No reply from manufacturer.
Project version: CW-ADLN-NAS-V10
Here is the latest bios for V10.
Last edited:
12gen N-series Nas motherboard (topton, cwwk, ... )
- Thread starter Steffe
- Start date
Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.
Project version: CW-ADLN-NAS-V10
Here is the latest bios for V10.
My Unraid server has been running for approximately six months without any issues. Over the past week, I’ve been experiencing system freezes, requiring a hard reset via the power button. The occurrences are random—sometimes after an hour, sometimes after a day—with no discernible pattern.
I suspect faulty RAM may be the root cause.
These devices struggling after six months makes me think fatigue in the electrolytic capacitors in either the power brick or the mini pc itself. 6 months is over 4000 hours and at max temperatures some caps are only rated for 1000 hours. Allow for lower temps (hopefully) and we might scale that up a bit, but 4000 might be enough.
I've been pretty sure it's the element that has killed my various broadband routers over the years. Those are the most annoying failures as they are literal 24/7 devices for everyone, but at one point 2 years was about my max.
I've been pretty sure it's the element that has killed my various broadband routers over the years. Those are the most annoying failures as they are literal 24/7 devices for everyone, but at one point 2 years was about my max.
That could very well be the case. I am still attempting to achieve stable operation (after more than half a year of functioning); otherwise, I will need to switch to a high-end board from a reputable manufacturer (although there are not many options available on the market)
Quite likely if they contain wet type capacitors (electrolytic), which are cheaper so wouldn't be a surprise. We would hope they would only use solid polymer capacitors with life spans of ten years or more, but these are built to be cheap and to be replaced often. Still whatever components they contain, you can be sure they are cheapest or otherwise factory scrap. There is a big market in China for components that fail testing, i.e. they work but are outside of normal parameters showing they are flawed in someway, these components that fail testing get sold, often for less than cost price, so the manufacturer at least recoups something, to anyone that wants them. These components make there way into most things on Aliexpress, which is why often we can buy things so cheap!
I have a 6 port network appliance based on a i7 7500U that has been running 24/7 for 7 years now, it only contains ceramic capacitors, clearly built to last and to a decent enough design that it idles at 4 watt, so runs fan-less and barely above ambient temperature, so stress on components is at the minimum. Sadly, can't buy anything similar to that now, they all idle at way too high wattages baking in their own inefficiency, and nothing is built to last.
It could be something as simple as one failing capacitor in the external supply. Then you get ripple or poor voltage and instability. I read a quick refresher on capacitors last night, and temperature is the real killer. Say they rate a cheap one at 1000h at 105c, apparently every 10c cooler you double the life. So at 85c the cheap one is 4000h. A reasonable 4000h capacitor will be at 16000h.
Now that critical temp is inside the capacitor - so power bricks, mini pcs and or VRM banks, start at a disadvantage due to lack of cooling in its surroundings. The caps are sitting in a hot spot. Depending on the load and the amount of ripple then they will be working harder which generates heat inside the cap. There was also a lot of talk about ESR (which I take to be the internal resistance). Expensive caps typically have low ESR and so make less internal heat under the same load & ripple.
I recall a few years back there were "standard repair" videos for the common routers, since the failure modes were so common. But if you couldn't DIY the fix a paid repairer would cost you more than a cheap router.
Now that critical temp is inside the capacitor - so power bricks, mini pcs and or VRM banks, start at a disadvantage due to lack of cooling in its surroundings. The caps are sitting in a hot spot. Depending on the load and the amount of ripple then they will be working harder which generates heat inside the cap. There was also a lot of talk about ESR (which I take to be the internal resistance). Expensive caps typically have low ESR and so make less internal heat under the same load & ripple.
I recall a few years back there were "standard repair" videos for the common routers, since the failure modes were so common. But if you couldn't DIY the fix a paid repairer would cost you more than a cheap router.
Good morning.
I’m facing the same issue and the board is out of warranty. Now I need to source a replacement board with similar specifications from a reputable manufacturer. Any suggestions? I require a power rating of N100/150/303.
Good morning.
I’m facing the same issue and the board is out of warranty. Now I need to source a replacement board with similar specifications from a reputable manufacturer. Any suggestions? I require a power rating of N100/150/305
If anybody needs it...
