n7qnm
Junior Member
Posts: 80 
|
Post by n7qnm on Jun 1, 2022 15:59:15 GMT -8
I'm trying to set up the WS-AQI and having a bunch of weird things happen.
I tried to place it next to my Weather Station (SkyWeather2); but apparently the range on the little transmitter isn't good enough (it's about 150' feet away from the receiver, behind and exterior wall). I've tried different "gain" antennas on the receiver, will be posting my experiences about those in a different thread.
So, I moved it to a much closer location and now I'm getting reliable communication, but the solar voltage and current are "confusing" to say the least. In full sun, I get almost 6V out of the two panels (fed through the combiner board); but I only see 2-3 MA of solar current, with OCCASIONAL spikes to a few hundred MA.
With the low power mod, the battery current is between 80 and 40 MA and the current is always between 20 and 25 MA; and the battery voltage is constantly above 4.
Is there something going on with the accuracy of the (power) sensors because of the low currents; or should I bring the thing inside and start looking for "intermittent" connections?
Thanks in advance!
Clay/N7QNM
|
|
|
|
Post by SDL on Jun 3, 2022 7:45:35 GMT -8
Clay,
Post some graphs if you would. Also, you will get garbage data out of the solar board if you have it plugged into a USB charger when you are running. That just confuses the poor INA3221 chip because of the diodes in the system.
BP
|
|
n7qnm
Junior Member
Posts: 80
|
Post by n7qnm on Jun 8, 2022 19:46:27 GMT -8
The saga continues - after I discovered the bad antenna on the transmitter board, I VERY CAREFULLY bent the pins on the board so it would angle below the USB cable, and then bent the antenna UP so it would clear the bottom of the case.
I ran rtl_433 with "-vv" and saw this error
switchdoclabs_weathersenseAQI_ask_extract: Error on 6to4 decoding low nibble: 35
(with a different nibble value) 8 times out of 72 packets, and the sensor "Buttoned up" about 10' from the reciever.
So, I took the whole unit outside and mounted it on the wall - same distance but through a double-pane glass window. Got NO data after 1.5 hours. I thought perhaps I had forgotten to turn on the main power switch, so I opened up the unit while it was on the outside mount, cycled the power switch and was rewarded with ONE packet in 2 hours.
I brought the unit back inside, and started receiving packets approximately once every 20 minutes. The ONLY difference is that when it's inside, the case sits with the back on my workbench (horizontal). Outside, the case is mounted a corner of the house, so the back of the case is vertical.
I'm wondering if the contacts or header for the transmitter board are intermittent. The header is actually a bit loose and the weight of the transmitter board bends the header a few degrees from perpendicular to the main board when it's mounted vertically. Tomorrow, I'm gonna pull things out of the case and examine that, and perhaps reheat the solder joints for the header.
What sort of range should I realistically expect, compared to the WeatherRack? I have the WeatherRack on a pole about 120' feet from the receiver, I did have to go to a dipole antenna om the receiver; but now WeatherRack is solid. I had HOPED to mount the AQI on the same pole.
