Now i just need to build an 1.8V and 2.5V board and then do testing with the RPI board.
personally I'd not bother with the 2.5v one, going by the schematics it's only used for the composite video and doesn't look like it'll ever have significant current requirements, so an LDO will be fine.
yes your right, there is no point in the 2.5V
However, the reality is
there's a universal truth in there somewhere that reality never meets the spec.
I wonder how much of the conservative engineers outlook of working out what's required, doubling it and then adding a bit of extra margin is because he knows that somewhere there's a beancounter who'll force a reduction for 'cost' reasons and to hell with the spec.
The Pi being quite up front about everything being to meet the $25 price point probably suffers in that there was never any margin for the beancounter to cut...
Ok, agreed, that's one crappy USB powersupply that won't power a 'pi.
That case design is quite common for UK plugtop supplies for cheap electronic stuff. I have had several just like it, most of which died fairly quickly. Not hard to see why.
Guys. Just FYI, my raspberry pi boots just fine on 3.7V.
While slowly turning the voltage down, it crashed somewhere between 3.7 and 3.5V. That's when the "link" leds went off. It didn't come back up when I slowly turned the voltage back to the "last known working" voltage of 3.7 and it DID come back up when I powercycled it from there.
My 3.3V has dropped down to 2.7V. So the dropout of the LM1117 RG2, is about 1V. That's not very impressive for an LDO.
So, if you want to keep your 'pi running, you have to keep your 5V above 3.7V. This is "actual device measurements". For "nominal operation" (e.g. to have the 3.3V within spec), you need about 4.3V, to have some margin, good engineering practise would set the require met at "at least 4.5V" at the input.
But notice that you have to go almost a whole volt below that before the 'pi stops working. That's proper engineering.
The things that eat away at this margin are:
* crappy powersupplies that don't deliver 5V at 400mA.
* crappy powercables.
* the polyfuse.
Roger Wolff wrote:
Guys. Just FYI, my raspberry pi boots just fine on 3.7V.
My RasPi system works down to 4.65V. Below that my DVI-D monitor (connected through an HDMI Switcher) starts to misbehave. I guess your peripherals are a lot more tolerant of low 5V than mine LM1117 dropout is indeed about 1V according to its spec sheet.
Most 3.3V parts allow 5% - 10% tolerance, so I wouldn't let it go too low. The previous low voltage I've heard of was 4.2V, which reduces 3.3V to 3.2V. Below 3.3V you'll start losing Ethernet drive, but if you're only driving a 2m cable instead of 50m you'll be fine. I would think you'd want to have at least 4.3V so RG2 is producing a precisely regulated 3.3V instead of propagating any misbehavior on the 5V0 net.
Proper engineering would require you to do the test at both end's of the Pi temperature range.
It would also require to repeat the CE testing at the lower voltage, as it will likely make it more sensitive to radiation.
I really don't see what the test is supposed to prove?
The second test was usefull to prove that it isn't representative for all Pi's.
My test shows the raspberry is pretty tolerant of "low voltage" at the input. The only peripherals that I had connected during my test was the ethernet. I'm not pretending you should run your raspberry at 3.7V. I'm not working on qualifying the 'pi for that. It's just that when people say that the raspberry doesn't work if your power supply delivers only 4.5V I get worked up. It simply isn't true. The raspberry pi itself works just fine downto 3.7V. All sorts of peripherals, even ones that should conform to the USB spec and therefore work downto 4.2V, drop out way earlier.
I left my 'pi running at 3.7V and it's still running with an uptime of over 16 hours now.
If you blame the wrong component when troubleshooting, the acutal cause of the problem won't be fixed.
So, the raspberry pi works down to 3.7V on it's 5V line. This means you have to provide 3.9V at the USB connector for the raspberry to function. Given that lots of peripherals need more than that (and are also behind the polyfuse) the requirements for the powersupply and cable increase.
I have tested a 500mA USB powersupply from china. It doesn't achieve its spec. But nobody claimed 500mA USB powersupplies would work.
Ok, so in a different thread I threatened to remove RG1 and do some current measurements on it's output after seeing those thermal images that show it's not generating any heat...
Well, I did it tonight. Some photos here: https://picasaweb.google.com/selsinork/RPi18v
The jumper pins in the output let me either just put a jumper on and verify the Pi boots ok, or wire a multimeter in series to get some current readings.
The results were interesting to say the least. I had to go back and check I was reading the multimeter correctly, that it wasn't broken etc.
On initial power up I see a negative current for a second or so which then reverses to about 0.5mA (yes half a milliamp, that's not a typo) for a few seconds while we get the first sd-card accesses. Once we're booted and sitting at the login prompt the current reading fluctuates from around 0.001mA to maybe 0.04mA.
I'm using the 40mA range on a decent Fluke multimeter, so I've no reason to doubt the results. There's obviously going to be some inaccuracy down at that level due to length of meter leads etc, but the result is fairly clear. You'll understand why I was checking the meter was working and I was reading it correctly though
So from there onto the next test, lets try completely disconnecting RG1 and see if the Pi boots while using the LAN9512 1.8v 'output'. Yes it does!
I think that's reasonably good indication that jamodio got it spot on, the lan9512 shouldn't be connected to the 1.8v plane and it's heat problems are going to be largely due to supplying current on it's 1.8v filter pin that it was never designed to do.
So anyone willing to pull RG1 off a Pi and verify my results ?