There isn't a bracket but the top is flat so the fan can just sit on top of the unit and the Noctua fans have rubber corners that keep the fan from vibrating. The Noctua NF-A14 5V comes with a USB to 3 pin fan connector. I used the Noctua NF-F12 5V PWM with the Noctua Noctua NA-FC1 to be able to adjust the fan speed to about 60% of full speed.
CWWK i5-1235U 6 port i226 report
- Thread starter fta
- Start date
Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.
I'll fix it on top of the case's fins externally, als @mxkied2 said. - The internal fan at the bottom that sometimes comes with the case has a thickness of 10mms, yes. Someone in this thread mentioned earlier it might possibly be possible to accommodate a 15mm fan, but I severely doubt it would fit. The 10mm fan of my unit was already blocked by an M.2 SSD mounted onto the NGFF-to-NVME adapter that came also installed. There's just not enough room.
Crucial, Samsung or Hynix RAM should work. With Crucials already mentioned in this thread you are on the safe side. Here is a somewhat euphemistic explanation by Crucial as to why that is: The truth about memory manufacturers. (I'm not associated with them in any way.)
No, there are no brackets. I'm going to try silicone noise dampers I have still lying around. If that doesn't work, I will switch to velcro straps if the case's temps allow for that. If everything fails I might go berserk and force it on with some screws or another. I'm planning to stand the case on one of the small sides for better air circulation especially at the bottom, so just placing the fan on top is not an option for me. It has to be fixed one way or another.
I'd like to power it with a Y-cable in parallel with the bottom fan, but that remains to be seen. In that case it would have PWM, yes.
I might even run it at top speed, although it would kind of defeat the purpose of a quiet passively cooled casing, I guess.
It's going to sit on top of a half-height IT cabinet where already some fan noise is present anyway, so it might go unnoticed there (I have to hear it first, though). If that would work, I could just power it via USB and be done with it.
Last edited:
I've just paid for my order with CWWK Store in AliExpress and they told me I could get some free gifts (vesa mount, 8010 fan, etc) and even the i5-1240P upgrade.
Now I'm planning to make an UPS for this Mini PC. Is there anyone here with experience in this area?
Since this Mini PC consumes less than 100W max, I was planning to buy this BMS:
Li-ion 3S 12V BMS
20A NTC (since the diference in price between 10A and 20A is minimal)
And these batteries (since 3S3P - 9 batteries, I'd probably get 10):
Since I can use the Mini PC power supply to be the power supply for this BMS, I think I would only need a DC Booster in the output to keep the voltage always in 12V, so when the batteries start to drain, it wouldn't reduce the voltage for the Mini PC (and my ISP ONT that is also 12V).
I'm just trying to understand what would be the schematics for this and what DC Booster I should get.
By the way, does anyone know the idle power consumption of this i5-1240P?
In this post in reddit someone said it's about 15W:
With this 3S3P I'd get about 12.7V when fully charged and 9000 mAh.
If this Mini PC with i5-1240P consumes about 15W in idle, that would be about 1200mAh, so I would have about 7 hours of power. Not bad.
---Edit
I actually just came with a better idea... Buying a 100W Powerbank with PD, and getting a 12V barrel plug adapter with PD chip.
Now I'm planning to make an UPS for this Mini PC. Is there anyone here with experience in this area?
Since this Mini PC consumes less than 100W max, I was planning to buy this BMS:
Li-ion 3S 12V BMS
20A NTC (since the diference in price between 10A and 20A is minimal)
And these batteries (since 3S3P - 9 batteries, I'd probably get 10):
Since I can use the Mini PC power supply to be the power supply for this BMS, I think I would only need a DC Booster in the output to keep the voltage always in 12V, so when the batteries start to drain, it wouldn't reduce the voltage for the Mini PC (and my ISP ONT that is also 12V).
I'm just trying to understand what would be the schematics for this and what DC Booster I should get.
By the way, does anyone know the idle power consumption of this i5-1240P?
In this post in reddit someone said it's about 15W:
With this 3S3P I'd get about 12.7V when fully charged and 9000 mAh.
If this Mini PC with i5-1240P consumes about 15W in idle, that would be about 1200mAh, so I would have about 7 hours of power. Not bad.
