#1: Bootloader got corrupted, no longer accepted uploads. Erratic behavior. Eventually gave "content mismatch" in the IDE.
#2: Voltage regulator burnt somehow with an input of 9V.
#3: Processor got borked. Only the power light was on. No activity on reset. Upload didn't work; IDE pretended it didn't exist.
#4: Onboard NRF died. Board threw a "programmer out of sync".
There must be a common cause, at least for 3 and 4 because they were on the same circuit. 1 and 2 were both on different circuits. 1 worked fine for a while but stopped working with a replacement board.
Show us your schematic - not a Fritzy diagram.
Burning out a regulator is unlikely to be the voltage ‘going in’, but more likely to be the current ‘going out’.
I would suggest as azon says "Please contact the seller directly for warranty information for this product. You may also be able to find warranty information on the manufacturer’s website." and ask for an exchange or your money back. I will take a SWAG and say you are drawing more power then the regulator can supply. Things are pointing to that. there is a lot of chatter on the capability of the Arduino 3v3 to power the nRF24L01. I believe the libraries defaults to max power pushing the regulator. This link has some information that may help: Using nRF24L01 with the maximum power mode. Is there anything else connected to that board?
I doubt that. The bottom side of that pcb shows a 3 terminal regulator, likely a 1117. This should be easily capable of powering an rf24 also at max PA power. Unless of course the 3.3V line is loaded excessively by the rest of the circuit.
Another possible issue could be excessive noise or spikes on the 3.3v line, destroying the usb/uart bridge. The symptoms really point toward something that kills the ch340, which is rather sensitive to begin with.
@lastchancename I will draw up a diagram of the circuit I'm currently using and then the one I plan to switch to at some point, but I'm pretty busy.
@gilshultz This is a random Chinese board so I doubt there's any warranty, especially after I desoldered the pins. I am running the NRF at max power but I have had no issues with that before.
@anon35827816 Voltage spikes on 3V3 are definitely possible because one of the things the Arduino is controlling is a BLDC motor which contains rapidly spinning magnets. Multiple times the motor got close to the nano during testing making me wonder if some tiny amount of induced voltage corrupted some data on the ATmega, created noise on 3V3, or something else. This is currently the cause that I think broke all these devices. How would I rectify this? Wrap the RF nano in foil?
EDIT - "All these devices" excludes the one with the broken 1117 because I think that was due to a dumb mistake on my part.
Nah. You won't destroy a chip on your Nano board through pure magnetic interference; the magnets you're likely to use are waaaaay to weak for that.
I'd sooner suspect something in how the motor is powered and possible back-EMF making it to the Nano board. As always, schematics help a lot.
@anon35827816 It's an optocoupler type ESC, so that's not it. The Arduino, however, is being powered through one of those cheap buck converters so I wonder if that is the issue, the buck converter giving a noisy input on VIN.
Hi,
What about your working environment?
Static electricity?
Power Supply performance?
I hope you do not work down on the floor, on a carpet?
Do you assemble using a breadboard?
How do you do your prototype assemblies?
Can you please post some images of your work environment?
Are you sure you power off and wait for your project to power down, before changing components?
Can you post a picture of your power supply and buck converter and how you have connected them?
I work on an ESD mat. The power supply is a 450mAh 75C 3S LiPo. This is being converted to 9V to power the Arduino through VIN but the battery is left raw for everything else.
EDIT - Input of battery goes to IN+ and IN- of converter and output of converter goes to VIN and GND on Arduino.
It's powered from the same source as your board. The optocoupler won't protect against gross problems on the supply line; the DC-DC converter should in principle, but no guarantees.
Anyway, it's a matter of systematically isolating variables. Start wirh running the rf nano board for an extended period of time just by itself, sending packets of data all the time. If that works, add the other components one by one until you start smoking up nano's again.
I tried process of elimination but that didn't seem to help. One setup would work, and then I would add something, and it would break. So I would get a new nano, try the first setup, and that would break. The common variable is always the DC/DC converter so I think it may be that. I do have a proper RC BEC on the way to try.
EDIT - I don't have a 'scope so there's no way for me to tell if the DC/DC converter is bad.
Your wiring is a royal mess and asking for trouble.
The photo is incomplete, the schematic contains something very weird around U3 (an rf24...?) which is being powered through its 3v3 pin from d13 on a nano (?) that for all I know runs at 5V (?).
Stuff doesn't add up as far as I can tell from here. It's at least inconsistent and chaotic. Meh.
U3 is the onboard NRF. The photo is incomplete because I disconnected the nano. The D13 thing is a result of not paying enough attention while making the schematic. It's supposed to be on 3V3.
