• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.
  • The forums have been upgraded with support for dark mode. By default it will follow the setting on your system/browser. You may override it by scrolling to the end of the page and clicking the gears icon.

Issues with RTX4090 FE waterblock (Bitspower)

Joined
Feb 3, 2012
Messages
204 (0.04/day)
Location
Medina, Ohio
System Name Daily driver
Processor i9 13900k
Motherboard Z690 Aorus Master
Cooling Custom loop
Memory 2x16 GB GSkill DDR5 @ 6000
Video Card(s) RTX4090 FE
Storage 2x 2TB 990 Pro SSD 1x 2TB 970 evo SSD, 1x 4TB HDD
Display(s) LG 32" 2560x1440
Case Fractal Design Meshify 2 XL
Audio Device(s) onboard
Power Supply beQuiet Dark Power 12 1000W
Mouse Razer Death adder
Keyboard Razer blackwidow v3
VR HMD n/a
Software Windows 11 pro
Benchmark Scores Heaven 4.0 @ 2560x1440 270.5 FPS
So about 2 years ago Microcenter had this 4090 block on sale for $50. Being bitspower, I figured it would be OK, but the deal seemed to good to be true.
Since I put it in my loop I have had issues with corrosion and green floating stuff (not even my main concern and has been mostly mitigated at this point)
the bigger issue to me is the hotspot temps. It seems like I have to reseat the card every 2-3 months. Temps will be great (45-50* core and 60-65* hot spot) but then I will check the temps after a while, today my core was 55-60* and hotspot was almost 100*C. Im getting sick of taking the loop apart to reseat this block. in the 15 years I have been watercooling GPUs, I have never seen this before.
Anyone have some experience water cooling the 4090 FE? Does this occur with all blocks? I will drop some $$ and buy an alphacool or EK (ugh) block right now if I dont have to deal with this **** anymore.
 
Last edited:
Corrosion and bio growth should be your concern as they can and will destroy your loop.

Corsair makes good and cheap blocks, so does AC.

Optimus if you want to splurge.
 
I mean i can drain the loop and refill in a few seconds. I have that issue mostly dealt with I just had to change up my additives a bit.
 
Picked up some TF9. Took the loop apart. Reseated the GPU block, CPU block, and disassembled both to clean the corrosion out. I would say the TIM was definitely pumped out of the GPU block. Not sure what cause the corrosion issues last year, but it was a thin green layer, i was able to scrape it off with my finger nail. vertical mounted the GPU while i was at it. hope the riser doesnt cause me any issues. 5 straight hours of back breaking work.

Resized_20250327_144220_1743100954259.jpeg
Resized_20250327_103833_1743100898066.jpeg
Resized_20250327_103829_1743100888891.jpeg
Resized_20250327_103822_1743100882353.jpeg
Resized_20250327_102010_1743100874946.jpeg
 
Looks like plastisizer. That might be from the tubing.
 
Use PTM7950.

The green stuff is probably residue build up from the liquid itself, just use a clear premixed coolant.

Looks like plastisizer. That might be from the tubing.
I don't think it's plasticizer, it takes ages for that stuff to build up.
 
Id had this issue before. Ended up being the bras fittings weren't really brass. Have to watch out everything these days...

In your case it doesn't look like corrosion at all, just debris from the tube or additives.
 
Last edited:
I use distilled water and algicide. After these issues I also added some corrosion inhibitor. ive had pretty squeaky clean loops for the last 15 years until either this bitspower block or the corsair rad.
 
Use PTM7950.

The green stuff is probably residue build up from the liquid itself, just use a clear premixed coolant.


I don't think it's plasticizer, it takes ages for that stuff to build up.
Was looking at it on the phone. Now that I'm in front of a normal monitor, think I may be mistaken, really hard to tell honestly.

I use distilled water and algicide. After these issues I also added some corrosion inhibitor. ive had pretty squeaky clean loops for the last 15 years until either this bitspower block or the corsair rad.
Is the radiator aluminum? If it is, I would ditch it and find a copper one instead/
 
I just found this thread after seeing the other one about pump-out. Just for reference, this is the block I have: HEATKILLER V PRO for RTX 4090 FE.

