I can't wait for that hype train to crash and burn so we can get heaps of cheap GPUs flooding onto the market. I'm sick of not being able to build anything. I know I know, wrong place for that discussion. Please ignore lol
RTX A2000 Full Copper Low Profile Single Slot Edition - aka A2000 SpongeBob Copperpants Mod
- Thread starter REVOCCASES
- Start date
I'm the one who derailed the thread
Back on topic, that just show how unimpressive Nvidia thermal solution is. I wish they had use some kind of heatpipe / vapour chamber with fin arrays with that blower ... something like that HD6950 (fin array + vapour chamber)
Even worse, it looks like it's what they are using on the RTX4000 (with flat heatpipes) and RTX A4000 (with vapour chamber)
Last edited:
Thought this might be interesting for some of you...
Very nice, I'm currently listening to it.
Do you think it would work in 2 slots thickness with the stock blower fan ? I'm wondering if the fan will provide enough air flow.
While searching for extra thick copper foam, I found a prototype of a fanless i7-4785T cooled with copper foam :
Yes I'd like to give it a try. I just sent you a note on your website contact form. Thanks!
Really cool ramyun noodle heatspreader and very professional job on this 
We were talking about this one a while back, the image shows the 'SilentPower' prototype (basically this image, and some vague specifications with very little follow-up) which was used to scam people into crowdfunding (outside an official channel, suss much?)Very nice, I'm currently listening to it.
Do you think it would work in 2 slots thickness with the stock blower fan ? I'm wondering if the fan will provide enough air flow.
While searching for extra thick copper foam, I found a prototype of a fanless i7-4785T cooled with copper foam :
Behold! The Most Insane Crowdfunding Campaign Ever | Hackaday
FYI, ignore Brian's comments about the heat radiation of copper / foamed copper versus aluminium etc., as direct radiation is only a tiny fraction of the dissipation ability of a heatsink. (He mentions aluminium being a better radiator than copper, and that everything should be black, but as we already know from our SFF experiences, those aren't the critical factors involved)
Heat transfer occurs via Conduction, Convection and Radiation.
Conduction:
Copper [~401W/mK] is a far better heat conductor than aluminium [~237W/mK], which is actually why it's so great for heatsinks and waterblocks. But it's because it allows the heat to conduct from the source (CPU/GPU/etc.) to the heatsink. (Or to the coldplate, then coldplate to the coolant/fluid, in the case of watercooling).
Given a heatsink is strapped to a heat-source (CPU/GPU/whatever) and not much else, heat conduction from the heatsink is approximately nil, as air itself is a very poor heat conductor.
Conductive Heat Transfer (engineeringtoolbox.com)
Convection:
This is the way most heat transfer will occur from a heatsink.
Air flow / current plays a significant role in removing heat from the heatsink, which is rather evident heatsink designs, even more so passive designs as they are HUGE and have large gaps to take advantage of ambient airflow. Even the smallest amount of forced air flow makes a massive difference to performance, too!
Convective Heat Transfer (engineeringtoolbox.com)
Radiation:
Pure heat radiation is by far the least significant way a heatsink will transfer heat away from itself.
Yes, aluminium has slightly higher surface emissivity than copper (although both have relatively low surface emissivity) and black materials would typically radiate heat more effectively than other colours. However the amount of energy that is lost via radiation is so truly miniscule compared to convection that it's a moot point.
Not to mention if you make something out of aluminium vs. copper, or paint something black with the hope of radiating more heat, you're likely to worsen overall performance, by losing conductivity from the heat source, or adding a material that's ends up functioning as insulation impacting the convection performance.
Radiation Heat Transfer (engineeringtoolbox.com)
Of course, in a vacuum or other environment, things would be very different. But thankfully, we're not, cause.. air is kinda important for us.
Emissivity Coefficients common Products (engineeringtoolbox.com)
For what it's worth, here's some comparison numbers, I'm sure we wouldn't suggest coating a heatsink in anything here with a higher emissivity...
| Surface Material | Emissivity Coefficient - ε - |
|---|---|
| Aluminum Highly Polished | 0.039 - 0.057 |
| Aluminum Anodized | 0.77 |
| Black Body Matt | 1.00 |
| Copper electroplated | 0.03 |
| Copper Polished | 0.023 - 0.052 |
| Copper Nickel Alloy, polished | 0.059 |
| Paint | 0.96 |
| Polytetrafluoroethylene (PTFE) | 0.92 |
| PVC | 0.91 - 0.93 |
| Rubber, hard glossy plate | 0.94 |
| Steel Oxidized | 0.79 |
| Steel Polished | 0.07 |
| Stainless Steel, polished | 0.075 |
| Wood Oak, planed | 0.885 |
| Wood, Pine | 0.95 |
The stock blower seems to have decent CFM and air pressure so I think that would work. Probably better than single slot thickness but not sure by how much.