---Edit
I actually just came with a better idea... Buying a 100W Powerbank with PD, and getting a 12V barrel plug adapter with PD chip.
Last edited:
Hello ,
Great post , thank you all for your tips, comments and great contributions.
The good news is that I have received my CWWK i5-1235U 6 port ordered from ALiExpress after 23 days. Delivery time was awful but the seller was responding quickly and accepted compensation of additional delivery fees (local) that were not communicated before.
The bad news is that when I started the unit , the bios date and time were wrong (device received June 23rd). I suspected that the battery was disconnected during transport. So I checked the battery connection, played with the unit for a few hours, installed Proxmox, VMs, CT,..... Then switched off and unplugged power. After 5 min, I checked again the bios time , it showed 4 hours less. I repeated this many times, always the same.
This what I did next
The seller said he will assist me to resolve or replace. Honestly I don’t know what else can be done from my side ?
Thank you for your suggestions
Great post , thank you all for your tips, comments and great contributions.
The good news is that I have received my CWWK i5-1235U 6 port ordered from ALiExpress after 23 days. Delivery time was awful but the seller was responding quickly and accepted compensation of additional delivery fees (local) that were not communicated before.
The bad news is that when I started the unit , the bios date and time were wrong (device received June 23rd). I suspected that the battery was disconnected during transport. So I checked the battery connection, played with the unit for a few hours, installed Proxmox, VMs, CT,..... Then switched off and unplugged power. After 5 min, I checked again the bios time , it showed 4 hours less. I repeated this many times, always the same.
This what I did next
- Changed the bios battery with a new one, still the same.
- Flashed the latest bios CW-MBX-AD12_2023-06-08, It seems to hold time longer but still going back 4 hours when unplugged
- Replaced the NVMe SSD with a Crucial P5 and Timetec 16Gb kit DDR4 2666Mhz (by the way I suggest not to order any NVMe or Ram with the unit, the ones I received are just crap ). Bios problem is the same
The seller said he will assist me to resolve or replace. Honestly I don’t know what else can be done from my side ?
Thank you for your suggestions
Attachments
Any ideas guys ?Hello ,
Great post , thank you all for your tips, comments and great contributions.
The good news is that I have received my CWWK i5-1235U 6 port ordered from ALiExpress after 23 days. Delivery time was awful but the seller was responding quickly and accepted compensation of additional delivery fees (local) that were not communicated before.
The bad news is that when I started the unit , the bios date and time were wrong (device received June 23rd). I suspected that the battery was disconnected during transport. So I checked the battery connection, played with the unit for a few hours, installed Proxmox, VMs, CT,..... Then switched off and unplugged power. After 5 min, I checked again the bios time , it showed 4 hours less. I repeated this many times, always the same.
This what I did next
- Changed the bios battery with a new one, still the same.
- Flashed the latest bios CW-MBX-AD12_2023-06-08, It seems to hold time longer but still going back 4 hours when unplugged
- Replaced the NVMe SSD with a Crucial P5 and Timetec 16Gb kit DDR4 2666Mhz (by the way I suggest not to order any NVMe or Ram with the unit, the ones I received are just crap ). Bios problem is the same
The seller said he will assist me to resolve or replace. Honestly I don’t know what else can be done from my side ?
Thank you for your suggestions
Last edited:
If you have a new device and having issues with it, contact the seller. If the seller wont help, lodge a complaint in the system and possibly dispute the charge.
What's your current timezone? +/- 4 hours?
If I remember correctly, depending the operating system, when you change the time and timezone behaves different changing the system time.
Maybe the host OS or some of the VMs are messing with the system time?
What I would try, reboot the computer with an Ubuntu usb drive, set correctly the time/timezone, reboot again directly into bios to confirm the time and unplug the computer.
Or even without Ubuntu usb drive, just enter in the bios, manually set the date/time, reboot again to confirm and unplug.
And after a while, confirm the time again going directly to the bios, do not let the operating system starts.
If I remember correctly, depending the operating system, when you change the time and timezone behaves different changing the system time.
Maybe the host OS or some of the VMs are messing with the system time?
