Eh, seems I’ve been burned (literally!) by this as well. I’ve been testing different power supplies, connecting them the breadboard like this -http://www.dx.com/p/mb102-breadboard-power-supply-module-for-arduino-dc-6-5-12v-148125 and from that to the VIN and GND pins. At one of them the particle led went barely red and then stopped.
After that Photon is dead, and the chip just below the usb port gets really hot (like the original posters). I realized
I had the jumpers set to 3.3V, which seems to be not enough for the Photon VIN pin, but I didn’t expect it to literally burn out!
I’ll try to connect the voltmeter to the outputs to see what voltage actually caused it to break.
After one year of planning and testing I finally wanted to solder my first real prototype (apart from the breadboard), using the first one of my 11 Photons. They will drive my automated home. The basic idea is to use LED drivers to not just power the room LEDs, but also to power a photon and peripherals like rotary encoders, button, sensors, relays and a MOSFET (the latter to dim the room LEDs).
Since I can not drive the Photons directly with 12V, I purchased a couple of buck converters (standard china boards based on the LM2596S). My first idea was to directly use the 3.3V pin, since the buck converter should be more efficient than the on-board voltage regulator. But then I decided solder my voltage to pin VIN, expecting it might be too low to operate.
After powering up for the first time, the Photon started blinking dim white, as I would call it. I couldn’t find this in the documentation so far. So I guessed it was a low voltage issue and increased the buck converter output to 4V. But this didn’t help. I checked the output of the buck converter: 2.6V! So this was still to low. But no matter how I changed the output of the converter, the voltage stayed at 2.6V. After removing the Photon, the voltage changed! So I set it to exactly 3.80V and tried again. After connecting the Photon it again dropped to 2.6V. I tried disconnecting and connecting a couple of times, but it stayed like this.
After testing for a couple of minutes, I noticed that the LED stayed dark! I removed the Photon from the circuit and plugged it into USB. And it is still dark. Nothing gets hot, nothing happens. I let it hooked on the USB for a couple of minutes, I tried reconnecting, nothing helps. The Photon is dead. Then I noticed one of the two capacitors of the buck converter was lose and broke off. Could this be related to the problem?
I have a couple of Photons and may just try the next one, but I don’t want to burn them all. What have I done wrong? Is the Photon really completely dead? What if I connect the next one? Why did the voltage drop to 2.6V? Was it already misbehaving? Is it really that sensitive? My whole project is about to crash, what should I do now?
Specs:
LED-driver Mean Well LPV-35-12, 12V, 3.0A, no LED attached, just the buck converter
@Orkus, you did have it hooked to VIN and not 3.3v? I would say the output is likely very noisey, and the caps falling off will only make that worse, the inductor and cap work together to smooth the output. the noise likely interferes with the regulator on the photon causing it to oscillate badly and probably break other things.
Do you have an oscilloscope to see what the output wave form is? might pay to get a high power wire wound variable resistor to load it up and see what the output does under load too.
Does the photon still work when connected to a good quality power supply?
We have seen a fair few people have issues with power supplies that dont have a clean output.
Thanks for replying and happy new year everybody. I have been abroad and wasn’t able to look here until now.
Yes, i connected my power source to VIN, not to 3.3V, because I considered that to be safer, due to the additional regulator and the schottky diode). Sorry for nor clearing that up earlier.
I do not own any oscilloscope, just a multimeter. Which always stays camly at any voltage as far as I can say. In case of heavy noise or bad oscillating, I would expect the readout to change at least a little.
The Photon stays dead when connected to the same Sony Ericsson USB power supply on which it has worked before. I have not yet dared trying to circumvent the voltage regulator and directly power the 3.3V input.
Oh, and I forgot to mention, that a MOSFET was connected to A6, as well. The Photon was running the Tinker software.
I will now solder a second board with a working buck converter (with both capacitors). Then I have the following options:
set it to 3.3V and power the 3.3V input of the dead Photon.
set it to 3.3V and power the 3.3V input of the next Photon
set it to 3.8V (or 4 or 5?) and power the VIN input of the next Photon.
I also happen to have a large amount of TTH resistors of many kinds. So I could first try to measure the buck converter’s behavior under different loads. Any suggestions?
It would be great if there would be a BIG WARNING on a datasheet that low voltage can FRY a photon board. I did definitely not expect this to happen, I would expect it not to work, but not to fry the whole board…
Is the photon soldered in place or on a breadboard/header ?
The eagle part for smd soldering the photon have very large pads, so if too much solder gets deposited onto them the photon will shortcircuit beneth the board.
I took option 3: Soldered a new board with a healthy buck converter, set it to 5V, soldered headers to the next Photon, tried tinkering when hooked on the USB power supply, and then tried the same on my board. It worked. Phew.
I’ve never seen digital electronics that can be permanently damaged by lower-than-spec power voltage (assuming no high voltage short duration spikes or LDO dropout faults).
@Mora If you are asking me, the photon is with the headers on a breadboard.
I measured the resistance between 3.3V and GND and it is 0, which to my eyes means a short and while probably my 2nd power suppy blew out when connecting it to 3.3 pin and GND.
Ok, so not related to the solderpads, just wanted to put it out there, so noone else needs to make my mistake anew
near zero ohm is indeed short, there was some threads where people found the metal shield was bend and short circuited something under it, in a way that the shield could be used as a reset switch...
Be aware that these very cheap LM2596 boards are a great source of counterfeit chips. The “work”, usually, but (as you found out) many haven’t passed quality assurance checks; worse, the fake chips seem to fail in a full “pass through full voltage” mode in over-current situations (such as driving into a dead short…)
I would NEVER connect one to the unregulated 3.3v/5v pins on a MCU board (Photon, Arduino or otherwise) - the danger of a failure wiping out the whole system is to great.
I know @BDub took a look at this back in July 15’ and didn’t find an issue, but if you do learn more about what happened leading up to what you saw, or if you have steps to reproduce the issue we’d love to be involved and learn / improve from any bugs or anything.
I’m experiencing similar symptoms to the original poster (chip below usb port getting really hot). I’ve had the Photon working for a while using USB for power but recently decided to switch to powering via VIN, I was supplying 5v via a LM78L05 voltage regulator. I suspect I may have at one point connect the 5v to ground and ground to VIN
