pingflip
Posts: 1
Joined: Wed Jun 20, 2012 7:32 am

power the Pi from 5V GPIO Pin possible?

Sat Jul 07, 2012 11:39 am

Hi,

Is it possible to power the Raspberry Pi from the 5V GPIO Pin (Pin 2) instead of the mico-USB port?

Cheers

pingflip

Fanjita
Posts: 17
Joined: Thu Jun 07, 2012 9:31 am
Location: Edinburgh, UK
Contact: Website

Re: power the Pi from 5V GPIO Pin possible?

Sat Jul 07, 2012 3:47 pm

I'm considering the same thing.

As I understand it, the 5V pin on the GPIO header connects to the SoC side of the input fuse from the micro USB 5V power input. So, it should be identical to connecting 5V into the USB socket, except that
  • You won't get the benefit of protection from that fuse, so in theory you could end up pulling too much current through the Pi. I'm not sure what circumstance that could happen in, however - some sort of accidental short circuit, I suppose.
  • On the flip side, you're not limited by that fuse, and could supply more current to e.g. USB devices (if you do something about the USB poly fuses).
Personally, I hope to power the Pi plus some motors from a single power supply, and this is probably the most convenient way for me to do that.

User avatar
jbeale
Posts: 3581
Joined: Tue Nov 22, 2011 11:51 pm
Contact: Website

Re: power the Pi from 5V GPIO Pin possible?

Sat Jul 07, 2012 4:12 pm

I've been doing that and it works fine. See below thread for a photo of a R-Pi running, with no microUSB cable connected (and very little else, actually!)
http://www.raspberrypi.org/phpBB3/viewt ... =44&t=9665

Ivovis
Posts: 4
Joined: Thu Nov 03, 2016 10:23 pm
Location: United Kingdom

Re: power the Pi from 5V GPIO Pin possible?

Fri Nov 04, 2016 4:55 pm

I know I am resurrecting a very old post but the OP question is still as relevant today as it was four year ago.

No one has suggested adding a fuse to the line supplying the GPIO, I have just had a rummage about and found the fuse below;

PPTC Resettable Fuse, MC36 Series, 750 mA, 1.3 A, 16 VDC, Through Hole, -40 °C

Available in the UK from here;
http://uk.farnell.com/multicomp/mc36245 ... dp/1861145

Now this fuse has a trip current of 1.3A the T075 on board fuse has a 1.5A trip current, as far as I can tell, you would have to have very heavy load on the USB sockets to even come close to this, so I am not expecting this small difference to be a problem.

The reason I have opted for this instead of an actual T075 is for ease of use on a breadboard prototype, does anyone have any reason why this is solution is a bad idea?

klricks
Posts: 6850
Joined: Sat Jan 12, 2013 3:01 am
Location: Grants Pass, OR, USA
Contact: Website

Re: power the Pi from 5V GPIO Pin possible?

Fri Nov 04, 2016 5:23 pm

Ivovis wrote:I know I am resurrecting a very old post but the OP question is still as relevant today as it was four year ago.

No one has suggested adding a fuse to the line supplying the GPIO, I have just had a rummage about and found the fuse below;

PPTC Resettable Fuse, MC36 Series, 750 mA, 1.3 A, 16 VDC, Through Hole, -40 °C

Available in the UK from here;
http://uk.farnell.com/multicomp/mc36245 ... dp/1861145

Now this fuse has a trip current of 1.3A the T075 on board fuse has a 1.5A trip current, as far as I can tell, you would have to have very heavy load on the USB sockets to even come close to this, so I am not expecting this small difference to be a problem.

The reason I have opted for this instead of an actual T075 is for ease of use on a breadboard prototype, does anyone have any reason why this is solution is a bad idea?
You need to look at hold current not trip current. Hold current is the maximum expected normal running current. A 'polyfuse' with 750mA hold was used on old RP 1B but is not enough for a RPi 2 or 3.
The 'polyfuse' on a 2B has a 2.0A hold and the 3B has 2.6A hold. You must not operate above the hold current or the RPi will became unstable due to partial trip condition. The trip current is usually 2X the hold current.
Unless specified otherwise my response is based on the latest and fully updated Raspbian Buster w/ Desktop OS.

J.Bryan
Posts: 42
Joined: Tue Aug 01, 2017 2:44 pm

Re: power the Pi from 5V GPIO Pin possible?

