Lite, WBJR, 1609 observations

[zoneblue]

This morning i fitted the WBjr, installed 1609 on the 150lite, and upgraded local app to 0.3.31. Did the jumper VMM by following the instructions on the downlaod page, it works fine.

For the record, a VMM in this siutation, did the following:

- reset battery voltage to 12v, voltage setpoints to default etc
- left my static address alone
- set aux2 to clipper control auto
- turned on lowmax, what ever it is
- reset most other things like temp compensate, current limit etc to defaults

Now local app status area now says

Classic 150V (rev 4)
Solar(On)
System Amps: 0

(Not sure what system amps used to say). But 0 is probably wrong (system is running at 500W solar, battery amps readout under guage says 11A.  Of course now i have no idea what battery amps means anymore.

WBJr led isnt flashing.  Instructions say: "simply go to AUX2 setup and pick the last function "WBJR" and press enter. Thats for MNGP, whereas local app has no entry for WBJR, last entry is float low.

So local app 0.3.31, even though the readme said it added support for WBJR didnt add config for it?

Can it be enabled using  modbus?

[zoneblue]

The May modbus document has a typo in address list for 4165 (needs some page numbers?).

4165  R/W  Aux 1 and 2 Function  [4165]  Combined Aux 1&2 Functions + On/Off.  See  Table 4208

The table itself is titled Table 4165 , in 4 parts.

Boy that register is kinda tricky.

Also table 4165-4  AUX 2 Function, has no value for WBJR. So looks like im stuck. Bob what value should i use for 4165?

[boB]

4165 (Aux12 Function)  should have  0x12 in the high byte.  So if it was  say,  0x0103  make it  0x1203.

That will enable the WB Jr. for aux 2.  But you are right, there is nothing released as of this weekend
for the Local App and WB Jr.  I will see if I can get you something maybe this weekend even.
At least something to show amps.

boB

PS...  I guess that's why it's called the Lite ?

[zoneblue]

Thanks bob, it works.  If nothing on aux1, set 4165 to decimal 4609.

Code Select Expand

./newmodbus-1.0.19-ARM 192.168.0.223 4165=4609


No hurry for the local app, its blackbox data for WB im wanting.

[boB]

Thanks bob, it works.  If nothing on aux1, set 4165 to decimal 4609.

Code Select Expand

./newmodbus-1.0.19-ARM 192.168.0.223 4165=4609


No hurry for the local app, its blackbox data for WB im wanting.

Ahhh...  Blackbox should work great !  That should also make for some pretty
good logging too.

boB

[zoneblue]

As for the value, register 4602... if my reading of the docs is right should get us:

Code Select Expand


$defns['ibat']= array(
     'name'=>       "Whizbang Current",
     'type'=>       'sampled',
     'store'=>      true,
     'interval'=>   'periodic',
     'method'=>     'get_register',
     'argument'=>   'BITS([4362],15) ? -(65536-[4362])/10 : [4362]/10',
     'comment'=>    '(decimal) battery current 1dp, positive=discharge, negative=charge ',
     'unit'=>       'A',
     'priority'=>   2,
);


Which produces a sane enough reading. Cool.

What is the difference between these two?:
(4362)         WbJrrRawCurrent      Whizbang Jr. current signed.  +-3,276.7 amps
(4371)         WzBangJrCurrent32Signed     WbJr.  signed  16 bits current   +-3,276.7 amps

Also if we are to compare apples to apples, we need to monitor battery current in the same timeframe as the controller's output is measured. You know how the controller output has these registers:

4117  R  IbattDisplayS  ([4117] /10) Amps  Average Battery Current (1 sec) 
4115  R  dispavgVbatt  ([4115] /10) Volts  Average Battery Voltage (1 sec)
4276  R  Output Vbatt   ([4376] /10) Volts   (peak V)  Battery Voltage Unfiltered

I am using 4115 for output voltage, and 4117 for output current, both of which you have previously said are pretty accurate. So if these are 1 sec average readings, is 4362 also 1 sec average?

Then  we can do the required subtraction to get load amps.  Where does the classic tare loss fit in this equation, do i need to subtract that as well, or is that part of Iout (4117)?

To avoid confusion, looks like you are going to have to rename some of those registers from {bat} to {output}, eg 4117   IbattDisplay is not the battery current, but the controller outout current.

