RasPi power usage measurements ALL Models

[fuffkin]

It would be really nice to get the Pi to boot quicker as it would open up a whole raft of applications like yours. I presume you'll have to run constantly to get the response you need or do you have a cunning plan?

I don't know enough about the OS side of things myself, but I live in hope that some software guru somewhere will come up with a really fast boot. To my simple brain, you've got an embedded processor and solid state disk; surely you could boot almost instantly? Maybe a Bare Metal approach is the answer, I don't know. I know I'm probably over simplifying things...!

[dom]

It would be really nice to get the Pi to boot quicker as it would open up a whole raft of applications like yours. I presume you'll have to run constantly to get the response you need or do you have a cunning plan?

Here's a demo of booting linux kernel and launching a game in about 3 seconds, so the boot can be quite fast:
http://www.youtube.com/watch?v=4Fjfqz6FxC8

[fuffkin]

Thanks for that link, that is very impressive!

[wildstray]

It would be really nice to get the Pi to boot quicker as it would open up a whole raft of applications like yours. I presume you'll have to run constantly to get the response you need or do you have a cunning plan?

Here's a demo of booting linux kernel and launching a game in about 3 seconds, so the boot can be quite fast:
http://www.youtube.com/watch?v=4Fjfqz6FxC8

Pretty impressive but I cannot reproduce... with a good SD, a Raspian 7.1 with unused daemons uninstalled, some optimizations (eg. cmdline serial console disabled, quiet, etc.) boot times are > 10 seconds. I need to be ready within a second.

[fuffkin]

Rumour has it, that if you use a USB stick for most of the OS files and filesystem and just use the SD Card as a read-only boot media you can speed up your boot some more.

I've not tried this yet and I suspect it'll still be a way off from your requirement of <1 sec. Has anyone tried this yet and what were the results?

[alexeames]

We seem to be drifting a bit away from the subject of power.

[wildstray]

We seem to be drifting a bit away from the subject of power.

...it doesn't seems so terrible to me The fact is that or RPi foundation (or anybody else good with reverse engineering) introduce a sleep/low power mode/hybernation or something similar, or the power drain is strictly related to halt/wake up and boot times.
Anyway, if there is a more appropriate topic about boot times I'll move there

[alexeames]

Fair enough. I'd like this thread to stay open though (and not get locked for going too far OT), as every time there's a new product launched, I like to put the power measurements here.

[hfc]

This thread has been kept alive for over a year...nice. The subject of power is of great interest to me and our organization, Hackers for Charity. We recently embarked on a path towards a truly portable educational classroom server. We are using a Pelican case to house all the components with externally mounted, waterproof RJ45, 2.1mm power and button for checking state and on/off.

You can check our progress here:
http://hackersforcharity.org/education

We are currently using Model B b/c of added RJ45 for updates, web connectivity (when available), etc. We require the use of one USB (high-power) for an AP offering up ~30GBs of educational material.

We are currently running off of solar, 110/220, or battery (tested 6Ah and 11Ah versions) with good success. Off the 6Ah battery, we run about 3 hours.

One of the products we use to handle ATX-style power control, is the ATXRaspi by LowPowerLab (http://lowpowerlab.com/atxraspi/). It when you push the power button for four seconds, it sends a proper shutdown -h now command to the Pi, monitors it until it enters its low power state, and then kills the power (conserving battery life).

If anyone has recommendations on solar, solar chargers or batteries, please let me know. I am always looking for ways to reduce power draw and extend run times (especially in low/no/erratic power locations, i.e. remote Africa).

Sam
sam@hackersforcharity.org

[Hix]

I bought Model A, I don't need networking but only to write pictures and videos on SD.
SleepPi is really interesting (and, let me say, a bit expensive...) but my problem is that I cannot wait at least 10s of boot time to have RPi ready. I don't need RPi always on, but I need RPi "always ready" My project involves a PIR sensor and I need to start taking from camera in milliseconds from trigger...

Hi - did you ever get anyewhere with rapidly booting the pi based on PIR trigger?
I am looking to build a remote wildlife camera which has near identical requirements. I'm aiming for a week's worth of running and have the camera film when the PIR triggers, recording ~3 mins of footage each time. If I could send the pi to sleep to conserve power this would be a win.

