Vin or 5v ? Arduino getting 1/2 power

Using Arduino IDE 1.x
1

Had an odd problem with my arduino Nano earlier, outputs that would normally be high (5v) were reading about half that. My power supply board delivers exactly 5v, but I've been wiring it to VIN on my nano anyway. So my first question is: Is that appropriate?. I have to do multiple units for this project and they are all wired/working that way currently (except this last one).

My workaround for this quirkly last module is to just move the wire from VIN to +5V instead- this seems to fix everything and the unit is now working perfectly.

Any suggestions? Thanks!
-k

2

Vin goes through the regulator. Regulators have dropout, the 1117 ones that the Arduino uses have a 1.something volts of dropout - you need like 7v on the VIN pin in order to ensure you're getting 5v out. You absolutely should not see 5v on the 5v pin, if you're supplying it with 5v on Vin, and it's not getting 5v from somewhere else.

You can power via the 5v pin, but if you plug it into USB like that it can damage the serial interface chip (I've only heard of this happening with official boards and clones using 16u2, but that doesn't mean it can't happen with clones with other serial interface)

The safer way to power off an external 5v power source is to supply the 5v via the USB cable (either from a USB charger, or by butchering a USB cable), that way you can't accidentally have it connected to external 5v and USB at the same time.

3

Thanks for the info! This unit will get power from both- usually it will come from the power supply board, and occasionally it will come from the USB (only during software updates). It's possible but unlikely that it would ever have both at the same time, but in my application it has to be this way.

Given all that, if you were me, would you chose to wire your +5v power wire (from the supply board) to the VIN or to the +5v pin?

4

salukikev:
Given all that, if you were me, would you chose to wire your +5v power wire (from the supply board) to the VIN or to the +5v pin?

As mentioned and explained above, 5V into Vin WILL NOT WORK. You've tested yourself, that it doesn't work.

If I were you, I would probably connect the 5V from the supply board to a usb cable. In that case, you can either have the USB cable from the PC or from the supply board plugged in, but never both at the same time.

Another solution (but not so fail-safe) is to ALWAYS power the arduino from the supply board, even during software updates. For this, I have a specially modified USB cable, where I cut the +5V wire, so that the PC can not supply 5V to the arduino.

And to explain why 2 different 5V power supplies at the same time are bad:
It's got nothing to do with the 16u2 chip, or "original vs. clone"

These 2 power supplies will never be EXACTLY same voltage. One will always be a bit higher. It will "fight" against the lower one, by supplying ever more current to increase the voltage. In this process, one or both of the power supplies might get damaged.

Note: 7 volt into Vin is also a 5V power supply, because these 7V go into the onboard 5V regulator. If this regulator burns through, then you will have 7V on the board. This can surely damage the 16u2 chip - and of course also the atmega chip.

I hope this clarifies the issue a bit.

P.S. Since this is a recurring issue for newcomers, I wonder if it would have been a better design idea to have a switch or jumper on the arduino board, where the user can select from the 3 possible power sources. On the other hand... you would probably get countless people who think their arduino is dead, because they forgot to choose the correct power source.

5

As I mentioned in my original post 5v into Vin HAS WORKED on the 12 units built so far- all wired to Vin, and all now with hours of running perfectly. I wouldn't have even thought to post about it here except that this one recent build has this anomaly, which I "fixed" by changing the 5v wire from Vin to the +5v directly. So my question was prompted by wondering if running through the regulated input would offer any protection or other advantages. I don't know what happened to cause this particular board to have issues, maybe its a factory defect, maybe something got shorted, spiked, or something else. Maybe it's good that if I have another board end up with the same issue that I have a backup fix to move to 5v direct. I don't really know.
I do understand your other explanations about power supplies fighting each other, but for this particular application, its going to be challenging to address them in the way you described due to logistics reasons rather than electronics or programming issues. I think the best solution is supplying these units with clear directions that avoid that occurrence.
Thanks for the help!

6
As I mentioned in my original post 5v into Vin HAS WORKED on the 12 units built so far- all wired to Vin, and all now with hours of running perfectly. I wouldn't have even thought to post about it here except that this one recent build has this anomaly, which I "fixed" by changing the 5v wire from Vin to the +5v directly.

You were lucky that the first 12 units worked. Go and measure the voltage on the high output on these working arduinos. Probably it is also very low. If it is (rather unexpectedly) 5V, then the 5V regulator on these 12 working units may be defective.
If it is very close to 5V, maybe your first 12 nanos have a low dropout voltage regulator? Do your nanos all come from the same source?

Also is it possible that your last build draws more current for some reason? In this case the low voltage problem will become more severe, and at one point apparent.

I would go so far as to say that the recent build is what to expect, and your 12 first builds are the anomaly.

(sorry for the messed up quote tags, I don't know how to get them working in Safari on the iPad)

7

I will measure as I'm almost done wiring another one- my expectation is that it will be just a bit shy of 5v. The thing about the one that didn't work is that the difference wasn't minor- it was right about half. Seems odd- if I had incoming AC I might suspect that it was only rectifying half the wave or something, but since it's DC, I don't know if there's some explanation that would cut voltage in half.. actually I should probably just stop speculating since I'm not an EE and just remembering random circuit explanation... haha.

Anyway, let me take some measurements once I get the next unit built, and I'll return with some numbers.

8

So if I run through Vin my pin (D9 if you want to be specific) is 4.3v So enough to run everything and seem pretty normal, but something definitely happened to the one nano that was outputting around 2.5v on every pin.
So back to my original question- I was really just wondering if there was any benefit to running through the regulator- specifically thinking of filtering benefits, but I think my answer is no. I should be running my 5v power supply directly to the 5v pin (which, as you would expect yields 5v at D9).

9

Not meant to bump old thread. But how did this work out for you?

I've had a 4 year project and device running for 4 years that operates this way. There is an ext. 5V reg on the motherboard and i occasionally update the device using usb. When the usb is inserted the device restarts immediately. I didn't think this was an issue.

I'll have to look on the nano schematics to see how i can bypass the ext. 5V reg and use the one on the motherboard.

But then again, this might have wored because the ext. 5V reg would be putting out 4.3V after it runs through the onboard regulator. So both the USB and the onboard 5V would be at about 4.7 and 4.3v respectively (i believe a schottkey diode is used on the nano so the dropout is not as bad),
The USB will win, i presume.