Jump to content

OrangePi Zero high temperature?


Lauwenmark

Recommended Posts

Hello,

 

I've received my OPi Zero a few days ago. I've installed Armbian 3.4.113 (Debian Jessie) on it. Everything seems to be working fine.

 

However, I'm worried by the temperature of the thing. With just the default installation and no extra installed, I get the following:

 


root@orangepizero:/home/sysadmin# armbianmonitor -m
Stop monitoring using [ctrl]-[c]
Time        CPU    load %cpu %sys %usr %nice %io %irq   CPU
11:53:58:  912MHz  0.00   1%   0%   0%   0%   0%   0%   67°C^C
root@orangepizero:/home/sysadmin# armbianmonitor -m
Stop monitoring using [ctrl]-[c]
Time        CPU    load %cpu %sys %usr %nice %io %irq   CPU
12:00:13:  912MHz  0.00   0%   0%   0%   0%   0%   0%   66°C
12:00:18:  240MHz  0.00   0%   0%   0%   0%   0%   0%   66°C
12:00:23:  240MHz  0.00   0%   0%   0%   0%   0%   0%   67°C
12:00:28:  240MHz  0.00   0%   0%   0%   0%   0%   0%   66°C
12:00:34:  240MHz  0.00   0%   0%   0%   0%   0%   0%   67°C
12:00:39:  240MHz  0.00   0%   0%   0%   0%   0%   0%   67°C
12:00:44:  240MHz  0.00   0%   0%   0%   0%   0%   0%   66°C
12:00:49:  240MHz  0.00   0%   0%   0%   0%   0%   0%   67°C

 

Ambient temperature is around 25°C, and I've not put any cooling device on. Power supply is through the microUSB port using a 5V 2A USB charger. I've read at various places on this forum that the expected temperature in that case should rather be in the 30-40°C range.

 

Should I worry, or is it a perfectly normal temperature? If not, what could I try to lower it?

 

Thanks in advance!

Link to comment
Share on other sites

1 minute ago, Lauwenmark said:

I've not put any cooling device on

So how does it feel if you put your thumb on H2+ chip? In case it's not too hot how does 'armbianmonitor -m' output looks like when pressing your thumb on the SoC for a minute?

Link to comment
Share on other sites

31 minutes ago, tkaiser said:

So how does it feel if you put your thumb on H2+ chip? In case it's not too hot how does 'armbianmonitor -m' output looks like when pressing your thumb on the SoC for a minute?

It is quite warm - I can leave my thumb for several seconds on it, but I wouldn't leave it for a whole minute.

 

When I leave my finger for a couple seconds, the temperature gets a little lower, to around 61°C.

Link to comment
Share on other sites

This is my OPi Zero before latest updates: http://sprunge.us/Uacg (you can see at the end at which versions the installation was regarding u-boot and kernel).

 

We had a known miscalibration of thermal data on H3/H2+ for almost a year (IIRC it was related to reading out the so called SID from the CPU and since this was broken thermal readouts were 10°C-15°C off -- too low). I now did the upgrade but temperatures didn't change that much after the reboot with most recent u-boot version (but the SID change led to another well known problem: changing MAC addresses once

 

