PvOutput response=Bad request 400

Aug 9, 2013 at 11:27 AM
This morning I upgraded from V8.4 to V11.2. Reason was that on daily basis at several times during the day no values were entered into PVoutput. A error 400 was returned.

Accessing the Arduino via a browser, the V11.2 has a better notification "PvOutput response=Bad request 400: Power value [18000] too high for sy".
Performing regular refresh, two values stand out: Actual and Peak. Actual now is around 2000 W, sometimes however it returns 18000, so the Peak states 18000 as well.

Where to start tackling this? The Arduino just died altogether.
V11.2
09.08.13 12:13:46
Uptime=0d+4h
ID SID Type Actual Peak Today Factor TodayCnt EEprom ppu Pulse Extra
0 14465 2 18000 18000 6661 1 13322 12191 2000 100
PvOutput response=Bad request 400: Power value [18947] too high for sy
DNS status=1
Last NTP update=09.08.13 10:00:18
WD ctr=0
WD val=2
Reset Day=255
Free=240
Aug 9, 2013 at 11:44 AM
Give te solarmeter a reset.
Coordinator
Aug 9, 2013 at 11:56 AM
Extreme high values from the S0 meter usually mean that there is a disturbance on the cable between the kWh meter and the Arduino.
The software measures the time between two pulses and if there is an extra pulse detected, it will calculate a value that can be very high.

There are a few options to eliminate disturbance on the cable:
  • Use shielded cable
  • Add a capacitor of 10 to 100nF (not higher because normal pulses will then be filtered also) between digital input and ground. The capacitor must be mounted as near to the arduino as possible
  • find an alternative route for the cable
Aug 9, 2013 at 12:09 PM
Edited Aug 9, 2013 at 12:11 PM
Thanks, will look into the cable then. Distance S0 (in the Kostal Piko) to Arduino is 30 cm though.
At the end of the day, the total production according to the Piko and the Arduino are the same.
Aug 9, 2013 at 3:40 PM
That doesn't seem to fix it. Cable is now shielded, with shield connected to ground. A 100nF cap is attached on the input on the Arduino.
PvOutput response=Bad request 400: Power value [18181] too high for system size [3185]
What is the Pulse variable doing (in the output data by the Arduino)? Seems to fluctuate on refresh between 2990 and 995..? TodayCnt does increase nicely.
Aug 9, 2013 at 3:43 PM
Aaaand.. it's gone. Dino is dead again, within minutes. I'll load another dino, see if that one lives longer.
Coordinator
Aug 9, 2013 at 3:47 PM
"Pulse" is the time in ms. between the last two pulses.

If your Arduino crashes, take a look at the power supply. If the voltage is too high (>12V) or too low (<7V) on the jackplug, strange thing can happen.
If you use a USB power supply, check that it is not overloaded (getting hot)
Aug 9, 2013 at 3:53 PM
Edited Aug 9, 2013 at 10:08 PM
A Duemilanove is now running, this replaced the Uno R3. See how the old one holds out,
Aug 10, 2013 at 12:41 PM
All the suggestions are implemented and checked. I'm of a mind to try another distribution altogether since there are still Bad Request errors from PVoutput, and it's not a very bright day today. Any other suggestions are still welcome.