Wed Aug 16, 2017 3:34 am

Sorry for resurrecting this old thread again. I would want to ask about the 5v power on the GPIO header of Raspberry Pi 3(Pin 2 and Pin 4). If i power my Raspberry Pi 3 through the use of Lipo battery by connecting 5v to pin 2, and ground to pin 6 of the GPIO header, i know that i am bypassing the polyfuse.

I am assuming that the polyfuse current rating is at 2.5A as the recommended power supply for the Raspberry Pi 3 is 5V 2.5A.

However, can I add in my a fuse of 2.5A before connecting to the 5v power and ground on the Raspberry Pi 3's GPIO header? I was thinking to use this fuse to compensate for bypassing the polyfuse from the USB side.

Is it recommended to add a fuse, and will I get any problem from it? Does anyone know if there is a maximum current limit on the 5v pin if I want to power the Raspberry Pi 3 through the GPIO header (Example, I can't put more than 2.5A to the 5v power pin)?

Thanks for any help!

User avatar
OutoftheBOTS
Posts: 711
Joined: Tue Aug 01, 2017 10:06 am

Re: power the Pi from 5V GPIO Pin possible?

Wed Aug 16, 2017 7:01 am

I haven't been on this forum very long and this question keeps popping up :)

I personnel mainly power via the 5v GPIO pin because it is convenient for me as I am mainly having mobile robots powered by RPi brain and it is easier for me to chuck a 5v step down buck converter on the battery(usually a RC hobby Lipo) that runs the robot and run the RPi off the buck converter. I am using a micro 360 buck converter and the manufacture quotes capable of 3amps but anytime I have tried to pull 3 amps through this baby (running motor not RPi) it gets very hot and shuts down so I think at best it can do maybe 2.5amps anyway.

There also was some discussion on adding diode to protect against reverse polarity connection (hooking it up backwards tends to let the magic smoke escape)

J.Bryan
Posts: 42
Joined: Tue Aug 01, 2017 2:44 pm

Re: power the Pi from 5V GPIO Pin possible?

Wed Aug 16, 2017 9:03 am

Hi OutoftheBOTS,

Thanks for your input.

I bought a 5V 2.5A step down voltage regulator to step down my 11v battery to connect to the 5V Power pin of the GPIO header. I am still waiting for the regulator to arrive so have yet to try it.

I intended to connect more sensors and motors to the raspberry pi 3's GPIO pin. Just curious do you run your motor from the same battery that you use to power your Raspberry Pi? or do you have a separate battery for it. I will go and look into adding diode as suggested. Thanks!

User avatar
OutoftheBOTS
Posts: 711
Joined: Tue Aug 01, 2017 10:06 am

Re: power the Pi from 5V GPIO Pin possible?

Wed Aug 16, 2017 10:32 am

I normally run a second step down converter in parallel for the motors/servos as it is normally better to have the motors on separate circuit to the electronics so that when motors kick in and out it doesn't spike the electronics.

I usually have a very small mini 360 set-down converter adjusted to 5v for all the electronics and a bigger XL4015 5amp 75w converter adjusted to 6v for my servos and motors.

mkirkrbi
Posts: 1
Joined: Fri Aug 31, 2018 4:56 pm

Re: power the Pi from 5V GPIO Pin possible?

Thu Sep 06, 2018 5:38 pm

I am new to this forum and glad to see this discussion about powering the Raspberry Pi from the GPIO port. My question is specifically about the Raspberry Pi Zero. Regarding the discussions of including a polyfused inline before GPIO pin 2, I read somewhere else that the R-Pi Zero does not even have a polyfuse inline with the USB power port? Can anyone confirm or deny that statement? Based upon the first answer, do you recommend adding a polyfuse when powering from GPIO 2?

Thanks!

User avatar
ameador1
Posts: 89
Joined: Fri Feb 22, 2019 5:43 pm
Location: Eastern - USA

Re: power the Pi from 5V GPIO Pin possible?

Mon Apr 29, 2019 8:03 pm

Would this be a good option for adding a "fuse" to the 5V GPIO #2 for powering the RPi via the GPIO instead of the microUSB connector?

https://www.mouser.com/ProductDetail/Sc ... UIyYcK4%3D - Based on the comment before from klricks. This is in relation to a RPi 3B+.
"When money ceases to be the tool by which men deal with one another, then men become the tools of men. Blood, whips and guns–or dollars. Take your choice–there is no other..."
- Ayn Rand - Atlas Shrugged

User avatar
Burngate
Posts: 6225
Joined: Thu Sep 29, 2011 4:34 pm
Location: Berkshire UK Tralfamadore
Contact: Website

Re: power the Pi from 5V GPIO Pin possible?

