Under-voltage warnings

[joan]

Is there any power warning for a non B+? I've seen one report of a red/orange warning on the top right of a screen connected to an ordinary B. Is this erroneous?

The red warning is over-temperature as does work on a model B. It means the chip has reached 85'C (and overclock has been disabled).
Try locating the Pi elsewhere (e.g. not on top of a receiver that generates a lot of heat), ensuring case has ventilation, or mounting the Pi vertically.

Ah, thanks. I had only heard of the over voltage warning.

[Burngate]

I've just edited my last post.
In fairness, the question only mentioned power. Though I suppose over-temperature could be construed as too much (thermal) power.
So an orange square means too little (electrical) power, a red square means too much (thermal) power. What level of what sort of power could give green or blue squares?

[dom]

I've just edited my last post.
In fairness, the question only mentioned power. Though I suppose over-temperature could be construed as too much (thermal) power.
So an orange square means too little (electrical) power, a red square means too much (thermal) power. What level of what sort of power could give green or blue squares?

There is no orange square. It's either red or rainbow, as described in the first post.

[meigrafd]

Hm im confused...

I thought the Voltage is not allowed to go below 4.75V? Is this limit with the new PI-Version now better/lower?
I also thought that the PI will reboot if the Supply Voltage gets below this limit - so how we than check GPIO35 befor it reboots?

Whats the different when i set GPIO25 manually to OUT? What else can i do with GPIO35?

Is this feature also with A+ available?


Thanx

[FTrevorGowen]

Hm im confused...
I thought the Voltage is not allowed to go below 4.75V? ...
Thanx

4.75V is the recommended/specified lower limit for reliable operation of attached (older?) USB devices. The Pi itself will carry on working at lower voltages** with reduced reliability below 4.6V ****. The B+ has better on-board regulation and current-load control for the USB ports.
Trev.
** At least one of my B's has continued working w/o problem down to 4.54V
**** Possibly network dropouts/issues when using the wired LAN?

[andrum99]

I am seeing what I believe might be a rainbow square under certain conditions. It is only when I try to boot Ubuntu Snappy Core on my Pi 2. The square is small and goes off after a few seconds, and the Pi stays on the rainbow splash screen. Colour-wise the square is a mixture of red, pinky-red and orange. Is this the rainbow square that you have implemented? It does not seem to be as big as the screenshots I have seen online. By small I mean that on my 14 inch TV it is about a centimetre in size.

[RaTTuS]

I am seeing what I believe might be a rainbow square under certain conditions. It is only when I try to boot Ubuntu Snappy Core on my Pi 2. The square is small and goes off after a few seconds, and the Pi stays on the rainbow splash screen. Colour-wise the square is a mixture of red, pinky-red and orange. Is this the rainbow square that you have implemented? It does not seem to be as big as the screenshots I have seen online. By small I mean that on my 14 inch TV it is about a centimetre in size.

that will be the power warning - if it only happens during boot and not all the time then it is probably ok ...
measure the voltage on the test pads to see what your getting

[grof]

Hi,

I don't know if this is the right place, but I have a problem.
I have 2 RPi2 devices and last couple of days I was running only one of them with success.
Yesterday I started using second one, which displayed rainbow square warning.

Both devices are using rip-branded 2A power supplies, so I assumed one of them is broken.

As a test, I disconnected the first RPi2 and replaced it so that test conditions are the same - rainbow square was still there.

I Measured voltage on across 2 and 6 GPIO pins - it's 4.79V.

What should I do? Should I return this second device, or should I just remove the warning and continue using it?

Thank you

[mahjongg]

sound like that on one of your PI's the polyfuse has triggered, that or the voltage that is presented to the PI is so marginal that it triggers the under voltage detector of one PI, but not the other. Normally the under voltage is triggered at around 4.65V, but a few percent each way of variance could be possible.

I suspect your power cable. (tinsel foil instead of copper wires inside).

you ignore low voltage warnings at your own peril...

[grof]

Hi, thanks for reply. I suspect polyfuse as well. Yesterday I tried connecting that Pi to my 10A led switching power supply which has a voltage-regulating potentiometer and can be boosted up to 5.5V, but I set it to 5.25V (USB top margin). It was powered through micro-usb connector, but still voltage was around 4.79 on GPIO.

How much time does it need for polyfuse to recover and what is the normal procedure (disconnect it, put it in a fridge (!?) )

Thanks.

[RaTTuS]

your micro usb lead may be built from wet string
putting it in the fridge will not reset the poly fuse any faster

[grof]

I'm not convinced it's the cable because other unit is working properly in same conditions.

[DirkS]

I'm not convinced it's the cable because other unit is working properly in same conditions.

That's no guarantee. If it's marginal it may work on one but not on another.

[grof]

Well, In my first post I stated that I also have a Raspberry-branded official charger and that first device works perfectly with it, the faulty one doesn't. So the cable isn't the problem.

