12gen N-series Nas motherboard (topton, cwwk, ... )

Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.

Outbound0918

Member
Jun 29, 2024
38
9
8
There was another insidence of this, a collision between the two standards. 6-32 vs M3
As I recall the case end of the standoffs is always 6-32 but the motherboard end can be either thread. It's more odd if the MB included M3 but they're too skinny for the MB holes.
So the standoffs *I think* are 6-32 (need a battery for my calipers) and the screws shipped with the MB *I think* are M3.
The hole in the motherboard are too skinny for 6-32 but allow the narrower M3 screw to pass thoguh.

So I need some M3 standoffs I guess?
 

Outbound0918

Member
Jun 29, 2024
38
9
8
For anyone wondering, the standoff size you need is: M3x6+#6-32x4

M3 is the female side x6 is the length so 6mm
+
#6-32 is the male side x4 is the length so 4mm
 

Outbound0918

Member
Jun 29, 2024
38
9
8
Yeah, the previous thread it seemed the case standoffs had 6-32 on both ends. While the MB expected M3. So you may need this type which are 6-32 male into the case and M3 female. #6-32 to M3 Jack Screw Standoff 50 Pack - Fasteners & Brackets | Computer Parts | StarTech.com
Yes! Thank you.
The key term is M3x6+#6-32x4

I have never seen #6-32 be used for the motherboard mounting..., usually M3 is used. But this motherboard FORCES you to use M3 too due to the diameter of the holes

Now see what other compatibility issues I run into, so far there have been 3:
- motherboard screws/standoffs
- gap between heatspreader and die
- included cpu cooler extremely close to touching caps

 

phil-2024

New Member
Sep 7, 2024
24
15
3
CPUs such as this shouldn't be using a thermal pad, they just aren't conductive enough, it will be okay if the CPU isn't using much power, but if something ramps it up it is likely to get hotter than it should and could start thermal throttling much sooner.

Thermal paste is only intended to fill the almost microscopic pits and gaps when two metal surfaces touch, you should really only need the smallest amount, and the interface between CPU and heatsink should be a good amount of metal to metal contact for the best conductivity. So when they ship these things with a huge amount of thermal grease trying to bridge any sort of gap, is just poor, and they basically don't care. It will probably work for a bit, but thermal cycling will mean the paste will end pulling apart or drying out leaving a gap. Adding extra copper shims helps but adds more 'resistance' to the heat transfer. If you are going to add a shim, it is better to add between the copper slug and the case/heatsink, this way it can be screwed down well for the best contact, and the metal slug is then directly on the CPU which helps to quickly get the heat away from the CPU die.

As we are finding, these things are just being flooded out of China, if they work its more luck than anything, if they fail in a year or two no one is going to go to the trouble of sending it back and any legal consumer rights we might have are impossible to enforce against these Chinese companies.

Cavet emptor
 

RolloZ170

Well-Known Member
Apr 24, 2016
6,363
1,944
113
Does anyone have some steps on how to update the BIOs on the CW-NAS-ADLN-K board? There is not documentation on how to do this, the update file is just an ISO?
extract with 7z or WinRAR, or create a bootable Disk/USB thumb with it.
 

Outbound0918

Member
Jun 29, 2024
38
9
8
extract with 7z or WinRAR, or create a bootable Disk/USB thumb with it.
Yeah, I figured it out shortly after asking :D

There is an EFI section in the ISO and the efi startupscript executes auto.nsh on the root of the drive. Then this script changes to the /CW-NAS-ADLN-K/bios directory and runs the 1.nsh script which uses fpt to flash the bios .bin file.
 

RolloZ170

Well-Known Member
Apr 24, 2016
6,363
1,944
113
Then this script changes to the /CW-NAS-ADLN-K/bios directory and runs the 1.nsh script which uses fpt to flash the bios .bin file.
afaik AFUEFI not FPT. sometimes you have to change the Afuxxxx.efi filename to AfuEfi64.efi
edit: you are right,
the first i checked was afuefi, other use fpt
 

Outbound0918

Member
Jun 29, 2024
38
9
8
Yes, firmware was already recent but I updated it anyway. Bios also updated to the newest version. Still fails when ASPM is enabled
What is the latest version? how did you check and update?
Please could you provide links to these?

What setting speicifically did you change to enable aspm?
 

Cromanin

New Member
Sep 11, 2024
12
2
3
CPUs such as this shouldn't be using a thermal pad, they just aren't conductive enough, it will be okay if the CPU isn't using much power, but if something ramps it up it is likely to get hotter than it should and could start thermal throttling much sooner.

Thermal paste is only intended to fill the almost microscopic pits and gaps when two metal surfaces touch, you should really only need the smallest amount, and the interface between CPU and heatsink should be a good amount of metal to metal contact for the best conductivity. So when they ship these things with a huge amount of thermal grease trying to bridge any sort of gap, is just poor, and they basically don't care. It will probably work for a bit, but thermal cycling will mean the paste will end pulling apart or drying out leaving a gap. Adding extra copper shims helps but adds more 'resistance' to the heat transfer. If you are going to add a shim, it is better to add between the copper slug and the case/heatsink, this way it can be screwed down well for the best contact, and the metal slug is then directly on the CPU which helps to quickly get the heat away from the CPU die.