What I would try, reboot the computer with an Ubuntu usb drive, set correctly the time/timezone, reboot again directly into bios to confirm the time and unplug the computer.
Or even without Ubuntu usb drive, just enter in the bios, manually set the date/time, reboot again to confirm and unplug.
And after a while, confirm the time again going directly to the bios, do not let the operating system starts.
Clocks are super simple. You shouldn't see that happen unless it's a software issue or a really bad BIOS issue that we'd all be seeing. right time -/+ 4 hours and it's likely the clock works but at @jmartinez50 is pointing out, it's probably a SW issue. When clocks go bad other than battery, it's resetting to 1 Jan 1900 or another date like that.
Thank you both @MiniKnight and @jmartinez50 for your replies and suggestions , indeed my time zone is GMT+4 . The date did reset to 1 Jan 2021 when I replaced the battery but other than that, it stays on same day. I will test with Ubuntu usb drive and check.
Many thanks
Has anyone used the 4p power adapter on the board itself yet? Was thinking of using a regular psu since I need a couple fans and 2X 3.5 spinner drives. Was going to keep bottom plate off, use sata extensions, and set on on my rack shelf with a fan mounted under the shelf (the shelf is slotted for air). Mount the 2 drives on the shelf next to it as well as the small psu to power it all. Wasn't sure if the 4p power plug on the mainboard was standard?
Thank you both @MiniKnight and @jmartinez50 for your replies and suggestions , indeed my time zone is GMT+4 . The date did reset to 1 Jan 2021 when I replaced the battery but other than that, it stays on same day. I will test with Ubuntu usb drive and check.
Many thanks
Hello ,
Good news, I booted from a Kali live usb drive, then found out that time zone was GMT so I made it GMT+4 and rebooted straight to Bios. The time was correct. Then switched off the device , removed the power and started again and all was good
. Also kept the device off for 24h and it was ok.I presume the device has no issues , unless there are other tests that I should perform
Thank you so much for your help guys
Installed pfSense 2.7 this weekend. Working fine with the I226 nics indeed! Used the settings for pfSense (and BIOS) mentioned by iceman_jkh. But I don't get expected results, or maybe my expectations are wrong.
My expectations are: when idle the CPU frequencies would be low, like <1000 MHz and when needed it can burst to 4.8GHz (because my box has an Intel i7-1265u).
Actual results: CPU frequencies stay at a certain level, with minor fluctuations.
My testing is as follows:
Running cat /var/run/dmesg.boot | grep "Speed Shift" gives me:
So I set the system tunables dev.hwpstate_intel.[0..11].epp : 95. And set hw.acpi.cpu.cx_lowest : C3. Running command sysctl -a dev.cpu | grep 'freq_levels\|freq' gives me:
Running it multiple times, I see no big fluctuation. Also, I would expect frequencies to be much lower when idle? Something below 1000 MHz? Or am I wrong here?
I tried changing system tunables system tunables dev.hwpstate_intel.[0..11].epp and set them to 100. No difference, same frequencies. Also setting them to 75 made no difference.
Just for checking if these setting are doing anything at all, I changed them to 50 and later 0. Then I saw the frequencies going up, so I think they do have an effect.
Any thoughts?
My expectations are: when idle the CPU frequencies would be low, like <1000 MHz and when needed it can burst to 4.8GHz (because my box has an Intel i7-1265u).
Actual results: CPU frequencies stay at a certain level, with minor fluctuations.
