Hi, I have two hEX PoE - RB960PGS devices both sitting in a room with 10-15°C but the first (older one) shows a temperature of 37°C while the other (new one) shows 62°C
The new one with high temperature just sits there idle with no load. Should I worry about that? Because it seems like a huge difference and very high above ambient.
The weird thing is both are Model RB960PGS Revision r2 but one has a default CPU clock of 800 with a range of 720-1000 while the other has a default 720 with a range of 500-840
So it seems like they have a different processor inside, is that normal?
Re: Weirdly high temperature on new hEX PoE
I'm not saying that definitely everything is fine. But with passively cooled devices it's kind of normal to see high temperatures of parts which generate heat. Are you sure that both are same model? Check /system/routerboard/print output ...
Re: Weirdly high temperature on new hEX PoE
They are now on different voltages but the temperature was high even when the hot one ran from 24V supply
Cold one
Hot one
Cold one
Code: Select all
[admin@Router] > system/health/print
Columns: NAME, VALUE, TYPE
# NAME VALUE TYPE
0 voltage 27.6 V
1 temperature 39 C
[admin@Router] > /system/routerboard/print
routerboard: yes
board-name: hEX PoE
model: RB960PGS
revision: r2
serial-number: HGB09Y95TA1
firmware-type: qca9550L
factory-firmware: 6.49.11
current-firmware: 7.17.2
upgrade-firmware: 7.17.2
[admin@Router] > system/routerboard/settings/set cpu-frequency=
720MHz 800MHz 900MHz 1000MHz
[admin@Router] > system/routerboard/settings/print
auto-upgrade: no
boot-device: nand-if-fail-then-ethernet
preboot-etherboot: disabled
preboot-etherboot-server: any
cpu-frequency: 800MHz
boot-protocol: bootp
force-backup-booter: no
silent-boot: no
protected-routerboot: disabled
reformat-hold-button: 20s
reformat-hold-button-max: 10m Code: Select all
[admin@MikroTik_Switch] > /system/health/print
Columns: NAME, VALUE, TYPE
# NAME VALUE TYPE
0 voltage 51.8 V
1 temperature 63 C
[admin@MikroTik_Switch] > /system/routerboard/print
routerboard: yes
board-name: hEX PoE
model: RB960PGS
revision: r2
serial-number: HD1086861FK
firmware-type: qca9550L
factory-firmware: 6.48.6
current-firmware: 7.17.2
upgrade-firmware: 7.17.2
[admin@MikroTik_Switch] > system/routerboard/settings/set cpu-frequency=
500MHz 600MHz 720MHz 840MHz
[admin@MikroTik_Switch] > system/routerboard/settings/print
auto-upgrade: no
boot-device: nand-if-fail-then-ethernet
preboot-etherboot: disabled
preboot-etherboot-server: any
cpu-frequency: 720MHz
boot-protocol: bootp
force-backup-booter: no
silent-boot: no
protected-routerboot: disabled
reformat-hold-button: 20s
reformat-hold-button-max: 10m Re: Weirdly high temperature on new hEX PoE
They're both r2 ... and the cooler runs at higher frequency which is a bit counter intuitive.
So it may be about dried thermal paste/pads in hotter's cooling path.
So it may be about dried thermal paste/pads in hotter's cooling path.
Re: Weirdly high temperature on new hEX PoE
What confuses me, is that they don't have the same frequency options it seems like they are not the same even when they should be
Re: Weirdly high temperature on new hEX PoE
It's not a couple of degrees, so first thing Is checking whether those readings are real.
There are reports of devices that have false temperature readings because of failed/defective components.
Ideally you should use a contactless thermometer, but as a first approximation you can feel them with your hand, one should be warm, the other definitely hotter.
39 seems to me "right" and 63 definitely "too high", particularly if the second has no or little load.
Another thing to check is - if you can - the amount of power each device draws (at a same voltage) when totally idle (no ethernet cables connected), there might be a sensible difference.
The difference in frequency is "strange" but unless the readings are way off it cannot be related.
There are reports of devices that have false temperature readings because of failed/defective components.
Ideally you should use a contactless thermometer, but as a first approximation you can feel them with your hand, one should be warm, the other definitely hotter.
39 seems to me "right" and 63 definitely "too high", particularly if the second has no or little load.
Another thing to check is - if you can - the amount of power each device draws (at a same voltage) when totally idle (no ethernet cables connected), there might be a sensible difference.
The difference in frequency is "strange" but unless the readings are way off it cannot be related.
Re: Weirdly high temperature on new hEX PoE
be aware voltage supply can influence temperatures