I have to say being able to finally distinguish these apart is just fantastic.!!

[zoneblue]

Regarding the classic modbus ethernet timeout issue, its too soon to say, but am keeping an eye on it, and will crank up the sample rate to test.

While a straight addition gets you Iload. I havent got to the other derived datapoints yet, ill do those later. Labour day hol here, gota go be nice to the SO.

[boB]

What is the difference between these two?:
(4362)         WbJrrRawCurrent      Whizbang Jr. current signed.  +-3,276.7 amps
(4371)         WzBangJrCurrent32Signed     WbJr.  signed  16 bits current   +-3,276.7 amps

I messed up on my description of  4362 WbJrrRawCurrent...  That is NOT calibrated in amps.
What WbJrrRawCurrent  is, is the raw output from the WB Jr.  In the usual normal case,
each count means  0.078125  amps.   This number is signed though.

4371 WzBangJrCurrent32Signed   is the corrected and rounded output that you want.

I need to change modify my original posting but I cannot seem to find it now.  I would like
to make that correct as well as add the Aux function setting for WB Jr. selection and
also the flag that allows you to reverse the polarity.

Also if we are to compare apples to apples, we need to monitor battery current in the same timeframe as the controller's output is measured. You know how the controller output has these registers:

4117  R  IbattDisplayS  ([4117] /10) Amps  Average Battery Current (1 sec) 
4115  R  dispavgVbatt  ([4115] /10) Volts  Average Battery Voltage (1 sec)
4276  R  Output Vbatt   ([4376] /10) Volts   (peak V)  Battery Voltage Unfiltered

I am using 4115 for output voltage, and 4117 for output current, both of which you have previously said are pretty accurate. So if these are 1 sec average readings, is 4362 also 1 sec average?

Then  we can do the required subtraction to get load amps.  Where does the classic tare loss fit in this equation, do i need to subtract that as well, or is that part of Iout (4117)?

No, it is not averaged in software at the moment.  There is however some hardware averaging on the
WB Jr. board itself.

To avoid confusion, looks like you are going to have to rename some of those registers from {bat} to {output}, eg 4117   IbattDisplay is not the battery current, but the controller outout current.

yes, you are correct in that it isn't ~really~ battery current.  But at the time I wrote that software it
was called Ibatt so I left it.

I have to say being able to finally distinguish these apart is just fantastic.!!

I would say you have done a great job interpreting all of this !
boB

[zoneblue]

Oh ok.  Clamp meter tallys better now.

Note that for 4371 the sign is reversed cf 4362. With 4371 its negative for discharging, positive for charging.

Is there a convention for that sign? Relative to the battery, positive for charging makes sense. Relative to positive bus sources, negative makes more sense. Shrug.

[zoneblue]

No, it is not averaged in software at the moment.  There is however some hardware averaging on the WB Jr. board itself.

Do you plan to populate another register with the 1s average? If so do you have the ability to make it the same 1s (as near as) as the Vcc/Icc datapoints?

Oh and its also worth mentioning that when i rebooted the classic yesterday it reset the value of register 4165 to decimal 1. (i have the MNLP dip set to custom). I just verified it just now by repeating it.

[boB]

No, it is not averaged in software at the moment.  There is however some hardware averaging on the WB Jr. board itself.

Do you plan to populate another register with the 1s average? If so do you have the ability to make it the same 1s (as near as) as the Vcc/Icc datapoints?

Oh and its also worth mentioning that when i rebooted the classic yesterday it reset the value of register 4165 to decimal 1. (i have the MNLP dip set to custom). I just verified it just now by repeating it.

Take a look at the  modbus document.  You will need to temporarily set the force flag to save the EEprom settings for
the Classic to remember the Aux setting if there is a power off and on.

Just set that flag once and it should save EEprom and the flag will clear itself afterwards.

A 1 second average will most likely happen, yes.

boB

[zoneblue]

Take a look at the  modbus document.  You will need to temporarily set the force flag to save the EEprom settings for the Classic to remember the Aux setting if there is a power off and on.

Just set that flag once and it should save EEprom and the flag will clear itself afterwards.

Didnt know that, thanks.

A 1 second average will most likely happen, yes.
boB

If you need more encouragement take a gander at this mornings curves. See when the cc ouput is changing alot, the WBjr gets out of sync... See the negative load watts, and the false spike up just before 10.