My testing is as follows:
Running cat /var/run/dmesg.boot | grep "Speed Shift" gives me:
Code:
hwpstate_intel0: <Intel Speed Shift> on cpu0
hwpstate_intel1: <Intel Speed Shift> on cpu1
hwpstate_intel2: <Intel Speed Shift> on cpu2
hwpstate_intel3: <Intel Speed Shift> on cpu3
hwpstate_intel4: <Intel Speed Shift> on cpu4
hwpstate_intel5: <Intel Speed Shift> on cpu5
hwpstate_intel6: <Intel Speed Shift> on cpu6
hwpstate_intel7: <Intel Speed Shift> on cpu7
hwpstate_intel8: <Intel Speed Shift> on cpu8
hwpstate_intel9: <Intel Speed Shift> on cpu9
hwpstate_intel10: <Intel Speed Shift> on cpu10
hwpstate_intel11: <Intel Speed Shift> on cpu11
Code:
dev.cpu.11.freq_levels: 2688/-1
dev.cpu.11.freq: 2188
dev.cpu.10.freq_levels: 2688/-1
dev.cpu.10.freq: 2188
dev.cpu.9.freq_levels: 2688/-1
dev.cpu.9.freq: 2188
dev.cpu.8.freq_levels: 2688/-1
dev.cpu.8.freq: 2188
dev.cpu.7.freq_levels: 2688/-1
dev.cpu.7.freq: 2188
dev.cpu.6.freq_levels: 2688/-1
dev.cpu.6.freq: 2188
dev.cpu.5.freq_levels: 2688/-1
dev.cpu.5.freq: 2129
dev.cpu.4.freq_levels: 2688/-1
dev.cpu.4.freq: 2088
dev.cpu.3.freq_levels: 2688/-1
dev.cpu.3.freq: 2051
dev.cpu.2.freq_levels: 2688/-1
dev.cpu.2.freq: 1989
dev.cpu.1.freq_levels: 2688/-1
dev.cpu.1.freq: 1919
dev.cpu.0.freq_levels: 2688/-1
dev.cpu.0.freq: 1889I tried changing system tunables system tunables dev.hwpstate_intel.[0..11].epp and set them to 100. No difference, same frequencies. Also setting them to 75 made no difference.
Just for checking if these setting are doing anything at all, I changed them to 50 and later 0. Then I saw the frequencies going up, so I think they do have an effect.
Any thoughts?
Last edited:
I must be doing something wrong. I got the 1215U and before the custom BIOS is was getting a CPU mark of about 9500. After flashing the HWP ON BIOS and setting the recommended settings in BIOS I'm now getting a CPU mark of ...... 7500. What?!
I'm testing in Win10 before I install pfsense.
Any thoughts? I've double checked all of the BIOS settings. Max temp only hits about 69C.
I'm testing in Win10 before I install pfsense.
Any thoughts? I've double checked all of the BIOS settings. Max temp only hits about 69C.
This might be your problem. This CPU has C1/C1E, C6, C8, and C10. By setting it to C3, you're limited to only C1/C1E, which won't let your cores turbo up as much. Remove this setting or set it to C10.
Thank you for the suggestion. I tried but unfortunately it didn't work. I set it to C10 and rebooted, but no change in frequencies. I also deleted the system tunable completely and rebooted, same result.
sysctl -a dev.cpu | grep 'freq_levels\|freq'
Code:
dev.cpu.11.freq_levels: 2688/-1
dev.cpu.11.freq: 2188
dev.cpu.10.freq_levels: 2688/-1
dev.cpu.10.freq: 2188
dev.cpu.9.freq_levels: 2688/-1
dev.cpu.9.freq: 2188
dev.cpu.8.freq_levels: 2688/-1
dev.cpu.8.freq: 2188
dev.cpu.7.freq_levels: 2688/-1
dev.cpu.7.freq: 2188
dev.cpu.6.freq_levels: 2688/-1
dev.cpu.6.freq: 2188
dev.cpu.5.freq_levels: 2688/-1
dev.cpu.5.freq: 2088
dev.cpu.4.freq_levels: 2688/-1
dev.cpu.4.freq: 2088
dev.cpu.3.freq_levels: 2688/-1
dev.cpu.3.freq: 2002
dev.cpu.2.freq_levels: 2688/-1
dev.cpu.2.freq: 1989
dev.cpu.1.freq_levels: 2688/-1
dev.cpu.1.freq: 1863
dev.cpu.0.freq_levels: 2688/-1
dev.cpu.0.freq: 1868
Code:
dev.cpu.[0..11].cx_method: C1/hlt
dev.cpu.[0..11].cx_usage_counters: xxxx
dev.cpu.[0..11].cx_usage: 100.00% last xxxxus
dev.cpu.[0..11].cx_lowest: C1
dev.cpu.[0..11].cx_supported: C1/1/0That does look like you’re still limited to C1. If you’re using my bios, it defaults to auto for c-states, which allows it to go all the way to C10.