Sunny Boy SB1700 - almost working

[Chris Phillips]

Hi,

I have a SMA Sunny Boy SB1700 DC-AC Inverter with Bluetooth Piggy-Back Module & SMA Sunny Beam with

Bluetooth which talk to each other fine with the default password.  I have installed sma-bluetooth on a Raspberry Pi: -

% uname -a
Linux pi1 3.18.7+ #755 PREEMPT Thu Feb 12 17:14:31 GMT 2015 armv6l GNU/Linux

% cat /etc/debian_version
7.8

Initially I had no joy with smatool.  I had to create the LiveData table manually, using the SQL I found in sma_mysql.c, as neither --INSTALL or --UPDATE did it for me.  After this, I then had the following error: -

% ./smatool -v                                                           
Auto set dates from 2000-01-01 00:00:00 to 2015-04-01 15:57:00
bluetooth signal = 74%
serial=77:4c:60:c0
bluetooth signal = 74%
bluetooth signal = 74%
$TIMESTRING extraction failed. Check password!!!

After reading this post: https://groups.google.com/d/msg/sma-bluetooth/CLl0ARROy8s/wZGSzxYIOcoJ I changed the Inverter value in smatool.conf from my best guess of 1700TL, to 3000TL and re-ran smatool.  It's only gone and worked!  Well, there's still nothing in the DayData table, but the output is much more exciting now: -

% ./smatool -v
Auto set dates from 2000-01-01 00:00:00 to 2015-04-03 10:26:00
bluetooth signal = 74%
serial=77:4c:60:c0
bluetooth signal = 74%
bluetooth signal = 74%
2015-04-03 10:26:32 Unit Serial          = 3821139 none
2015-04-03 10:26:32 NO DATA for 1f 82 = 8001 NO UNITS
2015-04-03 10:26:32 NO DATA for 20 82 = 9020 NO UNITS
2015-04-03 10:26:32 NO DATA for 21 82 = 601 NO UNITS
2015-04-03 10:26:32 NO DATA for 21 82 = 636 NO UNITS
2015-04-03 10:26:32 Total Power          = 393 Watts
2015-04-03 10:26:32 Max Phase 1          = 1700 Watts
2015-04-03 10:26:32 Max Phase 2          = 1700 Watts
2015-04-03 10:26:32 Max Phase 3          = 1700 Watts
2015-04-03 10:26:32 Line Voltage Phase 1Volts = 243 Volts
2015-04-03 10:26:32 Line Current Phase 1Amps = 2 Amps
2015-04-03 10:26:32 Grid Frequency       = 50 Hertz
2015-04-03 10:26:32 NO DATA for 1f 4a = 100 NO UNITS
bluetooth signal = 74%


28/4/1974 17:34:09  total=72058914471936.000 Kwh current=0 Watts togo=0 i=11 crc=0
14/7/1970 05:20:14  total=72057589071872.000 Kwh current=-15977278341120 Watts togo=0 i=23 crc=0Date Error! prev=136398849 current=16777214

This is most encouraging - that Total power value tallies with my SunnyBeam!  Here's what my DB contents looks like now (note the lack of DayData): http://hastebin.com/wijuyelefo.1c

I've got smatool running via cron, kinda hoping I can get it working soon.  I've put the contents of my "smatool.conf" file and the output of "./smatool -v -d" here: http://hastebin.com/fokizidina.vhdl

My house is Mac/Linux only - no Windows - so I don't know how to proceed in debugging and or fixing it & any advice is welcomed.  
Now, at this point, I'm not even sure if my inverter is supported!

I'm not exactly sure that I've got the hang of Google and Groups (I can't see my post(s) online when I last posted), so this may be an elaborate "note to self"...
--
Cheers 

[Daniel Kammermann]

I seem to recall errors like that setting up my system (it's been nearly 5 years...).

Looks to me like the date/time on the pi and/or inverter is screwy - the software is pretty sensitive to it being correct.

I remember things being better once I used the windows software (on my work laptop since the household is Linux only) to communicate with the inverter, do some configuration and set its date correctly.

IIRC there seems to be a date-setting portion of the messages when communicating with the inverter (from both the windows software and smatool), but that seems to act as more of a date offset for the purposes of that connection and didn't in my experience seem to "stick" like the explicit date-setting function of the proper windows software did.

In fact now I run the tool myself I see a familiar message:

[82] 2015-04-15 21:49:14 Found string we are waiting for

[83] 2015-04-15 21:49:14 Sending

reporttime=1429100351, idate=1429086354, Sleeping for 74 secs to allow inverter to catch up

...so I might need to run the windows software to set the date and remove the 74 sec difference.

On top of all this, there's a chance the inverter's communication is a bit different to what smatool is thinking it should be - those are some pretty big numbers for total kWh and a negative (large) current Watts value also doesn't look correct.

Cheers

--
You received this message because you are subscribed to the Google Groups "sma-bluetooth" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sma-bluetoot...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Stephen Collier]

Hi All,

Sorry for being absent for such a long time. I've had quite a few issues
and now I finally can get back to this project. I will first move to
github and make some progress there as I am aware google code is
closing. The mailing list will continue.

I plan to migrate the site in the next 24hrs. I'll post the new location
on the mailing list.

Cheers
Stephen Collier

[Chris Phillips]

Great news!

[Stephen Collier]

I've set up a new github repository at
https://github.com/sma-bluetooth/sma-bluetooth

It is now live and a work in progress.

All future development will take place here.

Anyone wanting commit access just let me know.

Cheers
Stephen Collier

[Ralph McKnight]

thanks steve, my system is still running nice (for years) now.

[Stephen Collier]

Run it with --debug  it gives more information. Once the date shows as 1970 or thereabouts nothing can be trusted about the data. It means the decoding was bad. I have an idea as to what is causing it but if you post with debug it may confirm it.

Cheers
Stephen

[Chris Phillips]

Hi Steve,

Welcome back!

For a bit of balance, I've not managed to get it fully working with my older SB1700 system :(

However. I'm excited that I'm more likely to get it sorted with an active and enthusiastic maintainer & the code living on GitHub :)

Cheers,

--
ChrisP