[alexeames]

So we have a new Pi in town. So some new power measurements are in order, particularly since the power circuit has been made more efficient. Let's see by how much.

So a side-by-side of the model B vs B+

Powered from a PiHub with HDMIPi screen on HDMI (also powered on PiHub).

with just kbd dongle...
Model B = 4.94 V 0.36 A = 1.78 W
Model B+ = 5.05 V 0.21 A = 1.06 W

with kbd dongle AND Edimax wifi dongle...
Model B = 4.93 V 0.41 A = 2.02 W
Model B+ = 5.03 V 0.25 A = 1.26 W

This is measuring the Pi's input, as with all my other measurements from previously (and using the same equipment).

So the new B+ certainly does use less power. Just over a Watt at idle with just keyboard dongle, VS the model B's 1.78 W. Nice!

[Douglas6]

That seems to be right in line with the announcement blog post: "Lower power consumption. By replacing linear regulators with switching ones we’ve reduced power consumption by between 0.5W and 1W."

Does that hold for the Pi in a shutdown state as well?

[alexeames]

That seems to be right in line with the announcement blog post: "Lower power consumption. By replacing linear regulators with switching ones we’ve reduced power consumption by between 0.5W and 1W."

Does that hold for the Pi in a shutdown state as well?

I'll look at that tomorrow

[jdb]

Alex,

Can you also post the results with the USB bus suspended (and thus the 9514 in a low-power state)?

doing a echo 1 > /sys/devices/platform/bcm2708_usb/bussuspend should turn it off (currently you can't get it *out* of that state once it's in it, but meh).

[alexeames]

Alex,

Can you also post the results with the USB bus suspended (and thus the 9514 in a low-power state)?

doing a echo 1 > /sys/devices/platform/bcm2708_usb/bussuspend should turn it off (currently you can't get it *out* of that state once it's in it, but meh).

OK will do

[alexeames]

Alex,

Can you also post the results with the USB bus suspended (and thus the 9514 in a low-power state)?

doing a echo 1 > /sys/devices/platform/bcm2708_usb/bussuspend should turn it off (currently you can't get it *out* of that state once it's in it, but meh).

OK done this now. It would only play ball after a sudo su. Running as sudo under pi it told me to get lost

So the scores on the doors are:
Before disabling USB, with just keyboard dongle:
5.03 V and 0.22 A = 1.11 W

With USB disabled
5.09 V and 0.15 A = 0.76 W

And then I got a chance to try out the "RUN" header to reset the Pi

[jamesh]

Hmm. Would be interesting to see camera power figures. I'm hopefully get a B+tomorrow so if no-one else tries will give it a go.

[jojopi]

(currently you can't get it *out* of that state once it's in it, but meh).

Do you mean that there is no longer a GPIO to the 9514's reset pin? It is still possible to unsuspend USB on a Model B.

[PiGraham]

Alex,

Can you also post the results with the USB bus suspended (and thus the 9514 in a low-power state)?

doing a echo 1 > /sys/devices/platform/bcm2708_usb/bussuspend should turn it off (currently you can't get it *out* of that state once it's in it, but meh).

OK done this now. It would only play ball after a sudo su. Running as sudo under pi it told me to get lost

So the scores on the doors are:
Before disabling USB, with just keyboard dongle:
5.03 V and 0.22 A = 1.11 W

With USB disabled
5.09 V and 0.15 A = 0.76 W

And then I got a chance to try out the "RUN" header to reset the Pi

Good stuff Alex. 0.76W will be very welcome. Good work RPF engineers.

I was curious about the disabling and possible re-enabling of USB and found an answer to the sudo puzzle:
http://www.raspberrypi.org/forums/viewt ... 72#p311072

It seems the enabling should be possible, but would either have to be on a timer or some gpio triggered code or UART terminal, since uSB keyboard , WiFi and Ethernet will be down.

It would be particularly useful if USB could be down most of the time and enabled when required with those 0.76W - 1.11W power levels.

If the camera draws negligible power when not running that would be good for Pi camera/camcorder type projects.

