Archived Forum

I have been able to get the emonTX shield working but only intermittently. I found the watchdog feature and used that but it only helps to get a few more transmissions before it just stops transmitting.

I am thinking I might need to replace out the RF module since nothing else on the shield is an active component.

Any ideas? Other things to try?

If I understand you correctly, when attempting to transmit data by radio, the sketch works for ~12 x 2 s, prints 12 lines to serial, then sticks. In other words, it returns from the call to send_rf_data(); for a while, then doesn't.

It seems as if the RFM12B module is failing.

Have you checked that you have all the components in the correct locations?

Have you soldered the RFM12B module in correctly - is the module aligned exactly with the pads on the main pcb (and is it the correct way round), and are all the joints good?

Are there any solder splashes causing a short anywhere?

Have you checked the supply voltages - especially to the RM12B module? (The maximum supply voltage is 3.8 V - it should be running on 3.3 V.)

Don't know if it's related to this:

Have you got the latest jeelib library?

See this thread: http://openenergymonitor.org/emon/node/1051 (read all 5 pages :) )

No problem like that has to my knowledge ever been reported with the emonTx. But that doesn't mean it can't happen, but equally I can't believe it hasn't been noticed if it does happen.

Thanks for the replies. I will check the supply voltages and I think I am going to try to put in a new radio module. I looked over my soldering and all looks well. I did also get the latest jeelib from git. I even just now tried a fetch to make sure I had the latest.

It is a little strange because these issues just cropped up when I did the install to my panel on my house. The datasheet on the RF12 says it can handle operating temps of 85C so maybe my panel box heated up too much and it got baked. 

Right now after more fiddling with it I am getting inconsistent readings but it doesn't just hang anymore. I took my laptop outside to see what was happening on the terminal and it returns some strange chars in part of my serial.print debug lines.

Ill keep a watch on it and when I get my new parts I will try again.

"it returns some strange chars in part of my serial.print debug lines"

As you suggest, if it did overheat, it could well have been damaged. I'd check the electrolytic capacitors too because they're normally only good to 85°C. And I'd suspect the crystal - if the serial output baud rate is wrong or varying, that could account for your garbled output.

It's nearly impossible to desolder the RF module without the proper tools - you need to melt all 14 joints simultaneously and quickly, otherwise you'll damage the parent board. If you're not planning to try to recover the module, I suggest you might want to use a burr or a router to very carefully destroy it in situ to leave just the 14 solder pads which you can then de-solder in the usual way. Beware, there are tracks underneath it.

I was going to use solder wick to get most of the solder off as well as my de soldering tools however the transmitter is  working as I have updated my sketch. I also put on a larger antenna (693mm) to make sure there were no reception issues.

I am not sure what I changed but the serial terminal still has the strange chars along with the rest of my normal serial print lines.