DAN A4-SFX: The smallest gaming case in the world

Does it work well? I have some 3mm black sticky back foam tape I was thinking about applying to the back of the psu too...

I realised a couple of days ago that I've been trying to optimise my CPU fan curve and stuff with prime95 but I've been totally wasting my time. The CPU is totally quiet while gaming, it's all the power supply....
I have not tryed with foad, but you can do it a test temps

I'm also interessed in this "solution".
I have the same problem since the beginning with my SilverStone SST-SX700-LPT!

What is this: corg? What is the thickness?
Is 3mm cork
home1PlacasCorcho.jpg

I test this originally to see if i can drop the ssd temps behind the motherboard, but was not posible.
However its works with the psu and, motherboard and gpu because I have two sealed sides.
 
has anybody else tried to put 2 92mm rads in this case? i knew you can fit one for cpu but i did it for the gpu too with a kraken g 12 bracket and move the psu outside the case?
 
Last edited:
has anybody else tried to put 2 92mm rads in this case? i knew you can fit one for cpu but i did it for the gpu too with a kraken g 12 bracket and move the psu outside the case?
Doesn't that just seem like an unnecessarily complicated way of dropping your gpu temps? If you want cooler GPU temps, sell your GPU and buy a similar one with a better cooler... I can almost guarantee that it'll be cheaper in the end... An asetek 545lc is 80$ alone, and then you have to get the kraken and develop some method of having the CPU outside of the case?
 
Doesn't that just seem like an unnecessarily complicated way of dropping your gpu temps? If you want cooler GPU temps, sell your GPU and buy a similar one with a better cooler... I can almost guarantee that it'll be cheaper in the end... An asetek 545lc is 80$ alone, and then you have to get the kraken and develop some method of having the CPU outside of the case?

buy a another 500+ vid card vs doing a 200 convert? i already have and its better then what i had and its not the cpu its the psu outside the case, just meant longer cords
 
Edited:

Test conditions: 25c ambient, NT-H1 and a9x14 used on both heatsinks, TDP set to 120w, fan speed set to max (2.5k), vcore 1.04, all core 4.3ghz, prime95 8k inplace FFTs for 10 mins inside dan a4sfx v2, no mods apart from delidded 8700k with CLU inside and hwmonitor was reset a few minutes into the test to clear idle values.

Test results:
L9i (bad paste coverage):
l9iqtsla.jpg


L9i retest (good paste coverage):
l9inew3fs4s.png


C7 Cu:
c7cu67sh9.jpg


C7cu
Pros: 45c ΔT
Cons: Turbulence starting from 1.3k rpm /w a9x14 and x2 the weight of alu verison.
20180424_145214ltsha.jpg
 

Attachments

  • 20180424_1416430vs8z.jpg
    20180424_1416430vs8z.jpg
    715.4 KB · Views: 1,622
  • 20180424_1543585gs75.jpg
    20180424_1543585gs75.jpg
    721.4 KB · Views: 2,102
Last edited:
My test wasn't very scientific because I had a messed up thermal paste application on the l9i during the test. Only realized after I removed it to put in the c7cu. I'm too lazy to retest the l9i with better thermal paste coverage, but at least the entire die was covered.

Here's some pics of my thermal paste coverage on the l9i and c7cu.
View attachment 68940
View attachment 68941
Don't worry, the plastic was removed (y)

Test conditions: 25c ambient, NT-H1 and a9x14 used on both heatsinks, TDP set to 120w, fan speed set to max (2.5k), vcore 1.04, all core 4.3ghz, prime95 8k inplace FFTs for 10 mins inside dan a4sfx v2, no mods apart from delid and hwmonitor was reset a few minutes into the test to clear idle values.

Test results:
L9i (potentially bad paste coverage):
View attachment 68942

C7cu:
View attachment 68943

C7cu
Pros: 45c ΔT
Cons: Turbulence starting from 1.3k rpm /w a9x14 and x2 the weight of alu verison.
View attachment 68944

Thanks for sharing your results! The thermal paste application may have affected the numbers somewhat, but the difference is staggering nonetheless. What metal clips did you use to attach the A9x14 fan to the C7cu so cleanly?

You have decent thermal headroom, would love to see you try to overclock a bit and see how the cooler performs. I bet you could do 4.6Ghz on all cores with undervolting, may even be able to push 4.8Ghz, but at that point you may be in the low 90s C.
 
My test wasn't very scientific because I had a messed up thermal paste application on the l9i during the test. Only realized after I removed it to put in the c7cu. I'm too lazy to retest the l9i with better thermal paste coverage, but at least the entire die was covered.

