EmonTx V3.4 no sign of life

gerombo · 2022-02-13T00:43:43.680Z

Hi,
I had a temporarily working setup with EmonTx V3.4 with 4 CTs, AC-AC adapter and an emonPi.
After a few calibration experiments the EmonTx is not showing any sign of life anymore:

not a single LED lights up when powering the EmonTX in AC mode nor with DC adapter in DC mode
no more output on FTDI serial console

I would appreciate any advice or pointers on how to try to troubleshoot, repair (or abandon) the EmonTx.

additional info:
-I was attempting a power diversion project in a 3 phase installation with single phase solar. Because of limited load and limited return to the net, the Solar invertor keeps tripping (reaching 250-260V) Adding additional load on this phase prevented the invertor from tripping, hence PV diverging would have been 2 birds with one stone.

-could a 250+ V peak on the ACAC 9V adapter have damaged the emontx?
-in an attempt to calibrate the CTs and make sense of the readings, I had connected all 3 CTs on the same phase. Could this have caused damage?

Any pointers appreciated

Robert.Wall · 2022-02-13T02:25:27.582Z

gerombo:

could a 250+ V peak on the ACAC 9V adapter have damaged the emontx?

That is possible - how much above 250 V? If only a few volts (say to 260 V) then damage is unlikely. If you mean 415 V, then you might easily have damaged both the emonTx and the a.c. adapter.

gerombo:

I had connected all 3 CTs on the same phase. Could this have caused damage?

Definitely not. Doing that is part of the standard 3-phase calibration procedure.

gerombo:

not a single LED lights up when powering the EmonTX in AC mode nor with DC adapter in DC mode

no more output on FTDI serial console

Both these things point to either a power supply problem or the processor has failed or the sketch isn’t running.

The power is the first thing to check. When you had the 5 V USB supply connected, did you remove the link? Did you put it back when using only the a.c. adapter?
Did anything happen (LEDs light) when you connected the programmer and 5 V from the computer?

Do you have (or beg, borrow?) a multimeter that you can use to measure voltages?

Can you measure the a.c. voltage from your a.c. adapter? It should be around 12 V.
With the programmer & computer only connected, do you measure 5 V d.c between screw terminals 1 (5 V) & 3 (GND).
With the 5 V USB power supply only connected, do you measure 5 V d.c. between screw terminals 1 (5 V) & 3 (GND).

If you have 5 V d.c. in both cases, with the 5 V present and the jumper link removed, do you measure 3.3 V d.c. between screw terminals 2 (3.3 V) & 3 (GND).

With the a.c. adapter connected and the link removed, do you measure 3.3 V d.c. between the side of the jumper link furthest from the screw terminals and GND?

When you replace the jumper link, and with no 5 V present, do you still measure 3.3 V d.c. between screw terminals 2 (3.3 V) & 3 (GND).

If you don’t have 3.3 V, then the emonTx may well be damaged beyond repair, but The Shop will need to confirm that.

If you do have 3.3 V, can you load a sketch - any sketch? The standard Arduino “Blink” is the simplest, though you must change LED_BUILTIN to [DIO] 6.

If you cannot load a sketch, then either the bootloader has been damaged, or the processor itself is damaged. It’s possible to install a bootloader, but the FTDI programmer won’t do it, you need either an ICSP programmer or an Arduino.

If the processor is damaged, then the emonTx is almost certainly beyond repair.

Let us know the results so far.

gerombo · 2022-02-13T22:04:59.754Z

Thanks a million @Robert.Wall answers in line. Spoiler alert: emontx came back to life

The power is the first thing to check. When you had the 5 V USB supply connected, did you remove the link?

Yes the jumper JP2 was removed.

Did you put it back when using only the a.c. adapter?

Yes put the jumper back to test with AC adapter only. (to enable internal AC DC psu)

Did anything happen (LEDs light) when you connected the programmer and 5 V from the computer?

no LEDs, nothing

Do you have (or beg, borrow?) a multimeter that you can use to measure voltages?

I do

Can you measure the a.c. voltage from your a.c. adapter? It should be around 12 V.

Yes verified, 12V

With the programmer & computer only connected, do you measure 5 V d.c between screw terminals 1
(5 V) & 3 (GND).

Yes verified

With the 5 V USB power supply only connected, do you measure 5 V d.c. between screw terminals 1 (5 V) & 3 (GND).

Yes verified

If you have 5 V d.c. in both cases, with the 5 V present and the jumper link removed, do you measure 3.3 V d.c. between screw terminals 2 (3.3 V) & 3 (GND).

Yes verified in both cases.

With the a.c. adapter connected and the link removed, do you measure 3.3 V d.c. between the side of the jumper link furthest from the screw terminals and GND?

I measure 4.5 VDC, with ONLY the AC adapter connected (and board receiving 12V) and no JP2 link

When you replace the jumper link, and with no 5 V present, do you still measure 3.3 V d.c. between screw terminals 2 (3.3 V) & 3 (GND).

Yes verified

If you do have 3.3 V, can you load a sketch - any sketch? The standard Arduino “Blink” is the simplest, though you must change LED_BUILTIN to [DIO] 6.

Yes this worked. (both with jp2 link and internal psu, as well as on 5VDC and without link.
Facepalm. So i guess i must have messed something up with previous sketches.

Thanks a million!

Robert.Wall · 2022-02-13T22:20:17.527Z

gerombo:

I measure 4.5 VDC, with ONLY the AC adapter connected (and board receiving 12V) and no JP2 link

Mine read high in that condition too - there’s absolutely no load on the (separate) 3.3 V regulator - I presume that’s the explanation, but it is unusual. Powering the emonTx, it was close enough to 3.3 V.

gerombo:

So i guess i must have messed something up with previous sketches.

Or a spike had screwed up the EEPROM and scrambled the sketch. That’s not something I can recall being reported, but I wouldn’t discount it. I’ve done a lot of bad things to an emonTx, but never managed to do that. I can understand the mechanism whereby it could happen.