Driver for the ITE IT8613E Super-IO Chip (Module for Linux / LM-Sensors)
GitHub - bogie/it87-sensors: ITE IT8613E in lm-sensors
GitHub - shauno8/it87: fix ITE IT8613E in lm-sensors
Driver for the ITE IT8613E Super-IO Chip (Module for Linux / LM-Sensors)
GitHub - bogie/it87-sensors: ITE IT8613E in lm-sensors
GitHub - shauno8/it87: fix ITE IT8613E in lm-sensors
I am looking to. Cwwk sende Like the perfect boards, but their warranty policy rules them out completely, topton too.
seems this market is not covered by anyone else
minisforum has nothing fitting unfortunetaly
one user on Amazon Write that downclocking the ram Speed to 4400 or 4200 seemed to eliminate crashes, I will Test that
Last edited:
Hi. I'd like some guidance from someone more experienced with the CW-NAS-ADLN-K motherboard (N100 purple model), if possible. My system OS is installed on a nvme and I'm trying to passthrough the SATA controller to a vm in order to use it as a NFS/SMB file server. I've done all the steps using 2 different distros (Arch Linux and Debian 13) and with both I have the same behavior. As soon as I boot the vm it takes a long time (I believe it's almost 10 min) to the vm becomes available and the SATA controller to become visible.
I'm quite sure everything is configured right because I have a spare AsRock motherboard and if I just insert the nvme and change the pertinent configurations everything works as expected. Thing is I'd rather using this CW because of its lower power consumption.
I imagined that this behavior could be related to an old BIOS version, so I've just flashed this one but it didn't change anything...
Here is some info I believe is useful:
Does anyone have had this issue? Maybe some BIOS configuration I'm missing? I'd appreciate any suggestion.
dmesg shows no errors on host. I'm quite sure everything is configured right because I have a spare AsRock motherboard and if I just insert the nvme and change the pertinent configurations everything works as expected. Thing is I'd rather using this CW because of its lower power consumption.
I imagined that this behavior could be related to an old BIOS version, so I've just flashed this one but it didn't change anything...
Here is some info I believe is useful:
Bash:
~$ sudo ./iommu.sh
IOMMU Group 11 02:00.0 SATA controller [0106]: ASMedia Technology Inc. ASM1166 Serial ATA Controller [1b21:1166] (rev 02)
Bash:
~$ sudo lspci -nnk
02:00.0 SATA controller [0106]: ASMedia Technology Inc. ASM1166 Serial ATA Controller [1b21:1166] (rev 02)
Subsystem: ASMedia Technology Inc. Device [1b21:2116]
Kernel driver in use: vfio-pci
Kernel modules: ahci
Bash:
~$ cat /etc/default/grub
GRUB_CMDLINE_LINUX_DEFAULT="quiet intel_iommu=on iommu=pt"
Bash:
~$ cat /etc/modprobe.d/vfio.conf
softdep ahci pre: vfio-pci
options vfio-pci ids=1b21:1166Hey @KevinR , thank you for taking the time to help.
I'm not sure if that's the case because I took the nvme that has the OS, put it on other motherboard (changing the pertaining ids with the ones on the new mobo) and everything worked. No timeouts and I can see the disks in the vm immediately.
And also, it doesn't happen only when the host and / or the vm is booting... if the host has already started and the vm is stopped, if I start the vm it takes the same long time to the disks to became available on the vm.
I have the purple CWWK N305 motherboard that I'm running Unraid on. Currently I have 2 SK hynix 2TB m.2 NVMe ssds in each of the m.2 slots on the motherboard. I also have 1) 2.5" ssd and 5) 3.5" hard drives that are occupying the 6 SATA ports. If I want to add more hard drives would I get an m.2 SATA adapter and move one of the SK hynix ssd to a PCIe to NVMe adapter? Or is there another option that would work better?
Thanks.
Thanks.
If you have the purple one. What do you have in the x4 slot? Could put an HBA in there or a PCIE Sata card.
Currently I don't have anything in the x4 slot. I was just wondering if it would make more sense to move the m.2 NVMe ssds that I have to a PCIe card. Don't know if it would better utilize the read and write speeds over the m.2 slots. Then I could just get a m.2 SATA adapter to use for the slower hard drives. If you don't think it would matter much then just getting a PCIe HBA or SATA card would be easier.
