My S2105 has been working fine for several months. Now it sends an off the scale soil temp reading once or several times a day. Other data from the sensor seems normal. It sends LoRa data to a RAK7244 which sends to my weather station. I have a SenseCap CO2 sensor which is setup the same way with no problems. See attached data. Any help would be appreciated.
Thanks,
Bob
Correction to my first posting. All three data sets are reading very high blips (off scale) at the same time—soil temp, moisture and conductivity.
maybe some type of power or grounding problem?
With a high spike like that, does it indicate a short or an open circuit?
Hi there,
so does it report the correct time stamp when the spike occurs?
just curious and lurking. can the timing of the sample taken be changed? try lower or slower.
Cool stuff for sure though, and looks like a ton of data being collected here. Very nice.
GL 
PJ
what dashboard is that you using?
Upgrading firmware to 2.4 seems (so far) to have fixed the problem. Thanks all.
1 Like
I have started having this same problem again. Please see below. Very high readings have been coming more often on all three sensor readings.
Thanks.
[email protected]
Date-time string", “S2105:Soil Moisture”
2025-01-24 08:28:00, 36.2
2025-01-24 09:28:00, 36.1
2025-01-24 10:28:00, 36.1
2025-01-24 11:28:00, 2000001.0
2025-01-24 12:28:00, 36.1
2025-01-24 13:28:00, 36.1
2025-01-24 14:28:00, 2000001.0
2025-01-24 15:28:00, 36.2
2025-01-24 16:28:00, 2000001.0
2025-01-24 17:28:00, 2000001.0
2025-01-24 18:28:00, 36.3
2025-01-24 19:28:00, 36.4
2025-01-24 20:28:00, 36.4
2025-01-24 21:28:00, 36.4
2025-01-24 22:28:00, 36.5
2025-01-24 23:28:00, 36.5
2025-01-25 00:28:00, 36.5
2025-01-25 01:28:00, 36.5
2025-01-25 02:28:00, 36.5
2025-01-25 03:28:00, 36.5
2025-01-25 04:28:00, 36.4
2025-01-25 05:28:00, 36.4
2025-01-25 06:28:00, 36.5
2025-01-25 07:28:00, 36.3
“Date-time string”, “S2105:Soil Temperature”
2025-01-24 08:28:00, 32.2
2025-01-24 09:28:00, 32.2
2025-01-24 10:28:00, 32.2
2025-01-24 11:28:00, 3600033.8
2025-01-24 12:28:00, 32.2
2025-01-24 13:28:00, 32.2
2025-01-24 14:28:00, 3600033.8
2025-01-24 15:28:00, 32.2
2025-01-24 16:28:00, 3600033.8
2025-01-24 17:28:00, 3600033.8
2025-01-24 18:28:00, 32.2
2025-01-24 19:28:00, 32.2
2025-01-24 20:28:00, 32.2
2025-01-24 21:28:00, 32.2
2025-01-24 22:28:00, 32.2
2025-01-24 23:28:00, 32.2
2025-01-25 00:28:00, 32.2
2025-01-25 01:28:00, 32.2
2025-01-25 02:28:00, 32.2
2025-01-25 03:28:00, 32.2
2025-01-25 04:28:00, 32.2
2025-01-25 05:28:00, 32.2
2025-01-25 06:28:00, 32.0
2025-01-25 07:28:00, 32.2
Readings are too fast… no need to read soil temp faster than once a second… actually once a minute is probably too fast practically… I would say 5 or 10 minutes would be fine… the sensor and computer are not capable of reading that fast… need to build in some delay
The readings are every 60 minutes.
Still having this problem. Even more in cold weather.
Any ideas? Thanks
Hi there,
So I see this and think two things, Even fast or slow, Not enough precision to get anything useful in terms of trends and coralatible data IMO. You need more digits and perhaps some form of temperature compensation. or Just more Sensors. 
I would use a heater probe and do some Zero and Span clibrations then do some Multi-point calibrations and see what the curve looks like. Then try it in the in-situ environment for comparison to the base-line stuff you collect. (this is much like UHP gas analyzing for FABS)
my .02
HTH
GL PJ
Thanks, PJ, that sounds too complicated for me. I have lots of other sensors using LoRaWAN into weewx. I’m thinking it’s probably some sort of short in the actual sensor prod. Still working on it.
Cheers,
Bob