Also, with rtl running, I'm seeing a BUNCH of errors from that. On startup, I see a flurry of these:
SDL_SM23: short package. Header index: 1248
bitbuffer:: Number of rows: 1
Then those go away, and I see messages like these repeated over and over between "good" JSON
{"time" : "2022-06-08 16:20:36", "model" : "SwitchDoc Labs FT020T AIO", "device" : 12, "id" : 78, "batterylow" : 0, "avewindspeed" : 17, "gustwindspeed" : 26, "winddirection" : 286, "cumulativerain" : 39875, "temperature" : 1215, "humidity" : 33, "light" : 13892, "uv" : 43, "mic" : "CRC", "mod" : "ASK", "freq" : 433.909, "rssi" : -0.606, "snr" : 4.637, "noise" : -5.243}
pulse_demod_manchester_zerobit(): SwitchDoc Labs Weather FT020T Sensors
bitbuffer:: Number of rows: 1
[00] {127} 00 29 89 c8 22 34 3d 37 87 09 7e 42 4c 88 57 82
Checksum error in SwitchDoc Labs F016TH message. Expected: e9 Calculated: 01
Message: {48} 4c 4e 41 11 a1 e9 : 01001100 01001110 01000001 00010001 10100001 11101001
Checksum error in SwitchDoc Labs Weather message. Expected: a8 Calculated: 4e
Message: {48} 46 cf 90 c2 0f d5 : 01000110 11001111 10010000 11000010 00001111 11010101
Checksum error in SwitchDoc Labs Weather message. Expected: ae Calculated: 7f
Message: {48} 46 cf 90 c2 0f d5 : 01000110 11001111 10010000 11000010 00001111 11010101
Checksum error in SwitchDoc Labs Weather message. Expected: ad Calculated: 6c
Message: {48} 46 cf 90 c2 0f f5 : 01000110 11001111 10010000 11000010 00001111 11110101
Checksum error in SwitchDoc Labs Weather message. Expected: a8 Calculated: 22
Message: {48} 45 58 74 bd 20 f5 : 01000101 01011000 01110100 10111101 00100000 11110101
Checksum error in SwitchDoc Labs Weather message. Expected: 00 Calculated: aa
Message: {48} 45 58 74 bd 20 f5 : 01000101 01011000 01110100 10111101 00100000 11110101
Checksum error in SwitchDoc Labs F016TH message. Expected: f5 Calculated: d5
Message: {48} 46 cf 90 c2 0f f5 : 01000110 11001111 10010000 11000010 00001111 11110101
{"time" : "2022-06-08 16:20:44", "model" : "SwitchDoc Labs F016TH Thermo-Hygrometer", "device" : 88, "modelnumber" : 5, "channel" : 8, "battery" : "OK", "temperature_F" : 81.300, "humidity" : 32, "mic" : "CRC", "mod" : "ASK", "freq" : 433.904, "rssi" : -0.117, "snr" : 5.270, "noise" : -5.387}
Any suggestions would be appreciated!
|
|
|
|
Post by SDL on Jun 9, 2022 16:39:49 GMT -8
Looks like lots of noise near the receiver (since it is messing up so many message). This would hose up everything.
Should be about ~100m in clear space. Put an external antenna on the unit if you can. Bigger antennas solve lots of problems.
If you bent the pins as shown in the manual, you should have a good solid contact between the transmitter board and the computer.
Do you have any other of our sensors that use that same spring antenna? If you do, swap and see what happens.
BP
|
|
n7qnm
Junior Member
Posts: 80
|
Post by n7qnm on Jun 9, 2022 20:20:21 GMT -8
OK - noise near the receiver makes sense, I'll try to relocate; but that may be a bit of a challenge.
Did some testing this afternoon with the AQI and am getting reliable reception at about 10M through a couple of exterior (insulated) walls.
I have a dipole on the receiver now and that seems to work for the WeatherRack and FT016s out to about 75M. I have at least 9 433 MhZ devices and the receiver is seeing stuff from neighbors, so I suspect part of the problem is just congestion.
I bought a whip on Amazon that was supposed to have 10 DbI gain - a dipole is 2.51; but the dipole hears the WeatherRack and the whip doesn't - go figure.
One way or another I'll get it working.
Clay
|
|
|
|
Post by SDL on Jun 12, 2022 16:08:08 GMT -8
Clay,
Odd about the whip. Does the whip receive anything?
BP
|
|
n7qnm
Junior Member
Posts: 80
|
Post by n7qnm on Jun 18, 2022 10:37:27 GMT -8
Hi - doing a bit of fiddling this morning - I have a second 433 SDR for my WeeWx sensors, right now that's on the whip. I noticed it was receiving a "Cotech" sensor, which I'm pretty sure is the WeatherRack2 (can you confirm that?).
Anyway - it was a bit intermittent - so, I moved the whip from horizontally oriented with a metal desk leg as a "ground plane" to vertically oriented mounted on top of our alarm box - now it's seeing that sensor just fine.
|
|
|
|
Post by SDL on Jun 19, 2022 10:04:40 GMT -8
No, the cotech driver has problems and is not a direct replacement for the SDL driver. I **think** it has the same demodulation system (it looks like they copied ours to tell the truth - which is fine - but they introduced problems). The SDL driver works.
BP
|
|