root@orangepizero:~# armbianmonitor -m
Stop monitoring using [ctrl]-[c]
Time        CPU    load %cpu %sys %usr %nice %io %irq   CPU
13:58:15: 1008MHz  0.26   1%   1%   0%   0%   0%   0%   55°C
13:58:20:  240MHz  0.23   2%   1%   0%   0%   0%   0%   54°C
13:58:25:  240MHz  0.22   1%   1%   0%   0%   0%   0%   53°C
13:58:30:  240MHz  0.20   1%   1%   0%   0%   0%   0%   54°C
13:58:36:  240MHz  0.18   2%   1%   0%   0%   0%   0%   53°C
13:58:41:  240MHz  0.17   2%   1%   0%   0%   0%   0%   56°C
13:58:46: 1200MHz  0.23  16%   2%   2%   0%  11%   0%   55°C
13:58:51: 1200MHz  0.30  16%   2%   2%   0%  11%   0%   58°C
13:58:56: 1200MHz  0.35  33%   1%  22%   0%   9%   0%   60°C
13:59:01: 1200MHz  0.48  50%   0%  50%   0%   0%   0%   61°C
13:59:06: 1200MHz  0.61  50%   0%  50%   0%   0%   0%   60°C
13:59:11:  240MHz  0.72  50%   0%  50%   0%   0%   0%   59°C
13:59:17: 1200MHz  1.06  61%   0%  60%   0%   0%   0%   63°C
13:59:22:  912MHz  1.30  61%   0%  60%   0%   0%   0%   67°C
13:59:27: 1200MHz  1.51 100%   0%  99%   0%   0%   0%   63°C
13:59:32:  240MHz  1.71  83%   0%  82%   0%   0%   0%   58°C
13:59:37:  240MHz  1.57  83%   0%  82%   0%   0%   0%   57°C
13:59:42:  240MHz  1.41  83%   0%  82%   0%   0%   0%   56°C
13:59:48:  240MHz  1.29   2%   1%   0%   0%   0%   0%   55°C^C
root@orangepizero:~# reboot
Connection to orangepizero closed by remote host.
Connection to orangepizero closed.
macbookpro-tk:~ tk$ slogin 192.168.83.138
tk@192.168.83.138's password: 
  ___                               ____  _   _____              
 / _ \ _ __ __ _ _ __   __ _  ___  |  _ \(_) |__  /___ _ __ ___  
| | | | '__/ _` | '_ \ / _` |/ _ \ | |_) | |   / // _ \ '__/ _ \ 
| |_| | | | (_| | | | | (_| |  __/ |  __/| |  / /|  __/ | | (_) |
 \___/|_|  \__,_|_| |_|\__, |\___| |_|   |_| /____\___|_|  \___/ 
                       |___/                                     

Welcome to ARMBIAN 5.25 stable Ubuntu 16.04.2 LTS 3.4.113-sun8i   
System load:   3.42            	Up time:       1 min		
Memory usage:  16 % of 241Mb  	IP:            192.168.83.138
CPU temp:      61°C           	
Usage of /:    8% of 15G    	

Last login: Sun Mar 12 14:01:55 2017 from 192.168.83.91
tk@orangepizero:~$ sudo armbianmonitor -m
[sudo] password for tk: 
Stop monitoring using [ctrl]-[c]
Time        CPU    load %cpu %sys %usr %nice %io %irq   CPU
14:02:43: 1200MHz  3.31  48%   4%   4%   3%  34%   2%   61°C
14:02:48: 1152MHz  3.29  47%   4%   4%   3%  33%   2%   60°C
14:02:53: 1200MHz  3.27  46%   4%   4%   3%  32%   2%   62°C
14:02:58: 1200MHz  3.16  46%   3%   3%   3%  32%   2%   59°C
14:03:14: 1200MHz  3.19  47%   3%   3%   3%  34%   2%   59°C
14:03:19:  240MHz  2.93  47%   3%   3%   3%  34%   2%   58°C
14:03:24: 1008MHz  2.70   2%   1%   0%   0%   0%   0%   58°C
14:03:29:  240MHz  2.48   2%   1%   0%   0%   0%   0%   57°C
14:03:34:  240MHz  2.28   2%   1%   0%   0%   0%   0%   57°C
14:03:40:  240MHz  2.10   2%   1%   0%   0%   0%   0%   55°C
14:03:45: 1200MHz  2.33  40%   0%  39%   0%   0%   0%   62°C
14:03:50: 1200MHz  2.46  40%   0%  39%   0%   0%   0%   62°C
14:03:55:  912MHz  2.59 100%   0%  98%   0%   0%   0%   69°C
14:04:00:  240MHz  2.38 100%   0%  98%   0%   0%   0%   58°C
14:04:22: 1200MHz  3.64  75%   2%  18%   0%  54%   0%   60°C
14:04:27:  240MHz  3.34  75%   2%  18%   0%  54%   0%   58°C
14:04:32: 1008MHz  3.08   1%   1%   0%   0%   0%   0%   57°C
14:04:38:  240MHz  2.83   1%   1%   0%   0%   0%   0%   56°C
14:04:43: 1008MHz  2.60   1%   1%   0%   0%   0%   0%   57°C^C
tk@orangepizero:~$ sudo armbianmonitor -u
/var/log/armhwinfo.log has been uploaded to http://sprunge.us/GKjY
Please post the URL in the Armbian forum where you've been asked for.

