SparkFun Forums 

Where electronics enthusiasts find answers.

Have questions about a SparkFun product or board? This is the place to be.
By ivonci
#179603
Hello

I am working on a weather station project, and my components are Arduino Mega + Ethernet Shield + RTC3231 + Weather Shield. To the Weather Shield, I have connected Weather Meters, and used code from this tutorial:

https://learn.sparkfun.com/tutorials/we ... nderground

I am loging total rainfall and maximum wind speed to SD card. This all works well. The problem is, that I get a random count for rain and wind speed, even if there is no rainfall or no rotation of anemometer.

What could be causing this random count of interrupt input? At this time I have the board connected via the USB wall addapter, could it be that there is to much power consumption and the voltage drops now and then?

I will try with 9VDC power supply, when it arrives.

Has anyone else got this problem?
User avatar
By Ross Robotics
#179608
They may just need calibrating..
By uChip
#179610
I have the USB Weather Board (retired :( ), not the weather shield, but I believe the circuit for the wind and rain sensors is the same. I have never seen a spurious rain or wind reading.

How long are the lines from the shield to the sensors? Have you extended those lines from the wiring included?

How do you know that the rain and wind sensors did not move? Are they inside for testing?

It might well be that too low a power can cause a glitch, but I run mine from a small lipo battery and have not had problems. I also program/test it running from USB powered by my computer, also no problems.

Good luck with it.
- Chip
By ivonci
#179621
Hello

Lenght of the lines is standard. Yes, they are inside, not moving. I have made the debounce of the rain sensor 100 ms long, but it stil has random readings... I'm thinking the problem is with power, because interrupt reads falling edge. I will have to try it with 9V 2A wall adapter, maybe 500 mA from wall-usb isn't enough, when let's say ethernet shield is serving web pages...
Once i had similar problem with UNO and wifi shield, when it didn't connect because of too much drawn current.
By uChip
#179644
Do each of the shields have their own regulators or are you relying on the regulator on the Arduino? If you count on the Arduino that's probably the problem. There's only so much "extra" current the Arduino regulator can provide.

If each of the shields takes raw power from the wall adapter and regulates it on the shield then that's probably ok. If they all depend on the Arduino regulator, I doubt that's going to work.

- Chip
By ivonci
#179645
I think that the ethernet shield has Power Over Ethernet, and has its own 9 V regulator on board. But it still probably draws some current from Arduino.

I will try to remove ethernet shield and just put Weather Shield on and will try to run it for several hours, looking at the data on the serial port. If I don't get random counts, then it has something to do with the ethernet shield. I will try this and report back.
By ivonci
#180576
Hello. I have tried powering the project from 12V 1500 mA wall adapter and that seems to fix the problem of random counts. I think that the problem is probably to much current drawn from ethernet shield, so that the power supply becomes unstable. I have also seen this if powered the board with the computer usb port only. I have now seen that the ethernet shield gets the power from the arduino 5V supply, but i can't find how much.

The voltage regulator on arduino board is getting pretty hot, is there any way around this, or is it best to just ad a smal 12vdc fan beneath the board to cool down the regulator? I will try to measure the tempretaure of the regulator with a thermometer to see how hot it gets.
By ivonci
#180611
Hello, just a little update.

I have resolved the problem with powering the board from a 5V wall usb adapter (samsung 2A). I have also added 1uF capacitor between pin 2 and ground and pin 3 and ground. The two capacitors debouce the reed switches and now i don't have anymore random counts.

When the board was powered from 12V wall adapter, the onboard regulator was overheating and arduino began crashing after about ten minutes (overheating protection of 7805 votage regulator).