Implement KW protocol

[bertmelis]

Implement KW protocol

[Empor-co]

BTW: The other LED diodes arrived: the Vishay TSHA4401 (IR LED 875nm) seem to work fine too - since the SFH487 family seems to be end-of-sales.

[bertmelis]

Developement takes places in the develop-KW branch. I made a quick first commit but it is completely untested!

[Empor-co]

Adjusted the datapoints and TX RX:
VitoWifi.addDatapoint("outsidetemp", "boiler", 0x0800, TEMP); VitoWifi.addDatapoint("boilertemp", "boiler", 0x0810, TEMP); VitoWifi.setGlobalCallback(globalCallbackHandler); VitoWifi.setLoggingPrinter(&Serial); VitoWifi.enableLogger(); VitoWifi.setup(&Serial1, 16, 17);

First read succeeds, afterwards keeps failing/rebooting. The addresses seem a bit odd...

`rst:0x1 (POWERON_RESET),boot:0x13 (SPI_FAST_FLASH_BOOT)

configsip: 0, SPIWP:0xee

clk_drv:0x00,q_drv:0x00,d_drv:0x00,cs0_drv:0x00,hd_drv:0x00,wp_drv:0x00

mode:DIO, clock div:1

load:0x3fff0010,len:4

load:0x3fff0014,len:588

load:0x40078000,len:0

load:0x40078000,len:9880

entry 0x400789d8

Setup finished...

reading datapoints

Datapoint outsidetemp READ action added

Datapoint boilertemp READ action added

READ request on address 0008, length 2

succes

Datapoint outsidetemp action successful

Value received

boiler - outsidetemp: 13.6

READ request on address 1008, length 2

succes

Datapoint boilertemp action successful

Value received

boiler - boilertemp: 43.2

READ request on address 1008, length 2

succes

Datapoint Guru Meditation Error of type LoadProhibited occurred on core 1. Exce ption was unhandled.

Register dump:

PC : 0x400e4af3 PS : 0x00060530 A0 : 0x800d106f A1 : 0x3f fca820

A2 : 0xfefefefe A3 : 0x3f40103f A4 : 0x00060120 A5 : 0x00 000001

A6 : 0x00060120 A7 : 0x3ffc8920 A8 : 0x800d1938 A9 : 0x3f fca800

A10 : 0x400d17d4 A11 : 0x3f40103f A12 : 0x80083f1c A13 : 0x3f fc0bb0

A14 : 0x00000003 A15 : 0x00060023 SAR : 0x00000018 EXCCAUSE: 0x00 00001c

EXCVADDR: 0xfefeff02 LBEG : 0x400014fd LEND : 0x4000150d LCOUNT : 0xff fffffd

Backtrace: 0x400e4af3:0x3ffca820 0x400d106c:0x3ffca840 0x400d09a4:0x3ffca870 0x4 00ddda1:0x3ffca890

Rebooting...

ets Jun 8 2016 00:22:57

rst:0xc (SW_CPU_RESET),boot:0x13 (SPI_FAST_FLASH_BOOT)

configsip: 0, SPIWP:0xee

clk_drv:0x00,q_drv:0x00,d_drv:0x00,cs0_drv:0x00,hd_drv:0x00,wp_drv:0x00

mode:DIO, clock div:1

load:0x3fff0010,len:4

load:0x3fff0014,len:588

load:0x40078000,len:0

load:0x40078000,len:9880

entry 0x400789d8

Setup finished...

READ request on address 0000, length 0

succes

Datapoint Guru Meditation Error of type LoadProhibited occurred on core 1. Exce ption was unhandled.

Register dump:

PC : 0x400e4af3 PS : 0x00060530 A0 : 0x800d106f A1 : 0x3f fca820

A2 : 0xfefefefe A3 : 0x3f40103f A4 : 0x00060120 A5 : 0x00 000001

A6 : 0x00060120 A7 : 0x00000000 A8 : 0x800d1938 A9 : 0x3f fca800

A10 : 0x400d17d4 A11 : 0x3f40103f A12 : 0x8008398c A13 : 0x3f fca720

A14 : 0x00000000 A15 : 0x3ffca7f0 SAR : 0x0000001f EXCCAUSE: 0x00 00001c

EXCVADDR: 0xfefeff02 LBEG : 0x400014fd LEND : 0x4000150d LCOUNT : 0xff fffffd

Backtrace: 0x400e4af3:0x3ffca820 0x400d106c:0x3ffca840 0x400d09a4:0x3ffca870 0x4 00ddda1:0x3ffca890

Rebooting...

ets Jun 8 2016 00:22:57`

[bertmelis]

I just tested myself with an esp8266 and it doesn't work at all 😢
I have to redo the complete optolink FSM I guess.

The addresses is something strange. Endianess? Is the ESP32 different from the ESP8266 wrt endianess?

[Empor-co]

the crazy part is that it worked in the first iteration. but yes, looks like the msb block moved

[bertmelis]

Could you try again with the newest commit?
I'm not completely if the "reset to KW"-method will also work on devices not compatible with the 300-protocol.

Keep in mind that the KW-protocol only supports datapoints up to 2 bytes in length.

[Empor-co]

no luck so far: "Setup finished.." is displayed all the time, but no readings...

[bertmelis]

You're using the globalcallback-esp32? (with your customization wrt pins)
I must have changed something on the serial-initialisation then... If you don't even see the "reading datapoints" message, we're back at where we started.

I can only test myself this evening.

[bertmelis]

I had the #ifdef-guard for esp8266 removed for some reason but it keeps the esp32 from starting up. I just committed the fix after testing with my lolin32 board.

[Empor-co]

I'll be offline for about 3 weeks :/ will test it directly as soon as I'm back.

[bertmelis]

If you want, you can also test with the branch develop-merge_protocols

You'll have to add VitoWifi_setProtocol(KW); to your sketch, before setup().

[Empor-co]

Will try to find some time this weekend to pick-up testing...

[bertmelis]

Deleted the KW-branch as you can use the develop-branch and set the KW-protocol in your sketch.

[Empor-co]

Just tested the dev-branch and it works nicely!
Just found some typos in OptolinkKW, line 163 & 164 as well as optolinkp300, line 310, 31
success with two 's' at the end ;-)

[bertmelis]

Good news! Thanks for your testing and feedback!