I am currently experimenting to make better thermal contact between the copper shroud, copper plate and the sponge so that it can "soak up" the heat from all sides.
Might depend more on how much you can direct the airflow tbh, more than the size of the foamed copper sink. But I haven't tested it so you're better placed to speculateThe stock blower seems to have decent CFM and air pressure so I think that would work. Probably better than single slot thickness but not sure by how much.
I am currently experimenting to make better thermal contact between the copper shroud, copper plate and the sponge so that it can "soak up" the heat from all sides.
Might depend more on how much you can direct the airflow tbh, more than the size of the foamed copper sink. But I haven't tested it so you're better placed to speculate
probably you would have to put the fan at the end of the card and make sure it (only) blows all the way through the foam.
btw, I've started working on a new cooler design without foam...
Last edited:
Now that is interesting. I was looking around for a big fin stack so I can deshroud mine and cool it with case fans.probably you would have to put the fan at the end of the card and make sure it (only) blows all the way through the foam.
btw, I've started working on a new cooler design without foam...
btw, I've started working on a new cooler design without foam...
Very nice! Looking forward to the results and comparison.
I've ordered an extruded aluminium heatsink of 60x150x20 that I will combine with two 60x25 fans soon. I may compare it to a copper skived version later but it will depend on the results I get with the aluminium version.
Only hard part will be to mill the cutouts for the caps. As I don't really have access to a proper cnc mill, I'll attempt by hand with a dremel router.
Found some time today to test the new CNC machined aluminum cooler prototype but unfortunately the turbo fan I ordered doesn't fit. The seller didn't mention any dimensions / specs but I had assumed that "5010" means 50mm diameter and 10mm thickness - same like with normal fans...
So until I can find a more suitable fan I had to use the DELTA one again. Far from perfect (e.g. it's too thick to assemble the shroud) but anyways I just wanted to see if a rather simple aluminum heatsink would/could do the job.
Stress test with the prototype aluminum heatsink, no shroud, manual 100% fan speed:
Stress test with the prototype aluminum heatsink, no shroud, AUTO fan speed:
In it's current form with the non-optimal fan and without the shroud it performs much worse compared to my Full Copper Sponge Heatsink but I'm somehow confident that a aluminum cooler like that would work OK with another fan and some design tweaks. Another approach could be using a 3mm copper plate and soldering some fins on it. Biggest challenge remaining is finding a good turbo fan with <51mm diameter and preferably <9mm thickness to get it down to a "real" 1-Slot card.
If you'd like to have a look or play around with the design, I've attached the STP files to this post.
So until I can find a more suitable fan I had to use the DELTA one again. Far from perfect (e.g. it's too thick to assemble the shroud) but anyways I just wanted to see if a rather simple aluminum heatsink would/could do the job.
Stress test with the prototype aluminum heatsink, no shroud, manual 100% fan speed:
Stress test with the prototype aluminum heatsink, no shroud, AUTO fan speed:
In it's current form with the non-optimal fan and without the shroud it performs much worse compared to my Full Copper Sponge Heatsink but I'm somehow confident that a aluminum cooler like that would work OK with another fan and some design tweaks. Another approach could be using a 3mm copper plate and soldering some fins on it. Biggest challenge remaining is finding a good turbo fan with <51mm diameter and preferably <9mm thickness to get it down to a "real" 1-Slot card.
If you'd like to have a look or play around with the design, I've attached the STP files to this post.
Shame that the fan you ordered did not fit, because the blower design is an essential match to the design of the heatsink. With the axial fan, I suspect it can't make enough pressure.
Shame indeed. I've been searching for days to find anything suitable but nothing seems to fit. Especially if I want to keep it a "true" 1-Slot card I'm limited to <9mm fan thickness - preferably 7 to 8mm.
If I have to stick with 10mm fan thickness it'll be about 2 ~ 3mm more than 1-Slot
PS: would there be interest in a CNC machined heatsink for de-shroud / fan mods?
Looking at my test results I think something like this would work well enough with two 50/60mm fans...
If I have to stick with 10mm fan thickness it'll be about 2 ~ 3mm more than 1-Slot
PS: would there be interest in a CNC machined heatsink for de-shroud / fan mods?
Looking at my test results I think something like this would work well enough with two 50/60mm fans...
There seem to be lots of slim blower fans designed for laptop cooling that are thin enough maybe like this. Im guessing these tend to come with their plastic enclosure, & not just as a bare fan, and that is the problem for you, wanting the bare fan. So you'd have to figure out how to cut the fan out of it's plastic enclosure, or alternately, keep the enclosure & design the heatsink around the shape of the fan enclosure and just incorporate the thing into the heatsink.
I've also thought about this. Only issue would be that notebook fans require 5VDC instead of 12VDC and I don't think I can find enough space somewhere to integrate a small stepdown converter in the design