I never see hotspots more than ~10C over the main GPU temp.

As for coolant, I first tried EK's pre-mix (clear) and it was performing poorly and leaving a residue on everything. I used Mayhems blitz system cleaner to flush and clean the whole system, then switched to Koolance 702 Clear. I had been using it at work for a while at that point and decided to use it at home too as it has been working great. So far so good with it. I have a loop at work I've been running (non-computer load, but plenty of heat pumping through a MO-RA3 and some cold blocks) for a few years now and there's no sign of any corrosion or issues at all. I'm going to change it out soon in that loop, but the statement stands: unless something changes, I'm very happy with this coolant and it's so simple I don't have to add anything or worry about it for years at a time. They rate it for 2-3 years before replacement is recommended too. Easy peasy.
 
Interesting. mine is a bitspower 4090FE block, not sure of the exact model. Now that everything is back together, my hotspot deltas are almost exactly 10*C as well. GPU temps never go above 53*C.
 
Distilled water only with brass and copper is the only way to go. Any additives just ruins these loops.

I don't know what all goes into it, but I've been using EK Cryofuel clear for 1.75 years with no corrosion or any growth in my loop with brass and copper. I am going to use it again once I finish my cooling upgrade.
 
I use distilled water and algicide. After these issues I also added some corrosion inhibitor. ive had pretty squeaky clean loops for the last 15 years until either this bitspower block or the corsair rad.
Ditch the damn thing then
 
Mixing metals in a Loop will cause issue’s, this is the cause. Plus all the coolants, that don’t cool or ain’t cool. I use anti freeze, never had a mixed metals issue but I have seen n read the stories(on here I think)……
There is a word for it, where you mix cooper and metal. You may need to separate the GPU loop and the CPU loop, unless you get the same company blocks.
You can clean the loop all you like, it will continue to happen!
 
I don't think it's plasticizer, it takes ages for that stuff to build up.
Looking back when I was using EK DuraClear tubes I'm pretty sure plasticizer build up occurred in as soon as a year of use. When I did a teardown and inspection of my distroplate whiteish residue (also inline with the clear tubes turning whiteish/yellowish during that time period) was lining all my gaskets in my distroplate. Since I changed to Primochill such residue hasn't built up in 2 yrs since.

Distilled water only with brass and copper is the only way to go. Any additives just ruins these loops.
I'm a bit wary of not having additives (using a premix solution) but I've been running an experiment and letting my EK Cryofuel run a year past it's expiration date when I'd normally change it and topping it off only with distilled when needed. So far everything seems to be fine after inspecting my blocks a few months ago and I'm due for another inspection coming up in next month which will complete the 1yr expired mark. I'm a bit undecided if I'll reup the EK Cryofuel, exchange all the fluid with distilled, or continue the experiment simply topping off the loop with distilled. My loop is all copper/nickel. Do you ever swap out your distilled water? If so how often? Is this because of potential ionization or PH changes in the water over time? Do you use any additional grounding in your loop from the blocks or fittings?

I use distilled water and algicide. After these issues I also added some corrosion inhibitor. ive had pretty squeaky clean loops for the last 15 years until either this bitspower block or the corsair rad.
Did you add the corsair rad at or near the same time as the block? The color of the residue makes me think oxidized copper is present in the residue. When using distilled are you using a "fresh" sealed bottle or has it been open for awhile or was exposed to the air for a long period of time? How did you prepare the rad before adding it to your loop?
 
Last edited:
Looking back when I was using EK DuraClear tubes I'm pretty sure plasticizer build up occurred in as soon as a year of use. When I did a teardown and inspection of my distroplate whiteish residue (also inline with the clear tubes turning whiteish/yellowish during that time period) was lining all my gaskets in my distroplate.
What this guy has doesn't look at all close to what you had, that's why I said it would take ages for it to build up in chunks like that.
 
What this guy has doesn't look at all close to what you had, that's why I said it would take ages for it to build up in chunks like that.
Yea that is a lot of build up. Something certainly fouled that loop. Not knowing if it biological I'd do a full breakdown and cleaning of the entire loop, sanitization of the rads, breaking down the blocks ensuring cleaning of the gaskets and channels incase something is hiding there. This is a situation where I would preserve a sample in a container and see if it grows to try to understand if it's biological.
 