I will try to power it over 5v GPIO PIN to get around polyfuse and let you know the results.

[clivem]

For anyone running headless, is it worth writing these errors to e.g. /var/log/syslog ?
Perhaps a software switch to write/not write them to log file...

Yep, as a starting point, would be useful to be able to query whether undervoltage or overthermal has been encountered via vcgencmd.

[Emexrulsier]

Just thought I would mention that I got a usb plug with my original pi that I got as a kit from Maplin. The charger is rated 5v 2.1a so I thought perfect for use on my pi2 yet I would always get the icon flashing up that I am under voltage regardless of what I was doing. I could just be sat staring at a desktop and the icon would come up every 30 seconds or so very annoyiny. Anyways I have since changed to a Samsung Travel Adapter (Can't remember now this either came with my Galaxy S5 or a Galaxy Tab3) which is rated 5.3v 2.0A and now all my voltage problems appear to have gone away. The usb devices I have connected are a Keyboard & Mouse dongle (logitech), xbox 360 wireless receiver and a Belkin N300 wifi dongle.

If anyone is interested the model number is EP-TA11UWE

[davidcoton]

rated 5.3v 2.0A and now all my voltage problems appear to have gone away.

Possibly. But if the low voltage was due to a poor cable, and you are still using that cable, you are relying on the loss in the cable to bring the voltage in spec. Should you ever change to a better cable, there is a risk (small, but present) that you will provide too high a voltage (the RPi is specified at 5V plus or minus 0.25V). I wouldn't say that damage was inevitable, but I would check carefully what supply voltage actually reaches the RPi .

[qwertinsky]

Oh great, now I am going to need a laboratory grade power supply and large gauge USB cable that has a micro USB connector on it.

How much more money is this $35 computer going to cost me?

Why in the frick did you chose a micro USB for a power connector instead of standard DC power connector that millions of 'wall warts' could plug into.

[ShiftPlusOne]

Oh great, now I am going to need a laboratory grade power supply and large gauge USB cable that has a micro USB connector on it.

Why? I've never had or needed such a thing for the pi.

How much more money is this $35 computer going to cost me?

$35 + shipping + tax and exchange rate padding, if applicable.

standard DC power connector that millions of 'wall warts' could plug into

Standard? What's the standard voltage for wall warts? Standard inner and outer barrel diameter? length? pin size? voltage polarity? current rating? AWG? I had a drawer full of wall warts collected over the years and no two were alike. With microusb power supplies, on the other hand, all you need to worry about is how trustworthy the information on the label is.

[davidcoton]

standard DC power connector that millions of 'wall warts' could plug into

That's a misattribution. I wouldn't wish barrel connectors on anything, let alone a Raspberry Pi.

All you need is a USB supply (5.00 to 5.25V, current to suit your setup but usually 2A is good) from a reputable supplier, with a decent preferably attached power cable. That allows for 0.25V drop on the cable, and no risk of overvoltage. It does, of course, depend on manufacturing tolerance, phase of the moon, etc, etc -- but there is no evidence here of any problems with that.

[mikerr]

TBH when the pi was first released I was one of the people who wanted mini-usb (more common back then )
- nowadays micro-usb is everywhere

[qwertinsky]

standard DC power connector that millions of 'wall warts' could plug into

That's a misattribution. I wouldn't wish barrel connectors on anything, let alone a Raspberry Pi.

All you need is a USB supply (5.00 to 5.25V, current to suit your setup but usually 2A is good) from a reputable supplier, with a decent preferably attached power cable. That allows for 0.25V drop on the cable, and no risk of overvoltage. It does, of course, depend on manufacturing tolerance, phase of the moon, etc, etc -- but there is no evidence here of any problems with that.

Well it seem I need more than that since I have tried micro USB adapters with current ratings between 1 and 3 amps and several micro USB cords, and plugging my keyboard and mouse into a powered USB hub. The best I could get is the rainbow square goes away only when during idle. Any activity and the square comes on and flickers.

[qwertinsky]

Obviously this problem is due to voltage drop in the USB cables . I am sure you would be lucky to find anything larger than 28 gauge wire in a micro USB cable. That can easily cause a .25 volt drop in a meter of wire at 1 amp.

I am going to modify my pi to accept a barrel connector so I can use a real power supply on it.

[rpdom]

Obviously this problem is due to voltage drop in the USB cables . I am sure you would be lucky to find anything larger than 28 gauge wire in a micro USB cable.

It seems you buy cheap micro USB cables.

Anything up to 20AWG is available if you look for it.

[davidcoton]

I am going to modify my pi to accept a barrel connector so I can use a real power supply on it.

As long as you remember to use only a 5V supply, if the voltage is switched tape over the switch. Banish all other PSUs with barrel connectors. And check polarity at least three times, on different occasions, before connecting! (I am writing from expensive experience.)

And, of course, never lend your modified Pi to anyone.

In this case at least sticking to the product design selection of a standard is in your own (and your Pi's) interests.