As we are finding, these things are just being flooded out of China, if they work its more luck than anything, if they fail in a year or two no one is going to go to the trouble of sending it back and any legal consumer rights we might have are impossible to enforce against these Chinese companies.

Cavet emptor
Of all options trying to fix this heat spreader gap problem tpad is the way, imho.
Tpaste is not an option at all. Copper shim is troublesome as it is problematic to determine what thickness to use, unless you have a selection and ready to go trail & error risking breaking die in process.
 
  • Like
Reactions: Outbound0918

Outbound0918

Member
Jun 29, 2024
38
9
8
I completely agree with your point. I have conducted three stress-ng tests: 1. using a 21W/mk thermal pad, 2. using only thermal paste (Noctua NT-H2), and 3. using a 0.5mm copper shim with thermal paste. The highest CPU temperature occurred when using the thermal pad, followed by using only the thermal paste, and the lowest temperature was achieved with the copper shim and thermal paste.
Can you post some numbers with screenshots etc? As I have a really hard time believing a lump of thermal paste to fill a ~0.5mm air gap results in a lower CPU temp than when using a thermal pad.

I will do my own tests to validate this.
 
Last edited:

Outbound0918

Member
Jun 29, 2024
38
9
8
It's too troublesome to remove the CPU cooler and perform three different tests again. Could you check it online using ChatGPT?
What? Use chatgpt to validate your tests?
That makes zero sense and is making me question if you actually performed said tests.


Thermal pads are generally not the best choice for CPU cooling due to several key reasons:
Yes, but ONLY when comparing them to thermal paste being used correctly, but in this case it is not being used correctly, as it is not meant to be used to fill air gaps of ~0.5mm.
 

Outbound0918

Member
Jun 29, 2024
38
9
8
Whether you believe it or not is up to you. Why not do some testing yourself or use online resources to do some research?
I just said I am going to do some tests to validate this.

What are you talking about "use only resources to do some research", I asked for proof of your tests unless you post these publicly there is nothing to search for...
 

Cromanin

New Member
Sep 11, 2024
12
2
3
The problem is that it is difficult to measure the gap between cpu and the sink. But we know that Chinese makers are quite relaxed about precision and quality (no offence), so we can assume it is big enough. And though Cgpt indeed reports a test simulating 0,5 gap where tpaste performs better than tpad, it still suggests using tpad mainly I assume because tpaste will degrade really fast, dry up and stop performing.
I myself plan to use Valor Odin with 15w thermal conductivity which should be more than enough for n305 if the gap is not too large and the pad performs up to spec.
 
  • Like
Reactions: Outbound0918

majoracehole

New Member
Apr 8, 2024
6
2
3
Of interest I have only ever seen this board POST with two beeps, likewise memtest passes fine and on the whole everything seems to work...so I just assumed that's how it is with this one


Hello everyone, thank you to all who have already contributed there is a wealth of info in this thread.

I have received the purple n100 cwwk board bought directly from them & I have updated the bios from their website which was nice and easy using ventoy usb. I installed a stick of Samsung DDR5 4800 16GB which is recognised the bios posts fine and I have run memtest from a USB drive which ran all 4 passes successfully.

However on each power on the motherboard gives two beeps which according to CWWK indicates a memory problem. I have re seated the ram but the beeps persist. There could well be a problem with the ram as it was bought from aliexpress so is almost certainly used/recycled from a laptop but I am unsure as it has passed memtest. CWWK said that have not tested this ram's compatibility with their board so before I ask the ram vendor to send me another stick I wanted to ask if anyone here has samsung ram working with their n100 board in case perhaps I am better just refunding this ram and buying a different brand

Also I have had no issues with connecting to the monitor with HDMI it was a 4k monitor and it worked straight away & if anyone wants to change the board so it doesn't turn on as soon as power is connected it is the small jumper located behind the audio in jack which you change from 2-3 to 1-2

Thanks
 

HaoSs

Member
Feb 27, 2024
51
27
18
my purple board is not changed from the manufacturer, i have a stoc intel cooler on it and it idles at 33 ( cooler fan is off) and goes up to 65 in full load before the fan ramps up and stabilizes at 55. I also have 4 out fans on an n3 case that run at a low 1000 rpm most of the time.
50 with pad seems way to high, also why you all stress so much about coolers, a stock intel one fits perfectly, touches nothing, and is able to dissipate 15W of heat with no problems
 
Last edited:

HaoSs

Member
Feb 27, 2024
51
27
18
they run at lower rpm/temp-dependent. I should test with CPU fan running all the time like a normal PC.
 

mediaNen

New Member
Jul 29, 2024
2
3
3
they run at lower rpm/temp-dependent. I should test with CPU fan running all the time like a normal PC.
I run my purple board N100 with the Jonsbo cooler directly from CWWK. I haven't touched any bios settings regarding cooling. Idle in Unraid with everything spinning (6x 3.5" HDD, NVME drive etc.) is about 24C for the CPU and 27C for the motherboard. In full load it will not go above 49C. My case has decent airflow, and is not in any way bottlenecking the cooler.
 
  • Like
Reactions: KevinR