds
 
Thanks for any info and tips as someone's first attempt at water cooling!

That GPU load temperature is a little high for a water cooled GPU.  How much does it drop with a higher pump speed?

rjbarker

bavor

Jinatsuu
Hey Guys, first time water Cooling and trying to figure out if I have to go through the pain of draining my loop and opening/tightening the water block some more or maybe change out the pads. My main components are in my signature as for what I have set up and water cooling parts below. Will mention the case I have is an open air case as well. 
So main thing is that this does seem abit too hot for the memory on the back but haven't really seen anyone else's numbers to compare against. I'm currently running stock clocks since I wanted to get a base line before I tried any OC. 

• EK - CoolStream Classic PE 360 x2
• K-Quantum Kinetic TBE 300 D5 PWM D-RGB - Plex
• Front and Active Back Plate for ftw3 with the revised standoffs
• x3 EK Vader Fans on both Radiators

 
GPU Temps - Tested with port royal
Idle: (GPU)27.8-30C (Mem)  36-38c (water) 29c 
Load: (GPU)57.8C (Mem)  74c (water) 37c-38c
Pump set to about 70% (still experimenting on where I want to adjust this)
Fans are set to 100% when water temp hits 35c 
(clockspeed) 1950 (memclock) 1219
 
This is at the 450 Watt Range ( 119% on power slider in msi after burner)
 
Edit: added wattage and clock speeds
 
Thanks for any info and tips as someone's first attempt at water cooling!

That GPU load temperature is a little high for a water cooled GPU.  How much does it drop with a higher pump speed?

Increasing pump speed will at most decrease GPU load 2c at best !!!  Unless the OP is running his / her pump at very very low RPM's.
 
2500 RPM to 3800 RPM no difference...to lower temps you need to increase rad space and fan speed .

I have two other 3090s that I've recorded temperatures on with EKWB blocks.  Both of them didn't exceed 50C in Port Royal stress tests while drawing about 480-500 watts.  His nearly 58C GPU temps seem rather high unless he is using a really small radiator.  However he has two 360mm radiators according to the above posts.  I've seen pump speeds make significant temperature improvements in water cooled 3090s under load.