 

New log here: http://sprunge.us/GKjY 

 

Since my Zero has an applied heatsink but is in a small enclosure I would say temperature readouts match.

Link to comment
Share on other sites

Okaie, thanks. In the meantime, I've also tried replacing the power supply by another one (also a 5V 2.5A), and with the exact same setup, I now only get 49°C idle! So maybe the first power brick I used was not very good. I'll try to get a hold on better ones or design my custom regulated 5V psu block.

 

Your readings are indeed reassuring. I'll put a small heatsink on the Zero, and it should do it. I don't plan any heavy duty on those anyway.

 

Thanks a lot for your answer!

Link to comment
Share on other sites

My first OPi0 had temp not less than 55-60 C (without heatsink) without any load. Moreover, after set max cpu frequency to 1008MHz SoC temperature was above 50 in idle. Distro was Ubuntu legacy (3.4.113). Have to mention that I used WiFi without power-saving. 

 

Updated:

I also was wandered about so low temperatures provided on the forum. Putting OPi0 outside of home (in box) get 25 Celsius degree (it was about a zero outside :-). 

I think low temp can be caused custom build with unnecessary staff removed (moved to modules).

I'm looking forward for second ordered OPi0 to play with kernel optimizations (broke the first one). 

Link to comment
Share on other sites

Ok, I've made quick tests (I've got guests home today :) ). I've tested four different USB PSUs, all rated 5V >=2A.

 

With two of them, I get 48°C idle. With the next two others, I get 51°C, so I'd say that although there may be some effect due to the PSU itself, there isn't any significant variation.

 

After that, I've retried the first PSU (the one that got me 60°C results), and now I get constant 49°C! What I'm wondering is that if somehow, the higher temperature wasn't caused by the first run of the Debian? Maybe some sort of background process was left running that caused it? I have three other OPi Zeros, so I'll definitely test things further. I admit this left me with some head-scratching :)

Link to comment
Share on other sites

My OPi0 hovers around 40°C all day, no heatsink, no real load, just toggling i/o in response to serial port input from an arduino.
If I put my finger on the H2+ it feels warmish, but just the act of pressing my finger on it reduces the temperature to 32°C.
There is no heatsink, ambient temperature is low 20's.

  ___                               ____  _   _____
 / _ \ _ __ __ _ _ __   __ _  ___  |  _ \(_) |__  /___ _ __ ___
| | | | '__/ _` | '_ \ / _` |/ _ \ | |_) | |   / // _ \ '__/ _ \
| |_| | | | (_| | | | | (_| |  __/ |  __/| |  / /|  __/ | | (_) |
 \___/|_|  \__,_|_| |_|\__, |\___| |_|   |_| /____\___|_|  \___/
                       |___/

Welcome to ARMBIAN 5.25 stable Debian GNU/Linux 8 (jessie) 3.4.113-sun8i
System load:   0.02             Up time:       4 hours          Local users:   2
Memory usage:  20 % of 241Mb    IP:            192.168.0.35
CPU temp:      40°C
Usage of /:    69% of 1.7G

The PSU is a homemade affair that outputs 5v on the nail with enough current capacity to weld with :)

 

 

Link to comment
Share on other sites

1 hour ago, billybangleballs said:

If I put my finger on the H2+ it feels warmish, but just the act of pressing my finger on it reduces the temperature to 32°C.

 

So assuming you're still alive and your body operates at 37°C the simple 'thumb test' revealed that your board's thermal readouts are wrong (real temperatures must be 8°C or even 10°C higher)

Link to comment
Share on other sites

Hi,

 

1st OPi Zero (512 MB, heatsink, no case, capped to 768 MHz, room temp 15°C, 2A power supply via USB, ARMBIAN 5.25 stable Debian GNU/Linux 8 (jessie) 3.4.113-sun8i):

20:06:21:  240MHz  0.06   1%   0%   0%   0%   0%   0%   64°C
20:06:26:  240MHz  0.06   1%   0%   0%   0%   0%   0%   63°C
20:06:31:  240MHz  0.05   2%   1%   0%   0%   0%   0%   64°C

 

2nd OPi Zero (256 MB, no heatsink, no case, capped to 768 MHz, room temp 10°C, 1.5A power supply via pin1+2, ARMBIAN 5.25 stable Debian GNU/Linux 8 (jessie) 3.4.113-sun8i):

20:05:35:  240MHz  0.17   2%   1%   0%   0%   0%   0%   60°C
20:05:40:  240MHz  0.16   1%   1%   0%   0%   0%   0%   60°C
20:05:45:  240MHz  0.14   1%   1%   0%   0%   0%   0%   61°C

 

So there seems to be no effect of power supply.