Mixing metals in a Loop will cause issue’s, this is the cause. Plus all the coolants, that don’t cool or ain’t cool. I use anti freeze, never had a mixed metals issue but I have seen n read the stories(on here I think)……
There is a word for it, where you mix cooper and metal. You may need to separate the GPU loop and the CPU loop, unless you get the same company blocks.
You can clean the loop all you like, it will continue to happen!
I have used copper rads and CPU blocks with nickel blocks for years with no issues. The only mystery here is what is in the corsair XR7 radiator.
Looking back when I was using EK DuraClear tubes I'm pretty sure plasticizer build up occurred in as soon as a year of use. When I did a teardown and inspection of my distroplate whiteish residue (also inline with the clear tubes turning whiteish/yellowish during that time period) was lining all my gaskets in my distroplate. Since I changed to Primochill such residue hasn't built up in 2 yrs since.


I'm a bit wary of not having additives (using a premix solution) but I've been running an experiment and letting my EK Cryofuel run a year past it's expiration date when I'd normally change it and topping it off only with distilled when needed. So far everything seems to be fine after inspecting my blocks a few months ago and I'm due for another inspection coming up in next month which will complete the 1yr expired mark. I'm a bit undecided if I'll reup the EK Cryofuel, exchange all the fluid with distilled, or continue the experiment simply topping off the loop with distilled. My loop is all copper/nickel. Do you ever swap out your distilled water? If so how often? Is this because of potential ionization or PH changes in the water over time? Do you use any additional grounding in your loop from the blocks or fittings?


Did you add the corsair rad at or near the same time as the block? The color of the residue makes me think oxidized copper is present in the residue. When using distilled are you using a "fresh" sealed bottle or has it been open for awhile or was exposed to the air for a long period of time? How did you prepare the rad before adding it to your loop?

I added the corsair rad at the exact same time as this block. both were on a fire sale and it was a whim purchase. did 5 minutes of research.
I dont think the issue has continued after the first few times I caught it (about a year ago). I removed my kill coil, and used different additives. (I was using my original 15 year old bottle of biocide) I just never bothered to scrub the block until this week.
 
I have used copper rads and CPU blocks with nickel blocks for years with no issues. The only mystery here is what is in the corsair XR7 radiator.


I added the corsair rad at the exact same time as this block. both were on a fire sale and it was a whim purchase. did 5 minutes of research.
I dont think the issue has continued after the first few times I caught it (about a year ago). I removed my kill coil, and used different additives. (I was using my original 15 year old bottle of biocide) I just never bothered to scrub the block until this week.
Ah ok. I highly suspect the rad released a butt load of flux (sorry for the insane imagery just now) with oxidized copper to explain the color in the absence of knowing if a colored fluid was used prior.
 
You guys will love this. I had a Mining rig with 2 6800XTs. It had 2 pumps and 2 rads. The 2nd pump was a cheaper Chinese unit. One day it burst a seal and exploded. I removed it from the loop and things were good. About a month ago I found a new algorithm that made Mining more viable. I restarted that PC and noticed that one of my cards was running warmer than it should. When I took the first card out I saw it. There was clearly damage. It seems the explosion of the pump put plastic bits into the loop. That restricted the flow and caused the GPU to scald the Block. GPU cooling is not for the uninformed. I wish I had one of those Electrolysis cleaners like what some of the bigger YT channels have.
 

Attachments

  • IMG_20250329_143332684.jpg
    IMG_20250329_143332684.jpg
    1.1 MB · Views: 29
Yes. WC is not for the faint of heart. It was not in my plans to spend a nice spring day in Ohio inside tearing my loop apart. diligence is key.
 
I took a screenshot from HWiNFO64 after a few hours of playing 7 Days To Die (4k, max settings...not an RT workload, but it still works fairly hard). I do have the GPU running at stock today:
1743298609208.png

Edit: just to sum up for those who don't want to go looking: max was ~10° hotter, but average was 6.5° hotter on the hot spot.
 
Last edited:
Back
Top