Hi. I'm measuring the 5.1v from the board where the power port is connected. I am aware that some (all?) Raspberry Pis have so-called "voltage measurement points", but I'm not interested in the voltages at various arbitrary locations on the board: I am interested in the supply voltage under load.ShiftPlusOne wrote:Where are you measuring the 5.1v, how reliable is the multimeter and does it have a fresh batteries?
Is the power warning (small colored box on top right of screen) on constantly?Sod Almighty wrote:Hi. I'm measuring the 5.1v from the board where the power port is connected. I am aware that some (all?) Raspberry Pis have so-called "voltage measurement points", but I'm not interested in the voltages at various arbitrary locations on the board: I am interested in the supply voltage under load.ShiftPlusOne wrote:Where are you measuring the 5.1v, how reliable is the multimeter and does it have a fresh batteries?
I don't know how reliable the meter may be. Lacking a magical device that can be relied upon to always output a known voltage, I know of no way to test it.
I have replaced the battery with a fresh one. I'm still getting the 5.1v reading.
A wireless keyboard dongle, some speakers, an ethernet cable, and a "PI supply switch" from http://www.pi-supply.com/product/pi-sup ... er-switch/klricks wrote:What else is connected to the RPi?
It is on when the PI Supply switch is connected, and I am running OSMC, and it is not in idle power-save mode (dimmed screen). If I disconnect the switch, the voltage jumps to something like 5.17, and the warning goes away.klricks wrote:Is the power warning (small colored box on top right of screen) on constantly?
It's an official Raspberry PSU by Stontronics, rated at 2 amps.klricks wrote:What is the mA rating of the power supply?
Actually, it would appear that the Pi 2 doesn't have any test points. I have, however, re-run the test from the fuse as suggested. The reading is now 4.85v. Oddly, the voltage from the port (that I was measuring before) is now also 4.85v. I'm confused.klricks wrote:Test at the fuse F1 to ground. The side closest to the edge of the board, (Or use test point PP1). That is power straight from the adapter (before the fuse).
Any other 5V on the board would be just as valid as long as the RPi is running. For example at the GPIO header pin 2 or 4.
Are you sure the RPi was fully booted and running when you did voltage tests? Also the power used by the RPi can vary depending on what is running.Sod Almighty wrote:It would appear that the dirt-cheap and unnecessarily-long USB cable http://www.pi-supply.com supplied with the switch is dropping 0.31v, compared to the switch circuit itself which only drops 0.02v.
Reckon a shorter cable would solve the problem? Or do I need a gold-plated one or something?
Incidentally, the voltage keeps dropping the more I measure it. The most recent reading is 4.7v. This makes no damn sense.
Pretty sure, yes. And the RPi was doing the same thing every time: sitting there in OSMC at the main menu.klricks wrote:Are you sure the RPi was fully booted and running when you did voltage tests? Also the power used by the RPi can vary depending on what is running.
Except that the culprit here seems to be the USB cable (not a defective cable, just a reasonably long one); and therefore bad connections aren't the issue.klricks wrote:Bad connections in power circuits tend to heat up and get worse over time.
You know, that's not a bad idea. I wouldn't try making an actual USB cable, because the signals degrade like hell if you use your own wires (I don't know why), but a power-only cable should be easy...klricks wrote:I am using a 5V 10A supply and still got a power warning until I made my own USB power cable. Got the micro USB end plugs off of ebay and soldered my own wires.
Hmm. I'll see if I can find that post.klricks wrote:FYI - The RPi has many test pads all over the bottom side of the board for various things. They are labeled with PPx designations. The primary purpose of these was to allow automated testing during manufacturing. Somewhere on the forums is a list of functions of each.
If you read any of the official documentation on the matter, it's not a problem until you get below 4.65 volts. Yours is at 4.8+ so is not an issue.
Most of the circuitry in the pi only needs 3.3v in fact, and the voltage is dropped to that level from 5v anyway.
The USB spec also allows for voltages down to 4.75 volts.
So what I'm saying is that it is not a problem if your device is seeing 4.8 volts. You don't seem willing to accept that fact, even though it is a published statistic from the manufacturers / designers.
Also, perhaps if you are worried you could try with a different USB cable like you said and see if that makes any difference? Do you have any other micro USB cables lying around you could test with?
I don't understand why you think that makes it "not fit for purpose"?
4.65+/-$some% actuallySod Almighty wrote:The "rainbow thing" is appearing. Therefore, logically, the voltage is dipping below 4.75v.
Except that he refuses to accept that I'm getting a low voltage. And that this must, by definition, be his fault.ShiftPlusOne wrote:He hasn't said anything wrong there,
What I have is PSU -> PI SWITCH -> USB CABLE -> PI.ShiftPlusOne wrote:It could be the cable, the fuse or anything else you have between the the supposed 5v source and the pi's 5v line. If you measure something that low at the actual supply output terminals while under load, then yeah, that's the problem. If you're seeing something near 5v on the output terminals and something much lower on the pi itself, the problem is elsewhere
I have gone over to using a PSU with inbuilt power leadSod Almighty wrote:"Is there anything else wrong other than the rainbow square? If not, then there isn't a problem is there?"
-- Aaron, http://www.pi-supply.com
Oh, I have one with a built-in cable. That exact one, in fact. But I'm using this clown's switching circuit as well, and that's causing voltage drops and rainbow warnings. And he just will not admit there is a problem.fruitoftheloom wrote:I have gone over to using a PSU with inbuilt power lead