However, it runs 50°C over room temp on idle... I wonder what will happen in the summer :huh:

 

To compare: OPi One, heatsink, case, room temp 15 °C, ARMBIAN 5.25 stable Debian GNU/Linux 8 (jessie) 3.4.113-sun8i:

20:11:09: 1008MHz  0.34   3%   0%   3%   0%   0%   0%   34°C
20:11:15: 1008MHz  0.40   4%   0%   3%   0%   0%   0%   35°C
20:11:20:  240MHz  0.37   4%   0%   3%   0%   0%   0%   35°C

 

Link to comment
Share on other sites

OK, maybe I was a little bit too fast yesterday... Tested another 2 power supplies, cpu is 10-15°C cooler now.

Checked voltage, the new ones (from my raspberry) deliver 5.25V, the old (from my tablet/smartphone) 5.1V.

So there seems to be an impact of the power supply on the cpu temp.

Link to comment
Share on other sites

1 hour ago, olti said:

So there seems to be an impact of the power supply on the cpu temp

 

Are you able to measure real voltage on test points (since PSU ratings...)? It would also be interesting whether other PSU characteristics (ripple, noise, whatever, I'm still an electrics noob) are important.

Link to comment
Share on other sites

Mine with a small 5v fan have temp beetwen 50-65c, I believe that it's just a coincidence that temperatures change with different power supply, but a few others test would be interesting. Anyway using the micro usb connector is not advisable, that pcb tracks are so small and the voltage drop drawing 2A maybe significantly. 

 

Update:

It stop working, no led turn on and the cpu have a small blob on it. Very disappointed, evidently it cannot work out the box without an adequate heatsink. 

 

Link to comment
Share on other sites

|PSU type|USB voltage|USB current|1.5V|3.3V|5V idle|5V with USB cam|
|Raspberry|5.25 V|250-300 mA|1.51 V|3.34 V|4.88 V|4.80 V|
|Smartphone|5.1 V|250-300 mA|1.51 V|3.34 V|4.70 V|4.60 V|
|Router (to PIN1+2)|-|-|1.52 V|3.34 V|4.90 V|4.85 V|

 

Unfortunately I have no equipment for further testing.

Link to comment
Share on other sites

Hi,

Just arrived my second OPiZ and I noticed that with absolutely the same conditions like room temp, PS, cable, etc new one is colder then previous (broken) in 10-15°C. During installing updates max temp was about 47°C without heatsink!

 

Link to comment
Share on other sites

On 3/31/2017 at 10:45 AM, krasoffski said:

Hi,

Just arrived my second OPiZ and I noticed that with absolutely the same conditions like room temp, PS, cable, etc new one is colder then previous (broken) in 10-15°C. During installing updates max temp was about 47°C without heatsink!

 

 

Call me crazy but I feel like my Orange Pi Zero decides to heat up or not randomly on every reboot. In the exact same place and env. temperature, only few minutes aparts and a reboot later it idles around either 40C or 50C depending on its mood.

Link to comment
Share on other sites

not sure how much i trust the temp readings, with the armbian installs on all my pi boards it jumps around a good bit , but on other installs it stays about the same, and armbian build temps do not a match with a thermometer.

heatsinks are cheap use one

Link to comment
Share on other sites

My OrangePi Zero idles at around 35-40 C (as shown by armbianmonitor -m, in ambient around 22 C) UNLESS I connect a RasPi GPIO Extension Board https://www.elecrow.com/gpio-extension-board-for-raspberry-pi-p-717.html to it, at which point it goes up dramatically.

