elp
New Member
Posts: 15
|
Post by elp on Aug 28, 2021 12:12:26 GMT -8
I have been using WeatherRack2 for about 6 weeks. I have been collecting data and storing each sample in an InfluxDB database. I'm interfacing to the WeatherRack from an ESP32 using the SDL_ESP32_WeatherRack2 library.
I am seeing an anomaly with the cumulative rain measurement that is received. Over the past 15 samples received in the last 26 minutes, the reading "jumps around". Specifically, it has moved from 1371 to 0 to 1371 to 33 to 1371 to 33 to 1371. This is after I did a reset on the WeatherRack2 to see if that could be causing the problem. (I have verified this "jumping around" from Serial.print statements in my Arduino code.)
It is a clear sunny day, so there is no rain.
What could be causing this? Is my unit defective?
Thanks for any insight that you can give me.
|
|
|
|
Post by SDL on Aug 29, 2021 10:10:16 GMT -8
elp,
This is interesting. We are not seeing anything like this from anyone on the Raspberry Pi receiving data from the WeatherRack2. That makes me think that this is a bug in the ESP32 WeatherRack library. Do you have a Raspberry Pi and an SDR to run SDL_Pi_WeatherSense and see if you are picking up this on the Pi?
It is very unlikely this is a defective unit due to it keeping the value when it comes back. Sounds like a transmission error to me or a reception error on the ESP32.
Do you have a bunch of other things running on the ESP32 which might interfere with the reception?
BP
|
|
elp
New Member
Posts: 15
|
Post by elp on Aug 30, 2021 9:01:02 GMT -8
Thank you for your response.
I am receiving data only from the WeatherRack2 and from the Indoor T/H sensor. I don't think that anything else in my application could be interfering with the reception. As I indicated, I have the ESP32 connected to my Wifi network so that I can store the data in InfluxDB. As such, I have the Wifi library, the InfluxDB client library, and the ArduinoJson library installed so that I can deserialize the incoming messages to prepare the data to store in InfluxDB. I only store data after receiving a message from the SDL_ESP32_WeatherRack2 library.
So, I don't believe that I have anything else interfering. However, what I will try is reverting to your example code from github without any of my additions to see whether I can duplicate what I'm seeing.
I'll be in touch. Thanks again.
|
|
|
|
Post by SDL on Aug 30, 2021 12:10:18 GMT -8
I'm really talking about processes running on the ESP32 interfering with the tight timing required by the demodulation system, not something external. Oh, and there are 433MHz signals in your area that probably do interfere on an occasional basis, but I really think it is a multi process problem.
If you have an Arduino, check out the software for that. Going back to the zero base software and finding out if it happens there is a great idea.
BP
|
|
elp
New Member
Posts: 15
|
Post by elp on Aug 30, 2021 12:16:46 GMT -8
I just got done running a test of the SDL_ESP32_WeatherRack2 app from the SwitchDocLabs github page. I have added an attachment with the terminal output from Platformio showing an example of the cumulativerain measurement oscillating from 36 to 1374 several times over the half-hour or so. This is the same thing that I was observing in my application that I described above. I have highlighted this oscillation i yellow on the attachment. Again, I am using the ESP32 software from your github page unaltered. I hope that this helps you find why this is occurring. Let me know if you need anything else. I really would like to use this to track rainfall over time by storing the data in InfluxDB. Thanks again for your help. Cumm-Rain-Error-Example-2021-08-30.rtf (10.14 KB) |
|
elp
New Member
Posts: 15
|
Post by elp on Aug 31, 2021 14:36:55 GMT -8
To provide you with additional information, I thought I would run the same test using my Arduino Uno, rather than my Adafruit ESP32. As you can see from the attachment, I am seeing the same phenomenon of the cumulative rain reading oscillating between 1557 and 243. (In the attachment, I have highlighted the values to make them easing to see.) This test was run using the Arduino Uno application & library from your github. Again, I'd appreciate any insight or fixes that you could provide. Thanks in advance for your help. Cumm-Rain-Error-Example-2021-08-31.rtf (25.72 KB) |
|
|
|
Post by SDL on Sept 2, 2021 15:52:27 GMT -8
Elp,
Thank you for the excellent information. I'll have to set this up in the lab duplicate it and the FIX it. we are not seeing this on our Python interfaces, so the problem has to be in the drivers for the Arduino /ESP32.
I get back in the office tuesday afternoon. Feel free to nag.
BP
|
|
elp
New Member
Posts: 15
|
Post by elp on Sept 3, 2021 6:19:02 GMT -8
Thanks so much for your help. I'll keep watching for when you have more information.
|
|
elp
New Member
Posts: 15
|
Post by elp on Sept 3, 2021 7:54:23 GMT -8
Not to be nit-picky, but if you're going to "have the patient open", could you check one more thing. I noticed on my ESP32 test, that the Battery Health indicator on the Indoor T/H sensor indicates "LOW" for my unit. I'm quite sure that the battery in my unit is good.
So, is there some polarity problem in that reading?
Again, thanks so much for your attention. I hope you have a nice Labor Day weekend!
|
|
n7qnm
Junior Member
Posts: 79 
|
Post by n7qnm on Sept 5, 2021 12:32:27 GMT -8
Hi, elp, et al - I see the same thing on my WeatherRack AND on an Accurite 5-n-1. In my case, it's wind related. The problem is that without a good support on the mounting "mast", if the sensor unit "wobbles" even a little but, it can tip the tipping bucket.
I'm doing a new installation, and right now, my mount is a 10' 2x4, with the bottom 4' tied to a 4x4 fencepost. Yesterday, we had 20+MPH winds and I picked up 2mm of readings. I'm trying to figure out a better permanent mount, I really don't want to guy it unless I have to; but, how to get 6' "above the weeds" height and an ABSOLUTELY stable mount is difficult.
I'm also adding a "RainOffset" configuration variable and then modified wirelessSensors.py to pick up the offset and subtract that from the reading.
|
|
elp
New Member
Posts: 15
|
Post by elp on Oct 2, 2021 8:00:42 GMT -8
You said feel free to nag. Any progress on addressing the cumulative rain issue? Or, do you have an ETA?
|
|
|
|
Post by SDL on Oct 3, 2021 13:14:40 GMT -8
Haven't even started the project. We are hammered in development right now. I'll put it on the list. I should be able to get to this by the end of the week.
BP
|
|
elp
New Member
Posts: 15
|
Post by elp on Oct 18, 2021 13:40:34 GMT -8
Thanks. I look forward to your response.
|
|
elp
New Member
Posts: 15
|
Post by elp on Oct 28, 2021 8:36:01 GMT -8
Looking for an update on your fix to the SDL_ESP32_WeatherRack2 library. We have been getting rain where I live for the past several days (after a drought for most of the summer) and the cumulative rain measurement is useless. I am very disappointed and losing patience.
Any update on your schedule for a fix?
|
|
|
|
Post by SDL on Oct 29, 2021 10:30:04 GMT -8
We have it scheduled for looking at it in early November. Please don't be disappointed and impatient. We are a small team and have a lot on our plate!
BP
|
|
