smurphy
Full Member
Posts: 169
Raspberry Pi: Yes
Other Device: many ...
|
Post by smurphy on Oct 19, 2020 0:03:19 GMT -8
So, since the lightning strike took out some of my equipment, I have replaced the Dust Sensor (AQI) with a new one. This one however only shows me 0 particles over time (as with the other one I thought to be dead.
1. Is there something else than running the testDust.py to troubleshoot it?
2. Is there something else that could have broken, like the "grove Powersave"? How could I troubleshoot that one? Could it theoretically be possible that the signal won't get through? Thing is that apparently the AQI is found but the lightning sensor is not (Ok, third party module). But it could be that the old module is still good, and the Powersave unit is the culprit).
For information, the 2 boards that were connected to a "grove Powersave" unit stopped working. So it could be that both are broken.
Thx for any hint.
|
|
|
Post by SDL on Oct 21, 2020 13:06:24 GMT -8
1) testDust.py is the best.
2) Bypass the Grove Powersave and see if it is working.
The Grove Powersave basically has a solid state relay on the VDD line so it is entirely possible that it was fried too.
BP
|
|
smurphy
Full Member
Posts: 169
Raspberry Pi: Yes
Other Device: many ...
|
Post by smurphy on Oct 24, 2020 3:18:39 GMT -8
Well, both the dust-sensor and the thunderboard stopped responding at the same time. Yet nothing was visible on the boards themselves. The powersave modules - you have the solide-stat relay part details? eventually a little howto how to test these. Maybe I'll have to replace these.
Thx
|
|
|
Post by SDL on Oct 26, 2020 11:03:33 GMT -8
Petty simple to test. Put a Grovecable to a I2C connector in the Input end.
Put another Grove cable in the Control Port from a digital output grove port
Connect up a digital voltmeter from VDD to Ground to monitor the output Grove Connect.
Write a program to switch the GPIO connected to the Grove port from 0 to 1 and then back to 1 checking the voltage on the VDD on the output connector.
It should turn on and off the voltage.
BP
|
|
smurphy
Full Member
Posts: 169
Raspberry Pi: Yes
Other Device: many ...
|
Post by smurphy on Oct 28, 2020 7:17:16 GMT -8
I'll check that. Even have a small oscilloscope I can connect to it to view it. Cheers
|
|
smurphy
Full Member
Posts: 169
Raspberry Pi: Yes
Other Device: many ...
|
Post by smurphy on Oct 31, 2020 5:07:51 GMT -8
ok. Everything looks Ok. I have also bypassed the Grove board all-together, and still no data from the AQI module (the new replacement module). So I suppose the error must be at another location then. The only module attached to the PI2Grove board is the AQI. This one does not work. The Thunder Detector does not work either (New one plugged in).
I think I will have to dismantle the Weather station completely, and test each and every component individually (Here, the PI Weather and the PI2Grover board). Is there any instruction on how to check these work correctly? Eventually I'll just order a replacement.
|
|
|
Post by SDL on Oct 31, 2020 7:16:35 GMT -8
Smurphy,
Is it possible that the Pi2Grover boards is blown?
Lightning can destroy everything (especially the lightning detector of course). Are you using a different Raspberry Pi? Are the i2C lights showing up on the Pi2Grover board (do a "sudo i2cdetect -y 1"
BP
|
|
smurphy
Full Member
Posts: 169
Raspberry Pi: Yes
Other Device: many ...
|
Post by smurphy on Nov 1, 2020 10:50:06 GMT -8
I have replaced the raspberry PI. The old one was starting to use 800ma (Raspberry PI Zero WH) and CPU temp going up to 80⁰C before I had to shut it down. So - yes, the PI2Grover board can be defective. Everything is detected as per instructions. But, that does not mean that some parts are not broken.
pi@weatherpi:~ $ sudo i2cdetect -y 1 0 1 2 3 4 5 6 7 8 9 a b c d e f 00: -- -- -- -- -- -- -- -- -- -- -- -- -- 10: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 30: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 40: -- -- -- -- -- -- -- -- 48 -- -- -- -- -- -- -- 50: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 60: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 70: -- -- -- 73 -- -- -- 77 pi@weatherpi:~ $ date Sun Nov 1 19:50:15 CET 2020
|
|
|
Post by SDL on Nov 2, 2020 9:20:58 GMT -8
|
|
smurphy
Full Member
Posts: 169
Raspberry Pi: Yes
Other Device: many ...
|
Post by smurphy on Nov 2, 2020 10:35:26 GMT -8
Using the old AQI. In fact, I just sent in an order for some parts. Weaherpi Board, PI2Grover board, 2 powersave units and a thunderboard. With that I should be able to come back to the old setup before the lighning destroyed the parts...
|
|
|
Post by SDL on Nov 4, 2020 8:22:02 GMT -8
Hmm. I think the old one is destroyed then by the lightning. I don't have another suggestion.
John
|
|
smurphy
Full Member
Posts: 169
Raspberry Pi: Yes
Other Device: many ...
|
Post by smurphy on Nov 21, 2020 6:52:29 GMT -8
So - after replacing the skyweather and the PI2Grover board, data is coming back in. So the lightning strike has left quite some damage on either (or both) boards. Kudos to the SDL Team for sending me the replacement parts so fast
|
|
|
Post by SDL on Nov 21, 2020 8:21:13 GMT -8
Sir,
You are welcome. As near as we can tell you were the first to really use SkyWeather as a lightning detector! The hard way.
BP
|
|
smurphy
Full Member
Posts: 169
Raspberry Pi: Yes
Other Device: many ...
|
Post by smurphy on Nov 22, 2020 4:45:00 GMT -8
*lol* Yes. Well, I can confirm that a lightning strike in the vicinity (around 20m away) did break quite some things (not only on the SkyWeather box, but neighbors had to replace TV Sets and other electronic equipment).
And I just realized, I probably have a spider inside the AQI - which reports only weird stuff (500pp/m³ or 0..). Seems it is obstructing the air intake. I'll have to figure something out.
Note - for those wondering in how to keep out small animals from your skyweather system, make sure you use hood foam. Best stuff to keep these little beasts out. Only, for the AQI, we need unhindered airflow - reason I didn't put any here. Will the new sensors have anything against it? or be handling these better?
|
|
|
Post by SDL on Nov 22, 2020 6:55:33 GMT -8
Basically the new sensor has the same problem. Sounds like we need to look at some solutions. Fine screen perhaps? Not too find.
BP
|
|