 

I have soldered a latching right-angle 2x13 pin connector to the Opi0 with the intent of connecting things to it despite having the NAS board connected (which hinders access to the header holes). I want it to be a data logger for arduino-type things that don't have enough local storage.

 

If I attach just the 26-way ribbon cable, the temperature stays about teh same. When I add the T piece to the other end, the temperature climbs (usually after a minute or so of doing nothing). It has just peaked at 89 C, now fallen back again to 85 C. The CPU is shown as running at 240 MHz almost all the time, total load almost always <10%.

 

Note that the T piece is not at this stage plugged in to a breadboard, though that is the goal. I have previously plugged it into a breadboard and I recall seeing the temperature reach 98 C as I desperately tried to shut it down gracefully.

 

Anyone know what is going on here? These are just passive components I'm adding.

 

What maximum temperature should I be aiming for?

 

Thanks, and sorry if I should have created a new thread instead.

 

Link to comment
Share on other sites

Sounds like you're shorting out pins. I had a quick look at the product page and it looks like the 26 pin header is "converted" into 20 pins. So most likely a number of the pins of the 26 pin header are connected together. 

Put the pinot of the RPI header next to the OPI header to see which shorts you are creating.. ..

Link to comment
Share on other sites

Somewhere in the 60-70 degree range is to be expected from an unmodified OPi Zero. To reduce the temperature:

 

1. Add a heatsink of appropriate size.

2. Using the `h3consumption` tool, change the maximum CPU frequency to 912 MHz. This will significantly reduce temperatures because it avoids sending an entire 1.3V to the CPU.

Link to comment
Share on other sites

22 hours ago, pzw said:

Sounds like you're shorting out pins. I had a quick look at the product page and it looks like the 26 pin header is "converted" into 20 pins. So most likely a number of the pins of the 26 pin header are connected together. 

Put the pinot of the RPI header next to the OPI header to see which shorts you are creating.. ..

Thanks- did that ... of the 26 pins, 5 are GND and they are all in the same place on Opi0 as on RasPi 1. Then 2 are 3v3, and again these are in the same places. and 2 are 5v, again in the same places. So there are 9 'power' pins all matching, and they are brought out as 5v, 3v3 and GND rails and a separate GND pin. Leaving 17 'data' pins on the Opi0 interface, which are brought out to 17 data pins on the adapter. Now, these are not necessarily labelled consistently, but there is no reason to think that any of them need to be shorted to each other by design.

 

But your suggestion got me thinking harder... and I noticed that while these pins are "in the same places", actually the layout could be mirror-imaged depending on the way that the 2x13-pin adapter is keyed. so I cut off the key from one end and put it on upside down.... and the problem seems to have gone away.

 

So I imagine that the adapter shorts together the 5v, or the 3v3, or the GND pins somewhere inside.... and when you put the adapter on upside down, that translates to a short between two pins that should not be shorted.

 

So thanks for the inspiration!

 

 

 

Link to comment
Share on other sites

Just a note that internal human body temperature is 37C, not surface temperature - unless you're bleeding profusely. The interface between internal (37C) and external (ambient) must necessarily be inbetween. This is why we stick the thermometer in our mouth and not on our finger. I would say 32C is the correct reading.

Link to comment
Share on other sites

My Orange Pi Zero reaches 85C - seems like the record for this thread... At such temperature sometimes only 1 of the 4 cores are available, and the CPU frequency and reduced to 480Mhz. Don't reboot the OPi often, but when I do it usually works for some time, like 1 hour. Have an external ssd attached. Run Google Assistant, which takes some of the CPU to have the hotword detection. Have PLEX, which requires the OPi a lot when video transcoding is required. Quite hot down here in Brazil at this moment, reaching 35C room temeprature. No heatsinks (I know its required for my use, waiting to be delivered).

I will check if the PSU can help reduce the temperature. Wann try to disable the WiFI. How can I do this?

 

Link to comment
Share on other sites

8 hours ago, Alexandre Prates Dias said:

My Orange Pi Zero reaches 85C - seems like the record for this thread...

