Modern fabrication process and materials are typically rated for sustained operations at up to 100C without any issue. My hottest E8450 idles around 63C and spike to near or over 70C when under load. I set my 3 E8450s using schedutil governor. CPU speed swings between min and max freq and in between. Haven’t seen any OKD … yet.
So I wouldn’t worry about it if I were you.
Really weird. Maybe your router board has broken solder joins for the IC chips? Cold temp likely shrinked the metal joins enough for proper contact. Maybe look at the flash chip IC solder joins first.
Hi everyone,
if someone here is willing to sell their OKD-bricked RT3200 to me, that would be great and help a lot with further analyzing the problem. The less messed-around-with, the better -- I'd prefer a device which only run the installer once and never had never been fixed via JTAG or mtk_uartboot. Please contact me, so we can make an arrangement.
Could you kindly share what you did to achieve this result? I am on OpenWrt 23.05.2 r23630-842932a63d.
TIA
I actually found some internal antennas with approx. the same cable length as the originals, so I might opt for those. Now I just need to find an internal header that provides 5V or 12V for a low profile fan 
The fan from a header would be tricky, since the device was not designed to support it. Although you could probably tap a GPIO or two for monitoring and control off one of the headers or elsewhere on the board, you'd still need to add the proper components to isolate the fan electrically and also to handle the current that the fan needs. If that weren't enough, you'd then need to write and maintain a driver for the custom component.
That said, why not instead connect to the solder points of the power input itself? That would reduce the work to two solder points and a bit of strategically-placed kapton tape for wire management. Of course, adding a fan means a bigger power supply regardless, but 12VDC supplies and fans are plentiful these days.
Originally I was thinking of desoldering the USB header and connecting a USB fan's pins to the board. I honestly don't want to add much control, even a small blower fan like this would do the job at its lowest setting.
If you use the USB header, you'll run a strong risk of killing the port entirely with any kind of short or overload. Although the MediaTek chip internally supports USB3 speeds and features, this router was only built with USB2 and all its limitations. In other words, you might get 500mA.
If the manufacturer cheaped out, there may be poor power regulation on the port and you might get the full 500mA with a risk of burning out the unit if any problems occur or if the fan draws too much. If the device instead follows the complete USB specification, you'll only get 100mA out of it unless your fan can communicate with the host controller and negotiate for up to the full 500mA.
Which kind of USB fan were you imagining when you think of 5V 500mA?!
This really huge thing is spec'ed 150mA
Should really be no problem on the USB port of the router.
I wasn't expecting that it should draw 500mA at 5v, only that many of these routers have rather crappy protection against something that does. Elsewhere, I've seen too many fried USB ports from people using them for purposes other than their original intent.
If the USB header is de-soldered and a pin header is installed or a device is hard-wired, it would only take an accidental short from accidental blobbing or even tin fingers, or even a failure of the fan for the whole thing to go down the drain. Of course, there is then the whole issue of whether or not everything follows the spec and whether or not the router can reliably push the power to a device that might exceed the USB specified default of 100mA. To be completely fair, I haven't tried to blow out a USB port on this model router.
Yes, I agree, you need to measure resistance after doing the soldering job, there is probably no over-current protection what-so-ever, and yes, shorting the 5V DC will probably fry the device.
Please don't 
hello openwrt forum, taking a look at a tr3200 i had setup awhile ago, and I'm looking to update, I don't really have important settings, so i was looking to just reinstall...this is what is currently installed:
OpenWrt 22.03.5 r20134-5f15225c1e / LuCI openwrt-22.03 branch git-23.093.57104-ce20b4a
When I originally installed it i followed this
owrt-ubi-installer
It seems like there is some new stuff going on, and not sure what I should actually try to update to... I did look around on github and saw a tldr, not to install some of the newer stuff, I was assuming that i could just flash
openwrt-mediatek-mt7622-linksys_e8450-ubi-squashfs-sysupgrade.itb
from v1.1.1 owrt-ubi-installer
but when i was about to do that, i got some warning and figured i should see whats going on in the forum
You can't flash sysupgrade snapshot image from the installer (unless you re-run the installer, and you should only do this if you know what you are doing and ready to run bleeding-edge snapshot with potentially severe problems).