Here's some pics of my thermal paste coverage on the l9i and c7cu.
View attachment 68940
View attachment 68941
Don't worry, the plastic was removed (y)

Test conditions: 25c ambient, NT-H1 and a9x14 used on both heatsinks, TDP set to 120w, fan speed set to max (2.5k), vcore 1.04, all core 4.3ghz, prime95 8k inplace FFTs for 10 mins inside dan a4sfx v2, no mods apart from delid and hwmonitor was reset a few minutes into the test to clear idle values.

Test results:
L9i (potentially bad paste coverage):
View attachment 68942

C7cu:
View attachment 68943

C7cu
Pros: 45c ΔT
Cons: Turbulence starting from 1.3k rpm /w a9x14 and x2 the weight of alu verison.
View attachment 68944

Nice test there mate, i can for sure say that the thermal paste application on the L9i was bad and we should not compare those two test, but the C7 Cu had some nice thermals!

What was your fan speed? Max or? And is it the Boxed Noctua 92x14mm fan or retail version?
 
Thanks for sharing your results! The thermal paste application may have affected the numbers somewhat, but the difference is staggering nonetheless. What metal clips did you use to attach the A9x14 fan to the C7cu so cleanly?

You have decent thermal headroom, would love to see you try to overclock a bit and see how the cooler performs. I bet you could do 4.6Ghz on all cores with undervolting, may even be able to push 4.8Ghz, but at that point you may be in the low 90s C.

Temps are really good. Gaming temps are 50-60c depending if I'm running 60 or 120fps with auto fans running at 1.3k-1.5k rpm, but I'm not pleased with the turbulence which is also audible in that rpm range.

Definitely have headroom to OC a bit, going to try go up to 4.6 sometime later.
These are the clips https://item.taobao.com/item.htm?id=562562405273.
You could use that design and make those yourself with some wire and cheap bending tool too.

Nice test there mate, i can for sure say that the thermal paste application on the L9i was bad and we should not compare those two test, but the C7 Cu had some nice thermals!

What was your fan speed? Max or? And is it the Boxed Noctua 92x14mm fan or retail version?

It's the boxed l9i fan, so it goes up to 2.5k rpm as you can see in hwmonitor and it was set to max manually in bios just to keep results constant.
 
Temps are really good. Gaming temps are 50-60c depending if I'm running 60 or 120fps with auto fans running at 1.3k-1.5k rpm, but I'm not pleased with the turbulence which is also audible in that rpm range.

Definitely have headroom to OC a bit, going to try go up to 4.6 sometime later.
These are the clips https://item.taobao.com/item.htm?id=562562405273.
You could use that design and make those yourself with some wire and cheap bending tool too.

That's amazing, you can find anything in China. I wish there was an easy way to shop from Taobao, I ordered some keycaps from them once and between trying to navigate/translate the site and figuring out how to pay, it was quite a pain. Great job. You could also delid, I mean you've gone this far.... That's going to bring down temps another 15 C at least. You can also get the window panels, the added clearance will help with turbulence.
 
That's amazing, you can find anything in China. I wish there was an easy way to shop from Taobao, I ordered some keycaps from them once and between trying to navigate/translate the site and figuring out how to pay, it was quite a pain. Great job. You could also delid, I mean you've gone this far.... That's going to bring down temps another 15 C at least. You can also get the window panels, the added clearance will help with turbulence.

You can order from Superbuy.com
i have ordered several item from there, very good service!
just paste in the URL and bam its translated to english everything
 
Ok, so we have one test run with the copper cooler and thermal paste and another one with the Noctua cooler that has barely any thermal paste on it. This review doesn’t help at all!
I would highly appreciate it, if you tested the Noctua again, this time with proper thermal paste. Please!
 
I have a A4-SFX v1 that I'm only using for music duties and some youtube/netflix. (I have even disabled two of the cores on a 4770k). I'd love to be able to passively cool the cpu in this case. My PSU is already fanless and the stock fan is adequate but makes a little noise. I've asked about the nh-l9x65 earlier in this thread. It's 51mm without the fan, and so I though I though 3mm might just be at the edge of what is recommended to re-bend the copper pipes. (However, I could use 4mm motherboard standoffs and that should solve it) But now I read in the thread that the A4 isn't strictly limited to 48mm height, but actually 50mm as the absolute maximum, that leaves only 1mm that needs to be adjusted by bending the copper pipes.

Anyone have any thoughts on that, or should I just order the nh-l9x66 and hope for the best?

Unfortunately the nh-l9x65 doesn't come in an all copper version..THAT I would like. But I guess it should still cool better than the Cryorig C7 Cu.
 
Do you really need a passive cooler? I can’t hear my NH-L9I when it’s running at just a few hundred RPM.
 