 

For this thread maybe but in other threads the numbers are even more off: https://forum.armbian.com/index.php?/topic/5250-improving-small-h2h3-board-performance-with-mainline-kernel/&do=findComment&comment=42050

 

 

Moral of the story: Seems to be numbers without meaning and there's something seriously wrong with calibration of the thermal readouts. Not fixable since all the people only report internal readings but not how hot the SoC is in reality (using a good IR thermometer or at least touching the chip)

Link to comment
Share on other sites

4 minutes ago, tkaiser said:

Not fixable since all the people only report internal readings but not how hot the SoC is in reality (using a good IR thermometer or at least touching the chip)

last Monday grg did wrote about his IR-readings and the thumb-test:

 

Link to comment
Share on other sites

On 16/03/2017 at 7:40 PM, olti said:

OK, maybe I was a little bit too fast yesterday... Tested another 2 power supplies, cpu is 10-15°C cooler now.

Checked voltage, the new ones (from my raspberry) deliver 5.25V, the old (from my tablet/smartphone) 5.1V.

So there seems to be an impact of the power supply on the cpu temp.

I've experienced something similar today.  I've checked my Orange Pi Zero (256MB) v1.4 numerous times over the past few days, and it's seems to follow a consistent pattern of working its way up to 58/59 degrees C over the first 10-15 mins, after which it remains at that temperature.  It's simply idling, not running any programs at the moment.

I then added on the expansion board with the 2 USB ports, and the temperate went up to 63 deg. C, and stayed there.

 

I was using a power supply with 2 USB outlets, which (allegedly) outputs 2.1 amps total across the 2 outlets.  It was powering the Orange Pi Zero and a Banana Pi.  Having read the above, I tried powering the OPi Zero (with expansion board) separately with a Raspberry Pi Charger I had handy (5V and 2.5A, I think, and probably better regulated), and the temperature levelled out at around 53 deg C - so a drop of 9-10 degrees.

 

System info:

ARMBIAN 5.38 stable Ubuntu 16.04.4 LTS 3.4.113-sun8i   

System load:   0.03 0.03 0.05      Up time:       9:05 hours        
Memory usage:  12 % of 241MB      IP:            192.168.1.163
CPU temp:      53°C               
Usage of /:    9% of 15G

 

 

Output from Armbianmonitor -m after changing power supply:
Time        CPU    load %cpu %sys %usr %nice %io %irq   CPU  C.St.
07:44:36:  240MHz  0.05   1%   1%   0%   0%   0%   0%   53°C  0/7
07:44:41:  240MHz  0.04   1%   1%   0%   0%   0%   0%   53°C  0/7
07:44:47:  240MHz  0.04   1%   0%   0%   0%   0%   0%   53°C  0/7
07:44:52:  240MHz  0.04   1%   1%   0%   0%   0%   0%   52°C  0/7
07:44:57:  240MHz  0.03   1%   1%   0%   0%   0%   0%   52°C  0/7
07:45:02:  240MHz  0.03   3%   1%   1%   0%   0%   0%   54°C  0/7
07:45:08:  240MHz  0.03   2%   1%   0%   0%   0%   0%   52°C  0/7

 

 

EDIT:

I thought I'd spoken too soon, as after running for around 10 hours at 53 deg C, the temperature climbed to 63 over 10-15 minutes.  But an hour or so later it was back down to 53.  Not sure what that was about.
 

I managed to lower it by another 5 degrees, to 48 deg C, by simply standing up the OPi Zero (with expansion board) vertically, on the edge with the pins (which is quite easy to do when it has the expansion board on it).  Previously it was sitting flat on a piece of wood, so very little ventilation to the underside.

 

Allegedly it should be fine up to 70 or 80 deg C, so 48 deg C doesn't seem TOO bad - although that's at idle, with no case, on edge; and sitting next to this OPi Zero is my Banana Pi working as an Openmediavault NAS (with pretty infrequent activity), running at a cool 34 deg.

 

2ND EDIT:

Still a bit unpredictable.  I booted up my OPi Zero last night and monitored it all night.  No case, stood on edge, and just idling.  Within 10-15 mins the CPU temp went up to around 57 deg C, and it stayed at that temp all night.  This morning I simply rebooted it, without waiting for it to cool down.  It went up to 47/48 deg C (only took a couple of mins, as it was already warm) and it's stayed at 47/48 for 4 hours.

 

Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
×
×
  • Create New...

Important Information

Terms of Use - Privacy Policy - Guidelines