Use a RPi3 with 5.5V

seprim
Posts: 5
Joined: Tue Mar 15, 2016 8:06 am
Location: Merzig

Use a RPi3 with 5.5V

Hello
Is it possible to supply an RPi3 with an 5.5V voltage regulator?

Thanks
You can never solve problems with same reasoning, by which they were created. (Albert Einstein)
I do not agree with the mathematics. I mean, that the sum of zeroes is a dangerous number. (Stanislaw Jerzy Lec)

RaTTuS
Posts: 10415
Joined: Tue Nov 29, 2011 11:12 am
Location: North West UK

Re: Use a RPi3 with 5.5V

5.5V is out of spec , 5.25V is max, YMMV but the polyfuse will blow,
what is wrong with supplying a correct voltage level?
How To ask Questions :- http://www.catb.org/esr/faqs/smart-questions.html
WARNING - some parts of this post may be erroneous YMMV

1QC43qbL5FySu2Pi51vGqKqxy3UiJgukSX
Covfefe

sora03
Posts: 193
Joined: Mon Dec 29, 2014 4:11 pm
Contact: Website

Re: Use a RPi3 with 5.5V

seprim wrote:
Fri Nov 24, 2017 9:39 am
Hello
Is it possible to supply an RPi3 with an 5.5V voltage regulator?

Thanks
I am currently powering my Pi 2B to 5.7V since it is powering a hard drive (2.5" SATA-USB). I use the MP1584 regulator, to test power I use the program (rpi-burn hosted at https://github.com/ronny-nilsson/rpiburn also test using \$ vcgencmd get_throttled) try use those softwares to determine if the Pi is happy at 5v then if it is throttled increase the voltage at a time.
NOTE: using a thick wire for microUSB will mitigate power problems ( a cheap USB to microUSB cable has too much resistance and thin wiring) there is always a voltage drop when the Pi is on-load.
Last edited by sora03 on Sat Nov 25, 2017 11:14 am, edited 1 time in total.
FLOSS enthusiast, Debian user (linux-libre)

joan
Posts: 14196
Joined: Thu Jul 05, 2012 5:09 pm
Location: UK

Re: Use a RPi3 with 5.5V

It is certainly possible. It is outside the USB specification of +/- 0.25V but I doubt any device you connect will care. The Pi itself should be okay. I've powered from 5.8V, although I would not knowingly go any higher, get too close to 6V and you may kill the Pi.

drgeoff
Posts: 9747
Joined: Wed Jan 25, 2012 6:39 pm

Re: Use a RPi3 with 5.5V

sora03 wrote:
Fri Nov 24, 2017 10:07 am
seprim wrote:
Fri Nov 24, 2017 9:39 am
I use the program (rpi-burn hosted at https://github.com/ronny-nilsson/rpiburn also test using \$ vcgencmd get_throttled) try use those softwares to determine if the Pi is happy at 5v then if it is throttled increase the voltage at a time.
That is nonsense on two counts.

1. The Broadcom chip gets throttled when it gets too hot. Increased voltage does not mitigate that.

2. The RPi has onboard voltage regulators supplying the Broadcom chip. Increasing the voltage of the 5 volts supply to the board does not significantly increase the voltages applied to the chip.

sora03
Posts: 193
Joined: Mon Dec 29, 2014 4:11 pm
Contact: Website

Re: Use a RPi3 with 5.5V

drgeoff wrote:
Fri Nov 24, 2017 5:08 pm
sora03 wrote:
Fri Nov 24, 2017 10:07 am
seprim wrote:
Fri Nov 24, 2017 9:39 am
I use the program (rpi-burn hosted at https://github.com/ronny-nilsson/rpiburn also test using \$ vcgencmd get_throttled) try use those softwares to determine if the Pi is happy at 5v then if it is throttled increase the voltage at a time.
That is nonsense on two counts.

1. The Broadcom chip gets throttled when it gets too hot. Increased voltage does not mitigate that.

2. The RPi has onboard voltage regulators supplying the Broadcom chip. Increasing the voltage of the 5 volts supply to the board does not significantly increase the voltages applied to the chip.
1. the throttling that I refer is when the RED LED blinks indicates insufficient voltage (<4.65V) not the CPU temp
2. yes but the USB peripherals will reduce the voltage (USB WiFi,USB HDD, USB 3G modem if not plugged into Powered USB Hubs)
FLOSS enthusiast, Debian user (linux-libre)

drgeoff
Posts: 9747
Joined: Wed Jan 25, 2012 6:39 pm

Re: Use a RPi3 with 5.5V

sora03 wrote:
Sat Nov 25, 2017 5:38 am
drgeoff wrote:
Fri Nov 24, 2017 5:08 pm
sora03 wrote:
Fri Nov 24, 2017 10:07 am

That is nonsense on two counts.

1. The Broadcom chip gets throttled when it gets too hot. Increased voltage does not mitigate that.

2. The RPi has onboard voltage regulators supplying the Broadcom chip. Increasing the voltage of the 5 volts supply to the board does not significantly increase the voltages applied to the chip.
1. the throttling that I refer is when the RED LED blinks indicates insufficient voltage (<4.65V) not the CPU temp
2. yes but the USB peripherals will reduce the voltage (USB WiFi,USB HDD, USB 3G modem if not plugged into Powered USB Hubs)
No-one except you refers to the red LED blinking as throttling. Everyone except you knows throttling refers to the CPU speed.

If the voltage drops under load, then either the power supply is deficient or the connecting wires have too much resistance. Increasing the power supply voltage is not the way to remedy those.

davidcoton
Posts: 4031
Joined: Mon Sep 01, 2014 2:37 pm
Location: Cambridge, UK

Re: Use a RPi3 with 5.5V

drgeoff wrote: No-one except you refers to the red LED blinking as throttling. Everyone except you knows throttling refers to the CPU speed.
Except that, IIRC, one of the consequences of detecting low voltage is to reduce the clock (throttle) in an attempt to reduce power consumption. There is of course absolutely no connection with a marginally high supply voltage, which will not induce throttling (because the voltage is NOT low, and the 3V3/1V8 regulators ensure that the SOC does not receive increased voltage, so does not overheat and throttle for that reason).

So, while there is a connection between low supply voltage and throttling, it is irrelevant here.
drgeoff wrote: If the voltage drops under load, then either the power supply is deficient or the connecting wires have too much resistance. Increasing the power supply voltage is not the way to remedy those.
Increasing the supply voltage while remaining in spec may mitigate cable losses -- even the official PSU has an above nominal voltage for that reason.
It is poor practice to raise the supply voltage above the spec maximum (5.25V) because of:
1. measurement uncertainties,
2. not knowing the exact point at which problems occur,
3. possibly causing an overvolt if the current consumption drops.
The correct solution is the combination of a properly regulated supply capable of the required current at the correct voltage (4.75V to 5.25V at the Pi input), and a short, thick (=low resistance) USB power cable.
Signature retired