Use the official release image instead: https://downloads.openwrt.org/releases/23.05.2/targets/mediatek/mt7622/openwrt-23.05.2-mediatek-mt7622-linksys_e8450-ubi-squashfs-sysupgrade.itb
thanks, Daniel, it worked
OpenWRT 23.05.0
Are you using 802.11ax and is it this issue?
Will be fixed in the next 23.05.x release and is fixed on the snapshots.
Thank you, but I think I got them from my donor router. Thank you.
Thank you! that is what I was missing.
Ok... Too much wine tonight. I think I bricked it MORE better! I get this, and UBOOT is gone:
[BLDR] Second Bootloader Load Failed
PL fatal error...
Anyone can point me in the right direction?
F0: 102B 0000
F6: 0000 0000
V0: 0000 0000 [0001]
00: 0000 0000
BP: 0000 0041 [0000]
G0: 0190 0000
T0: 0000 036C [000F]
Jump to BL
UNIVPLL_CON0 = 0xFE000000!!!
mt_pll_init: Set pll frequency for 25M crystal
[PMIC_WRAP]wrap_init pass,the return value=0.
[pmic_init] Preloader Start..................
[pmic_init] MT6380 CHIP Code, reg_val = 0, 1:E2 0:E3
[pmic_init] Done...................
Chip part number:7622B
MT7622 Version: 1.2.8, (iPA)
SSC OFF
mt_pll_post_init: mt_get_cpu_freq = 1350000Khz
mt_pll_post_init: mt_get_mem_freq = 1600000Khz
mt_pll_post_init: mt_get_bus_freq = 1119920Khz
[PLFM] Init I2C: OK(0)
[BLDR] Build Time: 20200522-165358
==== Dump RGU Reg ========
RGU MODE: 4D
RGU LENGTH: FFE0
RGU STA: 0
RGU INTERVAL: FFF
RGU SWSYSRST: 8000
==== Dump RGU Reg End ====
RGU: g_rgu_satus:0
mtk_wdt_mode_config mode value=10, tmp:22000010
PL P ON
WDT does not trigger reboot
WDT NONRST=0x20000000
WDT IRQ_EN=0x340003
RGU mtk_wdt_init:MTK_WDT_DEBUG_CTL(590200F3)
[EMI] MDL number = 2
[EMI] DRAMC calibration start
[EMI] DRAMC calibration end
[EMI]rank size auto detect
[EMI]start_addr[0x40000000]=0x12345678, test_addr[0x48000000]= 0xEDCBA987
[EMI]start_addr[0x40000000]=0x12345678, test_addr[0x50000000]= 0xEDCBA987
[EMI]start_addr[0x40000000]=0xEDCBA987, test_addr[0x60000000]= 0xEDCBA987
[EMI]rank0 size: 0x20000000
[MEM] complex R/W mem test pass
RAM_CONSOLE wdt status (0x0)=0x0
[BBT] BMT.v2 is found at 0x3FF
[PLFM] Init Boot Device: OK(0)
[PART] blksz: 2048B
[PART] [0x0000000000000000-0x000000000007FFFF] "PRELOADER" (256 blocks)
[PART] [0x0000000000080000-0x00000000000BFFFF] "tee1" (128 blocks)
[PART] [0x00000000000C0000-0x000000000013FFFF] "lk" (256 blocks)
Device APC domain init setup:
Domain Setup (0x0)
Domain Setup (0x0)
Device APC domain after setup:
Domain Setup (0x0)
Domain Setup (0x0)
load lk (ret=-1)
[BLDR] Second Bootloader Load Failed
PL fatal error...
I should have mentioned that all devices mentioned in the original question are 802.11ac devices.
Please ignore my original concern about ~10% packet loss for now. After digging out the floor plan, the location is further away and less friendly to the 5 GHz band than I'd originally thought. I'll edit the original post.
But I still wonder if >=1% packet loss is acceptable when I have line of sight at 10 ft using the 5 GHz band. Client side RSSI is -41 dBm at 10 ft with line of sight. Packet loss >1%. 5 GHz radio transmission power is at 20 dBmV.
Edit:
Thanks for the reassurance. May I ask how long your three devices have operated under those conditions?
Thermals reached 64C average during regular daytime use with spikes upward of 66C. I tend to be overcautious. I'm hoping the hardworking devs come up with a solution soon. In the meantime, I'll try the performance governor and point an externally powered 120mm fan toward the side of the device. The fan dropped the average about 4C.
