VM211 Earth Listener - VMA343 Lightning sensor does not report any lightning

3.7_2 running
Boot-up seems faster (also due to no 2 seconds wait for I2C address,
but also the startup time of the AS3935 sensors is faster.

Two thoughts by me (OK, four):

  • THE most important is getting the lightning sensor under control

  • Writing to SD-card each second is way too much
    Maybe set to 1 or even 5 minutes Changing in the code would be acceptable

  • add optional RealTimeClock

  • possibility to dim the screen during night, I know that switching of the backlight leds cannot be done. But using darker colors or NO colors might be good to prevent screen burn-in. Especially the Velleman logo (that I like!) that is always on the same spot.

I know this should not be in THIS item, but my question is make a separate item for each of them?
As it is I need to do updates like this each time…

Thanks Pieter,
Bert

The download location of the zip file is in Pieter’s post above:

And if you like to try it and fiddle a bit with your EarthListener, then do so.

You already know how to update the device.
In fact simple:

  • download and unzip into you sketch directory
  • compile and upload … ready.

(Press crtl-k in the arduino GUI to get into your sketch directory)

We do not know if this firmware is better by trying at as many places as possible :wink:

Bert

@bert.test Aww…I missed the link in the post above, it didn’t display for me the first time :slight_smile:

1 Like

@bert.test @VEL342 One more question - does the 3.7 include changes in code from post 24 and 41?

Yes, interrupt reason ZERO is no longer ignored in the mainline.

@VEL342
Yesterday I’ve put in software3.7 and it works. Startup is faster.
However still no reaction to “artificial lightning” from a cigaret lighter or the spark-system from the stove.
When 3.4 was still in, I made connections from IRQ pin on the I/O board to GND or 3,3V. No reaction. (Removed the wire from the sensor first) Also, connecting pin 18 or 19 to GND has no visable effect.

1 Like

@bert.test @VEL342 3.7 is up and running. Waiting for the thunderstorm:)

1 Like

Hi Guus, you are probably also a bit of a hardware-man (I am guessing)

Usually, if a button is used on an Arduino the pin is connected to the switch and via a resistor to GND and the other side of the switch to the plus.

So, it might be that the IRQ-pin has a pull-down resistor and the interrupt pulls the pin UP?

On an Arduino UNO I experimented a bit using this schematic:

This schema along with the second sketch on Using interrupts

So, pin 2 (in this case) goes to 5V on button-press.

:relieved: (It worked better after I used the correct pin (2) as button pin instead of (1))

I did not yet dare to attempt something like this on The Earth Listener.

But it should be doable to simulate a trigger to the interrupt pin.

My progress for today,

Bert

Hi @Guus.Assmann,

something is definitely wrong. If you disconnect the I18 pin on the sensor board from the lightning module and connect it to the GND of the sensor board, the EarthListener should give an lightning detection but with the error IRQ source not recognised.
Please only do this for testing purposes.

Kind regards,
VEL342

@bert.test @VEL342

Rain is falling, clouds are rumbling, but there are no lightnings:(

There was one lightning cloud-cloud I guess but the station did not record it.

1 Like

@VEL342 and @bert.test
Today I tried again, with both software 3.4 (unmodified) and 3.7
I removed the wire, see picture in post 68 above, the white one.
And I used a pair of tweezers (Pincet) to connect the pin to GND a couple of times.
Now it does report an interrupt and error, just as it should. And it’s on the lightning screen.
This proves that the interrupt circuit is working, as is part of the software.
So if there’s an error, it’s either the sensor (very unlikely and I have 2 of them) or the software has an issue. I would expect a sensor setup not 100% correct, but have not found that yet,

Somewhere in software 3.4 I read that the capacity that’s witten on the sensor board should be entered in the correct spot. (Think it was in “setup” )
I have not done this, or checked if it’s right.

@bert.test
You guessed right. In my day job I do troubleshooting, repair and training on electronic stuff.

Thank you for giving a bit insight. I am a quality assurance engineer (Software) in heart and soul and only a moderate electronics hobbyist.

Hi there, I am new on this forum and suffering from the same issue as you described. I think I solved the issue. The AS3935 is a very sensitive 500KHz radio which is also very sensitive to noise. The way it is designed now will never work. The antenna is simply too close to the TFT screen and the Arduino micro processor. They both are very noisy. Moreover one should use a ‘silent’ power supply and not a switch mode power supply, which is too noisy. Last what you should do is to extend the wires of the AS3935 to about 40-50 cm so that the sensor is no longer close to the noise sources, and then it will work.

@bert.test @VEL342 @beva

Update:

  1. There were two thuhnderstorms yesterday, both with cloud to ground lightnings and the station did not record any disturbances. As a proof I can say that one of the crop fields got fire from the lightning strike.
  2. My sensor did not show “too much noise” or “too move the senson” message not even once.
  3. The status is still “none” as on the pictures from my posts above.
  4. Maybe beva is right?

Thanks for testing. If you have a wire available, then please connect the AS3935 IRQ pin to earth for a short moment. This causes a high to low on PIN19 and I predict you will then get the message: too much noise. If that is the case, it is worthwhile to place the sensor far away from the TFT screen and Arduino. But also be sure that you feed the Earth Listener with a silent power supply (e.g. a battery pack). When I did that, the IRQ pin became low by itself and I could see the system work by applying sparks via a lighter normally used to fire the gas stove. I hope this explanation will help you to solve you this problem.

@beva IRQ to earth?:wink: How to connect Earth Listener to battery pack?

1 Like

Hi @beva are you sure you mean PIN 19 or should it be PIN 18?

@bert.test: sorry that I add to the confusion about the PIN numbering. It is indeed PIN 18.

@Pz83: on post 59, from VEL342, there is a clear picture of the AS3935 sensor. There you can see 5 connected PINS: VCC, GND, SDA/MOSI, SCL and IRQ. Take a small wire a shortly connect GND and IRQ. Then I predict the interrupt will be activated and the screen will show: too much noise.

The Earth Listener can be fed via the USB port or by the battery port (round connector with a small PIN in the middle, positioned below the USB port). Use that last connector to feed the Earth Listener with 9V DC. This could be a battery pack or a simple power supply (transfomer with a rectifier). But it should not be a switch mode power supply as used for loading a smart phone).

1 Like

@beva When I did this I only get the message as in VEL342 post #69 “IRQ source result not expected” and nothing more. I had disconnected the I18 on the sensor shield and connected it to GND. No “too much noise” message.

I do not know what else to do.

@bert.test @VEL342 help please :slight_smile: