NRF24L01 radio problems. (Bermuda triangle shit)



  • I was had a horrible project with a design that i build for "smart lamp" The idea was simple. Build a standard PCB that can be build into a standard light housing that can control the light, can read the output of a PIR, and sent temp/hum data. This is what i made:

    0_1482849340625_upload-7061811c-3893-4c2c-8303-33d6718a64dc

    It was working great until i tried 230V. The lamp would not turn on. After a few seconds a big flash killed the solid stat and the fuse in front of it. I had NO idea on what could be wrong here. Checked everything 10 times, placed a new solid state and fuse, and tried again. The same thing happened and i had no clue what was wrong. The lesson i learned eventually is that you should never take print terminals from a sample box without checking. the terminals for the output where internally fused together by design 😕. Meanwhile i was out of solid states but i had learned that the solid stat was not compatible with my LED bulb anyway because it needed a 100mA current to be able to switch of. My 3W led light did not supply such current. What a nightmare... The simple solution was to add a relay module so that problem was fixed. The module preformed without any problems now.

    On to the real issue...

    Next i installed the light outside. All the electronics where build in the base of the lamp which was plastic only. I used hot glue to fuse the electronics to the housing. (the glue is non conductive). Before screwing on the rest of the lamp (aluminium) i tested it again, but the module would not respond to Domoticz.

    The LED's on the arduino would behave strangely. The on LED would turn solid. Another LED started blinking, slowly. After a couple of seconds it stopt blinking, and stated again. Next the cycle repeated and repeated. I'm not used to this. Usually 1 led is always on, and the other blinks so fast it looks like a dimmed led. I do not understand a lot of the communication between the arduino and the radio, but i'm guessing the "dimmed" light must be the radio doing its job.

    The arduino is powered by the Hi-Link module. Its powering the 5V pin of the Arduino and is decoupled with 470uF and 100nF. The radio is connected the the 3.3V output of the arduino. Its also decoupled with 47uF and 100nF. I measured the DC voltage on both 5V and 3.3V, and its nominal. I also measured AC, and my cheap ass DMM gives a very small AC ripple on the 5VDC. I have no idea of the actual ripple as my cheap ass DMM is probably not fast enough. I do not expect there to be a actual noticeable AC ripple tho.

    That said, I could not get my creation to work. I switched the arduino, i switched to a different brand of arduino, i switched radio's, etc. In the end i got it working with this radio module. The kind of arduino didn't matter.

    I also experimented by putting the radio's in another in another Mysensors board, with the same arduino's. The same thing happened with almost all radio's. A serial print showed that the radio failed almost every time. I tried about 4 different boards without external antenna, and 3 with antenna. This one (from now on called straight antenna) and this one (from now on called angled antenna). I even took some radios out of older mysensors projects that i know where working fine.

    It seems like my build destroyed every radio except for the one with the straight external antenna. I have NO IDEA what could have caused this, but it looks like i now have a bunch of dead radio's. I'm not sure how to test this.

    Do you guys have any idea's on how this could have happened and how i should proceed in the future?


  • Admin

    Sorry to hear about your troubles. We all have gone through a few magic smoke moments. Frustrating but hopefully you learn a thing or two in the process.

    It's hard to say what the problem could be without looking at the circuit diagram of your board. Distances between hi/low power parts and how well you've decoupled things.



  • I for sure learned to double check any print terminal i take out of sample boxes. Damn...

    The 230V part of the board has a 5.08mm pitch for all lanes. I cut out the PCB pads and via's of the experimental board on both sided for insulation. The decoupling for 5V was done with 470u and 100n as close as possible to the Arduino. Same for the radio with 47u and 100n.

    There is only 1 real way to do this. Same goes for all the other connections. There are no other components in there. Except for the psu output and the radio voltage i cant think of anything that causes the issue with the radios.


Log in to reply
 

317
Online

6.7k
Users

7.6k
Topics

80.3k
Posts

Looks like your connection to MySensors Forum was lost, please wait while we try to reconnect.