Tue Apr 30, 2019 8:45 am

Trip current and hold current appear right, but the maximum voltage is only 6v, whereas the one used on the 3B+ is 16v

In normal use, that shouldn't make any difference - there shouldn't be more than 5¼v - but then, normal use isn't what the fuse is there for!
And we hear tales of people putting 24v on their Pi - the 16v fuse wouldn't protect in that instance, either!

User avatar
ameador1
Posts: 89
Joined: Fri Feb 22, 2019 5:43 pm
Location: Eastern - USA

Re: power the Pi from 5V GPIO Pin possible?

Sat May 04, 2019 5:55 pm

Burngate wrote:
Tue Apr 30, 2019 8:45 am
Trip current and hold current appear right, but the maximum voltage is only 6v, whereas the one used on the 3B+ is 16v

In normal use, that shouldn't make any difference - there shouldn't be more than 5¼v - but then, normal use isn't what the fuse is there for!
And we hear tales of people putting 24v on their Pi - the 16v fuse wouldn't protect in that instance, either!
So would using a similar polyfuse with a hold current of 2.6A, a trip current of 5.2A, and a max voltage of 24V be a good option - better than the 16V one on the RPi? The supplier I linked to doesn't have one with a max of 16V - at least not that I found. Is the max voltage the max voltage before the polyfuse itself is damaged? If say 20V were put on it - does that mean that up to 2.6A at 20V would make it to the RPi? That is why I was looking at the first one that was max of 6V - trying to keep a higher voltage from getting to the RPi.

How about this one: https://www.mouser.com/ProductDetail/Bo ... Pu0NjmQ%3D
The datasheet is here: https://www.mouser.com/datasheet/2/54/flsmf-777634.pdf
"When money ceases to be the tool by which men deal with one another, then men become the tools of men. Blood, whips and guns–or dollars. Take your choice–there is no other..."
- Ayn Rand - Atlas Shrugged

User avatar
Burngate
Posts: 6225
Joined: Thu Sep 29, 2011 4:34 pm
Location: Berkshire UK Tralfamadore
Contact: Website

Re: power the Pi from 5V GPIO Pin possible?

Sun May 05, 2019 10:18 am

I don't know.

When an ordinary fuse blows, the heat causes the metal to melt, and an air-gap forms. No residual current flows, and it cools down, but still no current flows even with any reasonable high voltage applied.
(Actually, in some circumstances, the metal vaporises instead of melting, and can form a conductive path inside the casing of the fuse)

When a polyfuse trips, it goes high-resistance, but not infinite resistance. The residual current flowing keeps it hot and in its high-impedance state.
But how high that resistance is, and how large the current can be for any given voltage, and what happens if you exceed the maximum voltage, is not something I've been able to find.

About all we can say is that far less than the trip current will flow, as long as we stay below the maximum voltage.

Oh, and we also know that RPT chose the polyfuse they did for a very good reason.
That may have been for very good electronic design and safety considerations, or it may just have been price and availability.

User avatar
ameador1
Posts: 89
Joined: Fri Feb 22, 2019 5:43 pm
Location: Eastern - USA

Re: power the Pi from 5V GPIO Pin possible?

Sun May 05, 2019 3:57 pm

Burngate wrote:
Sun May 05, 2019 10:18 am
I don't know.[...]
Well, maybe for safety sake - I'll go with the trip and hold current listed - but I'll go with the 12V max version. It sounds to me like that would cause the polyfuse to fail even before the one on the RPi. I'd rather have to remove the SMD polyfuse and replace it more often than replacing the RPi. There's no reason for more than a little over 5V to go to the RPi - so I can understand the 6V model I asked about before being a bit close for comfort in terms of damaging the polyfuse - but 12V - no reason to be near that and if the polyfuse fries - so be it.

Thanks!
"When money ceases to be the tool by which men deal with one another, then men become the tools of men. Blood, whips and guns–or dollars. Take your choice–there is no other..."
- Ayn Rand - Atlas Shrugged

Return to “Interfacing (DSI, CSI, I2C, etc.)”