Hi,
Can you post a fully connected schematic, including
Show power supply and please connecting wires.
Your diagram is okay for PCB manufacture but a hopeless "search a word" for trouble shooting.
Use gnd and 3V3 symbols, but please connect everything that is connected.
You have no bypass capacitors anywhere in your circuit, you need also to add the suggested capacitors to the 1702 as in the datasheet.
Tom... 
But they are 'burning out' in a most peculiar way. The I\O pins are apparently shorting, and not the same ones, which suggest that they, the I\O pins are getting an excess or reverse voltage of some type.
if you have the
First time, the DIO0 pin was shorted with Vcc.
This time with the changed circuit (reverse orientation), the MOSI and MISO pins are shorted to GND.
I have checked the connections again and again for any shorts or dry solder points but have found none. But if you insist then I can make another board and see if I get the same problem. Otherwise, I only have one option but to replace the MCP1702 regulator (which I don't have at the moment and is a bit difficult to get).
Another piece of info is that the circuit is taking around 150mA of current when transmitting (for a couple of hundred milli-seconds) and 30mA when the receive window (with a timeout of 20sec) is open. But if the current limit of the regulator was exceeded then the Arduino should reset or something, the RFM95W should not burn out IMO.
I have removed the power LED and on-board regulator of the Arduino Pro Mini. I thought maybe the pro mini's regulator already had capacitors connected to it. And since, I'm soldering the MCP1702 to the Vraw, Gnd and VCC pins of the Pro Mini, that I don't need external capacitors.
Do you think I will still need capacitors? @srnet thinks that some sort of overvoltage or reverse voltage killed the RFM module. Will capacitors be able to avoid this situation? Also, why isn't the INA219 or Arduino getting damaged from this reverse voltage (since they are also powered by the same regulator)?
Hi,
That is how you connect the 1702 by soldering it to those points, BUT the datasheet points out the need for those two capacitors.
From this reference;
snoa842.pdf (270.0 KB)
Even though TI say LM series do not need caps, their datasheet shows them in suggested applications. This is to bypass any noise, signal, RF current that maybe generated from the load that may influence the regulation.
Tom...
ok. makes sense.
Please let me know if my understanding is correct - The load here is the RFM95W. It is generating stray RF currents at times which in turn is causing a voltage spike on the MCP1702 as the filter capacitors prescribed in the application note are missing?
I will put these filters capacitors, use a new RFM95W module and get back with the results!
Hi,
Yes that is a possibility along with the digital noise from the UNO.
Tom....
I make my own Pro Minis, with an MCP1700\MCP1702 fitted, they only have the 10uF capacitors, never had a problem using them with LoRa RFM9Xs.
I use the MCP1700\MCP1702 so the board has a deep sleep currint on circa 1.5uA.
great. so that means theoretically my circuit should work fine.
Is it safe to program the Arduino when it is plugged into the circuit? Or do I have to remove it while programming it? I have programmed it while it is plugged into the perf board so far.
Also, I am unable to program the Arduino when it is connected to an external power supply. I have to remove the power supply otherwise avrdude reports a not in sync error.
Could this indicate that something is wrong with the Arduino? Or is this normal?
Not normal.
Can this issue be related to solder flux residue? I have not cleaned the circuit using isopropyl alcohol and I have used paste type flux (NOT no-clean flux) to solder the connections.
And is it too late to clean now since I have been using it for a month? I am also preparing a new board at the moment!
Make sure it is clean and shiny before soldering.
Tom...
The board is clean and shiny...but black residue gets built up between the pins. It is the solder flux paste residue I think. Anyway, here is the new one (in process):
Hi,
What solder are you using?
What soldering iron?
Thanks.. Tom...
Solder - 0.5mm 2%flux
Soldering iron - 25W regular soldering iron with a broad tip.
Lead free or Pb-Sn ?
Tom...
Not sure. The seller says 35% tin, so maybe Pb-Sn.
It's this onw: https://www.ebay.com/itm/402498569966
So, after almost a week of testing with the new hardware board that I made, I have not encountered any problems. In the new hardware board, the solder joints are cleaner and all flux residue has been thoroughly cleaned. The wiring is same as before. So, it can be concluded that it is definitely not a problem of the arduino, voltage regulator or RFm95W module.
The only possible cause of error that was causing the earlier board to fail seems to be a connection problem as @srnet had suggested from the very beginning. However, I'm unsure if this connection problem was being caused by the flux residue. After watching a couple of videos on the topic, I found out that flux residue can cause problems and unwanted behavior. But, is it supposed to happen after just a couple of weeks/days of use or does it happen after a time span of months? This is the only thing that is unclear to me.
Thanks @srnet and @TomGeorge for your support!
This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.
