WBJr odd readings

[zoneblue]

Bob et al,

I had previously thought there to be an issue with a different sample rate than the classics 1sec averaging. This shows up when you subtract WNjr from classic out data ( to get load data).

But lately as i study the WB data more closely, it seems there is another unexplainable thing showing up. See graph below. Its real wide. These are 1sec samples out of blackbox.

The odditys are the circled areas. WB appears to be reading load spikes durint bulk, when there is no load. What coudl be causing that? The PV was eratic from clouds and mist at that time. It settles down upon float.

Edit: ive added a data sample as well.  These are mostly raw modbus values (only /10 etc), the 'stg' column is the raw value of register 4120. The circled reading is an example of one of the 'ghost' load spikes. Notably its charge state does change for that second from 1027 to 1028.  MSB 4,  LSB 3 to LSB 4. (4120 LSB: Internal state 3,4,6  3,4,6  MPPT or Regulating Voltage ). Also the array voltge goes up so something to do with a mppt sweep.

(The column "dur" is the time in milliseconds the reading took)

[boB]

Interesting data...

I see on that one line (row off data) where the WB Jr. goes UP from basically 0 amps net to +5 amps
into the battery itself, but the Classic charging current (Ibat) goes DOWN with the associated
PV input voltage going UP to keep the battery voltage from going UP (if in a constant V stage).
Then, everything goes back to the prior conditions.

I would also think that if the Classic was just doing a sweep at this one time (you might catch it doing
this once in a while), that the WB Jr. battery total current would NOT go UP but would go more negative,
or, i.e. DIS-charging some.   This is a positive WB Jr. battery current so that indicates charging and
the -0.1  -0.2 Amp net current indicates idle from the Classic itself....  Just like if it were in a Constant
Voltage stage. (Float, Absorb, EQ)


One Question is, what charging stage was the Classic in at this time ?  Is it in constant voltage
(Absorb, Float, EQ) at this time ?...  Or was it in BULK MPPT or FLOAT MPPT at that time ?  If in,
say, FLOAT, the Classic would NOT be sweeping BUT it might be right on the edge of going
from MPPT to FLOAT or back to MPPT and you may not have caught the battery voltage
dropping in part of that 1 second interval... There are a few Classic and WB Jr. samples
during one second of time.

I also see that Ipv went down at this time by an amount that makes sense from the battery
voltage to PV input voltage ratio.  (i.e.  PV current change is lower than battery current change)

What this makes me think is that there is something either adding a pulse of charging current
TO the battery, making the Classic back-off its charging to keep the battery voltage constant,
OR some load temporarily turning off...

Hard to say for certain but that's what my initial thought is by looking at the data.

boB

[zoneblue]

It occurs during bulk. Confusingly my WB figure is sign backwards. Negative is charge, positive is discharge (got that from the previous WB (raw) register, and havent got round to changing it). ON the first image (chart, you can see the bulk descent, and absorb climb back to zero. Those spikes are LOAD spikes not charge spikes.

I have no idea whether thats a false WB reading, or if thats just part of the sweep mechanism. (or even some obsure blackbox bug).

At the moment this resolution data has been accumulating for the last couple of weeks. 700K per day.  Its all like this, more or less. The following day was sunny, and its less spiky, but you can still see a rhythm of sorts to it. The spreadsheet is here:

http://zoneblue.org/files/2014-01-31.zip

[boB]

Backwards huh ?   You can reverse that by setting  bit 0x1000  (WBJrAmpsInvert)
in the  EnableFlags3  register,  address 4181  (register 4182).

OK, now I will have to look at that again now that I know it is charging when it is discharging.
Oh wait ?   You know what I mean.

boB

[Wxboy]

I'm not sure if this is related but several times I have seen strange instantaneous readings displayed.  As a matter of fact I just saw it happen.  I was watching the local app. and it was reading about 3 amps going into the batteries then it went to -.1 amps for a split second  and then back to 3 amps.  There was no change in solar output that caused this to happen.  I've seen it happen at least 3 other times but figured it probably doesn't affect much because it comes and comes and goes so quickly. 

[Wxboy]

I just saw it do again so I think it happens pretty regularly.  This time it went from 5.6 amps to 0.1 amps for about 2 seconds and then back.  Again, no change in sky conditions.  It think it held for about 2 seconds because I am watching the local app.  From what I remember when I've seen it on the Classic's display it was only displayed for a split second. 

[Vic]

Hi Wx ..,

Am no expert,  but believe that the Classic makes a very short Sweep about every minute.  These seem to last for less than a second +/--.

This may not be what you are seeing.   FWIW,  Vic