My configuration has an Electron that gets 5VDC from a TI reference design switch mode power supply PCB, applied to the Electron’s Vin. The SMPS is capable of supplying 3A. In the real world, the SMPS will be sourced from a 12V battery on a boat. It is currently hooked to a lab power supply delivering 12VDC at up to 3 Amps.
When I connect the 12VDC to the board, the 5VDC comes up to 5.0 Volts, but the Electron does not start up. I get a dim red LED. The 3.3V pin reads 1.4VDC. So. figuring the 5V did not have enough current capacity, I tested the 5VDC supply. I was able to draw 2.5Amps (while it was still connected to the Electron), though with the 5VDC sagging to 4.5VDC. I attached a 330uF capacitor to output of SMPS and that did not change anything.
However, if I then connect the battery, the Electron comes up. If I subsequently disconnect the battery, the Electron stays running (and I get the rapidly blinking red LED since no battery is attached).
Is there some sort of startup sequencing I need to do?
Good to know but then the datasheet should be changed I guess - it appeared you do not need the battery. Maybe it is out of date. From the datasheet, a cut and paste:
VIN
The input voltage range on VIN pin is 3.9VDC to 12VDC. When powering from
the VIN pin alone, make sure that the power supply is rated at 10W (for
example 5VDC at 2Amp). If the power source is unable to meet this
requirement, you’ll need connect the LiPo battery as well.
I just powered my Electron up from my bench power supply set at 5V 2A max and the Electron in Safe Mode pulled a max of 350mA @ 5v and powered up just fine. The RED charging LED flashed because no battery was attached.
You may have a issue with the Power Management chip on your board if it keeps acting up.
I’m powering the Electron directly through VIN using a 5.5V 2A power source. It boots up fine but looses the cloud connection soon after I transmit anything through Particle.publish. I’ve tried adding a 100uF electrolytic capacitor along with a 1uF ceramic capacitor in parallel with the VIN/GND pins but that did not help. Any advice on how to get the Electron stable without a LiPo? Should I just give and connect the LiPo? The Electron is going to be a data logger permanently connected to a 12V power source so I though I don’t need the LiPo.
Yes, after Particle.publish the Electron changes to blinking green and after a while back to breathing cyan. The problem has to be the power source and it’s inability to handle the current spikes (Electron works fine using LiPo). I have a set of beefier 1000uF electrolytic capacitors and I’m going to try one next. I’m curious whether someone else has managed to get Electron stable without the LiPo and what kind of setup they have.
5VDC 3A is not telling the full story. The supply must be capable to supply 1+A very fast.
Since it works with the LiPo I suspect the response time of your supply is too slow.
For fast response time just going bigger on the caps doesn’t really help.