I have a A4-SFX v1 that I'm only using for music duties and some youtube/netflix. (I have even disabled two of the cores on a 4770k). I'd love to be able to passively cool the cpu in this case. My PSU is already fanless and the stock fan is adequate but makes a little noise. I've asked about the nh-l9x65 earlier in this thread. It's 51mm without the fan, and so I though I though 3mm might just be at the edge of what is recommended to re-bend the copper pipes. (However, I could use 4mm motherboard standoffs and that should solve it) But now I read in the thread that the A4 isn't strictly limited to 48mm height, but actually 50mm as the absolute maximum, that leaves only 1mm that needs to be adjusted by bending the copper pipes.

Anyone have any thoughts on that, or should I just order the nh-l9x66 and hope for the best?

Unfortunately the nh-l9x65 doesn't come in an all copper version..THAT I would like. But I guess it should still cool better than the Cryorig C7 Cu.

Theoretically, with 4mm standoff you have a maximum of 53mm, might be less. I think I read somewhere that the cryorig c7 cu behave same as the l9x65.
 
is their going to be a another version of this case? cuss i can almost put 2 92mm aio(one for cpu, one for gpu) inside this case with the psu too if only it was a little wider
 
Do you really need a passive cooler? I can’t hear my NH-L9I when it’s running at just a few hundred RPM.
I have fans in my other A4-SFX that's geared towards gaming/desktop duties, so I don't mind fans. But this one is geared towards high end audio, and if I can get away without the fan, then I'm thinking "why not?". No mechanical noise and/or moving parts is somewhat the goal. If the absolute max really is 50mm (not the official 48mm) and the L9x65 is 51mm without the fans I figure I should be able to bend the heatpipes just that tiny amount. I can then use the included fan with a LP53 or C7 Cu in my other A4, so it would be a win-win. (I also have the standaalone version of the fan, but the specs of the included fan in the L9i/L9-65 are somewhat better suited for the A4 cooling duties)
 
@galletabah: cork worked perfectly - many thanks!!!
I added the 3mm foam just to the psu, I can't tell much difference but it's hard to know with no fan speed monitoring. :-/ I've ordered this fan adaptor so I can use the fan from my noctua cooler. I'll report when it arrives if it's worth the risk of electrocution!
 

Attachments

  • IMAG3513~2.jpg
    IMAG3513~2.jpg
    651.8 KB · Views: 0
Finally had my Dan A4 build up and running!! (photos taken by Tantricmodz who assembled my pc)

Intel I5 8600
Asus Strix Z370i
16gb G.skill Trident Z 3200
Cryorig C7
Asus Strix GTX1060 OC
512gb Samsung 960 evo
Corsair SF450
Be Quiet Silent wings 2 92mm
Custom cables by UPC

Will post the custom panels I had my pc modder did by the weekend and prolly test results aswell

received_1662179987163504.jpeg
 

Attachments

  • FB_IMG_1524713152858.jpg
    FB_IMG_1524713152858.jpg
    15.5 KB · Views: 0
Last edited:
Who have fitted a Noctua 92x14 fan under the Mobo?
It seems when i try to do it the mobo hits the back of the fan and which causes the mobo to not line up with the 4 standoffs...
 
Who have fitted a Noctua 92x14 fan under the Mobo?
It seems when i try to do it the mobo hits the back of the fan and which causes the mobo to not line up with the 4 standoffs...

I used the rubber bits to secure the fan and didn't have that problem; try those?
 
I used the rubber bits to secure the fan and didn't have that problem; try those?

There are no need to secure the fan if its under the Motherboard, think we are misunderstanding each other, im trying to put 1 additional fan in the bottom compartment of the case.
 
There are no need to secure the fan if its under the Motherboard, think we are misunderstanding each other, im trying to put 1 additional fan in the bottom compartment of the case.

I think we're on the same subject; I'm just saying that I used the rubber pieces from the fan kit to hold it down as much as possible so it barely touched the motherboard at all.
 

Attachments

  • 20180427_070151.jpg
    20180427_070151.jpg
    335.8 KB · Views: 0
Ahhh i see, maybe it gives the last mm i need for it to fit, so the fan dosent hit the Mobo at all?
 
Fair enough, i am gonna try do that soonish, is it the retail og boxed Noctua 92mm fan you got there?

Correct; I'm not sure I would encourage using the 2.5K rpm fan because the turbulence past 900 rpm is unfortunate.

Edited for typos.
 
Correct; I'm not sure I would encourage using the 2.5K rpm fan because the turbulence past 900 rpm is unfortunate.

Edited for typos.

Ye thats true, so the cutout in the bottom lined up perfectly with the rubber mounting i see?
 
Back
Top