[RaTTuS]

...
OK done this now. It would only play ball after a sudo su. Running as sudo under pi it told me to get lost
...

sudo command >somewhere
is not what you expect
you need to
sudo sh -c 'command > somewheret'
there are other ways ....

[alexeames]

Hmm. Would be interesting to see camera power figures. I'm hopefully get a B+tomorrow so if no-one else tries will give it a go.

OK James. Have camera, will do it
My meter is charging, so it'll be a couple of hours before I can get you a result.

[alexeames]

B+ with camera & kbd dongle...

idle: 5.12 V, 0.21 A
video: 5.07 V, 0.27-0.33 A current seems very dependent on what you point it at. bright light = 0.33 A, darker subject 0.27 A

But at worst it's 0.12 A over the B+ itself.

Shutdown state:
5.23 V, 0.05 Amps

[boyoh]

Got a Pi from RS in the post today

I've seen several power related threads where people are wondering how much the Pi uses and a few have attempted to make measurements. I've got a fairly accurate piece of kit I bought for measuring the current/voltage use of RC planes, so I thought I'd put it to use here.

Here's the setup.
Power source 3 cell lipo (12.5v)
Connected to Hyperion E-meter calibrate shunt (and then on to emeter)
Connected to LM2596S switching reg (allegedly 90% efficient) stepping voltage down to 5.0V
Connected to USB A port
Containing Samsung USB phone connection/charge lead
Connected to the Pi

Pi was connected to...
Network cable
HDMI monitor
Power in (described above)

No USB for now. I logged into the pi using SSH without attaching USB devices.

Raw data (needs converting later)...

On booting, the Pi drew 0.20 amps @ 12.5v
Once booted, it dropped to 0.19 amps @12.52v

While logging in via SSH, current went up to 0.20 A again
then fell back to 0.19 A.

Plugging in Logitech wireless dongle for kbd/mouse
current went from 0.19 A to 0.20 A and stayed there.

Typed startx on wireless kbd.
While loading 0.21 A
Back to 0.20 A on idle (all @ 12.5V)

Start midori 0.21A while loading
Back to 0.20 Aonce loaded.

Load Raspi forum main page
0.21 A fell back to 0.20 once loaded

Load Raspi site main page
0.21 A fell back to 0.20 once loaded
then fell back to 0.19 on idle

Then shut down. 0.21 A while shutting down.

So what does this tell us?

1) Assuming the 90% efficiency is true, the idle power usage with just network and HDMI monitor connected is 0.9 * 12.5 * 0.19 = 2.14 watts

2) My logitech wireless kbd/mouse dongle required only 0.9 * 12.5 * 0.01 = 0.11 watts, which is nice and low

3) loading X or a web page or a program use a similar amount of power as 2) ~ 0.11 watts

I haven't had a chance to stress the unit yet, but will make further measurements when I do. I'm fairly confident of the accuracy of the measurements, but less so about the efficiency (90%) of the reg. But if it's less efficient, that means the pi uses less.

Bottom line. Pi uses about 2 watts at idle.

Stressed Pi measurements to come later. I've got a fairly inefficient python anagram algorithm I wrote a few weeks ago that gobbles up processor power. And I might hook up the USB data logging facility on the emeter if I can remember how it works to give you some pretty graphs and stuff.

All good fun. Hope this is useful for someone, but either way, I'm having fun.

Testing two identical Pi modules, would probably not give the same results
This might be due to component tolerance between the Pi.

If you are testing just the one Pi...Put a 10R resister in the 5v+ lead ,
Then test the voltage drop, then with a little help from Ohms Law, you will
Get what you want

[alexeames]

Testing two identical Pi modules, would probably not give the same results
This might be due to component tolerance between the Pi.
If you are testing just the one Pi...Put a 10R resister in the 5v+ lead ,
Then test the voltage drop, then with a little help from Ohms Law, you will
Get what you want

I already use a calibrated shunt

[chorlton2080]

Interesting thread. What is the most achievable configuration to minimise power consumption whilst still have the GPIO and SD card available for data logging? I note the methods for disabling the USB and Ethernet chip, can the video/audio subsystem be similarly disabled with tangible benefits on power draw?