emonTx restart every 3 seconds

hi, I builded an emontx v2 which I used for one month with a ds18b20 and a battery pack with no problems. after I started using it as monitor fv/consumption in my sister's house, with a emonglcd (this is my third emontx with emonglcd to monitor fv/consume).

that I see is that after few days turned on, emontx simply reboots after 2-3 seconds infinitely; I can see on serial output only the first rows and the restarted again; if I leave emontx turned off for one day, after I can see that emontx goes ok, but after few hours start again to reboot.
It seems that when it become "hot" it starts rebooting infinitely.

this my tests with no results:

- I changed power supply;

- change image of atmega;

- left emontx without sensors connected and it continues rebooting.

suggests? thanks

Robert Wall's picture

Re: emonTx restart every 3 seconds

Is "fv" photo-voltaic generation?

Does it work in your sister's house when powered by batteries? And when powered by the programmer? Are you using a power supply you bought from the shop or from somewhere else? See "Not all USB power adapters are created equal - highlighting importance of selecting good quality power adapters"

Can you measure the power supply voltages - when first powered, then when it starts to reboot?

"It seems that when it become "hot" ... " Which components are hot? What temperature (approximately) do they reach?

Which sketch are you using?

modenet's picture

Re: emonTx restart every 3 seconds

yes, fv is for photo-voltaic production.

I haven't tried to power by batteries.

I have tried 3 different power supplies (one was bought in this shop) and every times emontx started rebooting after some hours; even powered by the programmer I saw through the serial output that it was restarting continuosly.

I'm using emonTx_CT123_Voltage sketch.

ps when I told "hot" I was not indicating the temperature but that it become unstable after some hours. if I keep turn off for one day I can see that I work for some hours, after it becomes unstable.

 

Robert Wall's picture

Re: emonTx restart every 3 seconds

I have not seen this problem reported before. We do not know for certain that the problem is new, or whether it has always been there but never appeared because you never used a mains power supply until now. I think it would be very useful to try powering it with batteries to see if runs properly as it did before, or whether it reboots.

modenet's picture

Re: emonTx restart every 3 seconds

hi I'll come back after some weeks, I did more tests: with battery and ftdi power supply emontx keeps reboot, even with no CT and AC connectors.

emonTX CT123 Voltage example
OpenEnergyMonitor.org
Node: 9 Freq: 868Mhz Network: 210
1516 -16451 55.40
0 760 0.30
0 0 0.23
emonTX CT123 Voltage example
OpenEnergyMonitor.org
Node: 9 Freq: 868Mhz Network: 210
1527 -14416 55.59
0 827 0.31
0 0 0.30
0 0 0.23
emonTX CT123 Voltage example
OpenEnergyMonitor.org
Node: 9 Freq: 868Mhz Network: 210
1516 1497 55.38
0 0 0.25
emonTX CT123 Voltage example
OpenEnergyMonitor.org
Node: 9 Freq: 868Mhz Network: 210
1492 1483 54.96
0 0 0.22
emonTX CT123 Voltage example
OpenEnergyMonitor.org
Node: 9 Freq: 868Mhz Network: 210
1513 -17339 55.33
0

modenet's picture

Re: emonTx restart every 3 seconds

I modified the sketch: if I disabled this line my emontx keeps running (without data over rfm12b)

send_rf_data();                                                       // *SEND RF DATA* - see emontx_lib

 

so the problem was the rfm12b module; I re-soldered again all points and now  seems to be running from 8 hours.

calypso_rae's picture

Re: emonTx restart every 3 seconds

If there are any problems with the RFM12B connections, the main program (sketch) is likely to stop running.  I think this is because its waiting for a reply that never arrives.  If the RF chip is not present, the code won't get past the initialisation stage in setup().

I have always found it very difficult to attach the RF module to the emonTx V2 board.  A good way that I have found is to apply lots of solder so that the pins are shorted together, and then remove most of the excess solder using solder-wick.  The resulting joints are always good because enough solder gets sucked underneath the pads by capillary action. 

Robert Wall's picture

Re: emonTx restart every 3 seconds

I just put enough on so that there is a triangular fillet (http://en.wikipedia.org/wiki/Fillet_%28mechanics%29) of solder between the plated-through half-hole on the edge of the RFM12B and the PCB.

calypso_rae's picture

Re: emonTx restart every 3 seconds

The tip of my soldering iron is not ideal  for fitting the RFMB to the emonTx V2, but the solder-wick approach provides an easy workaround.

In a factory, surface-mount components are generally fitted by running the board across a solder bath followed by an air knife to remove the excess.  My solder-wick idea is doing much the same job on a smaller scale.

On my own PCB, I use a different method of attaching the RFM12B device.  I find this approach to be much easier because the connection can be made at the top of the pads, rather than at the bottom where it can be difficult to get the solder to flow reliably.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.