Archived Forum

Thank you !

I'll test it soon ;)

Hello,

No result.

It transmit at start the battery power and nothing after. The total blackout...

I puted a pulldown between 3 and 4, tried the other wire to 2, 5 or 6 : no response.

I am beginning to think that my emonth is broken...

What is your emonTH, V1.4 or V1.5? 

The pulse input is D2 on the V1.4, D3 on the V1.5 - so you might need to change "const int PULSEPIN=2;"

[ERIC: You need to add that in a comment.]

Hello,

Sorry, i precised it on another thread. It is V1.4.

Connecting my reed is not clear. Where connect it ?

I use a pull-down resistor between pin 4 and ground to fix 0v when no working on reed.

I tried to connect other wire to 3v to send 3v to pin 4 when sollicited.(pin 1).

"Sorry, i precised it on another thread. It is V1.4."

That is what happens when you cross-post. Please do not do that, it wastes everyone's time. You cannot expect anyone to remember a post from 2+ weeks ago on another thread.

"Connecting my reed is not clear. Where connect it ?"

Look at the circuit diagram - you will find it in the Wiki. You will see Pin 4, the pulse input, has provision for a pull-down resistor to GND. Therefore you connect your reed switch between Pin 4 and 3.3 V, which is Pin 2. If no resistor (R2) is fitted on yours (it is underneath the battery holder), use an external 10 kΩ resistor between Pin 4 and Pin 3.

It wasn't a cross-post but an other sketch from Eric :) So i missed to precise it on this one.

Otherwise, thank you for your answer. It confirm my situation : My Emonth doesn't work. For the pull down i didn't put it behind battery pack yet because i wanted to test it before.

You confirm my 3.3V  use for it. So, Dig2 is protected and can accept 3v load. I wasn't sure and i supposed that i fused it.

I will make other tests, before to decide to trash it... and buy an other one. I think this one is broken. No transmission after first start and first battery value sent.

Thank you.

Changed JeeLib libraries because i had old one (2014) No way.

Tried to resold weldings : no way. (electronic lab fully equiped)

If you have a programmer, you can try a simple sketch:

#include <JeeLib.h>

void setup()
{
  pinMode(9, OUTPUT); digitalWrite(9,HIGH);   
  delay(10000);
  digitalWrite(9,LOW); 
}

void loop() 
{
  digitalWrite(9, digitalRead(2));
}

With your reed switch and pull-down resistor connected, the LED should come ON for 10 s, after that it should come on when the magnet is near the reed switch, and it should go off when the magnet is away from the reed switch.

Thank you, i test it now.

EDIT :

Working well !

So my connection is good.

Will try to understand sketch programmation... :)

EDIT2 :

Well...

Nothing. No reaction. No back to console too.

So, the problem is either the sketch is not working as it should, or your reed switch is "bouncing" so much that the sketch cannot recognise the transition from 0 to 1.

The main loop runs and then sleeps for 500 ms. Every time, that is every 500 ms, it reads the switch. The last 8 switch states are recorded in the bits in a byte, and when you have 00001111 (0x0F), you have detected the switch closing.  If you have a really bad switch, it might be bouncing for a long time. But it should not bounce for more than 500 ms, and the switch must remain in each state long enough to give that bit pattern. i.e. it must be in each state for at least 2 s.

Thank you for your precisions.

Are my fingers and a peace of wire can generate boucing phenomen ?

Ill will try to modify some parameters to taste more things?

Yes, a piece of wire or a very cheap switch are both very good ways to get contact bounce. Using the 'interrupt-driven' pulse sketch, I could see anything up to about 40 pulses with pressing the button once only.

Thank you for the precisions.

Hi,

Attached is a new version of the sketch described above but for EmonTH V1.5.

It also sends the temperature/humidity every minute. It's derived from the emonTH_DHT22_DS18B20_RFM69CW sketch (V1.6.1).

NB : When reading temperature and humidity with the DHT22, one pulse may be ignored because this sensor is very slow (2s).The DS18B20 sensor which is faster (375ms) is less problematic

Eric

Thanks, example has been pushed to github

https://github.com/openenergymonitor/emonTH/commit/4ae519baecd59627ffdbc...