Possibly stupid question: won’t ESPHome output the right raw ir codes needed via configuring the transmitter component i.e.
remote_receiver:
pin:
number: GPIO23
inverted: True
mode:
input: True
pullup: True
dump: raw
id: ir_receiver
tolerance: 55%
?
in theory it should but I had better luck with the rpi setup and using LIRC. The format recorded is I believe a little different.
@hagak I got a Pi set up with a receiver and lirc. I’m able to capture the commands however I’m a little lost on this part:
Should you wish to use this code the timing values are defined in the beginning of the program, you will probably need to adapt it.
It’s probably staring me right in the face but I’m not sure what he means about adapting. I capture the raw output, I get the 6 timing values at the top of the columns but not sure what to tweak. When I compile and run his decode program I get a string of gibberish from the most part, so something is off.
Am I missing the obvious? I can send you a raw file, for example the one I did when pressing on/off if that helps.
his example is using a remote protocol very similar so the timings he has are fine.
Something is off then. Example, here’s a “baseline” from pressing on/off (standard settings) capture:
sudo mode2 -d /dev/lirc0 -m > off_daikin_new
File output:
Running as regular user pi
11725325
11697837-space
9837 9793 9866 9724 4634 2470
379 349 350 987 317 976
393 308 395 940 358 351
390 342 347 388 427 287
403 940 327 382 384 347
398 938 359 346 384 347
384 349 389 933 363 342
386 346 417 305 396 348
380 947 359 351 396 338
384 942 365 346 381 350
419 906 360 350 379 353
375 358 384 351 346 386
381 352 347 385 382 351
358 964 330 383 378 352
383 352 379 354 350 383
376 357 376 356 376 946
334 379 376 356 383 358
378 926 377 945 365 345
380 349 380 352 381 955
351 349 416 311 393 345
378 346 379 938 367 920
379 924 381 352 377 359
375 925 377 20305
348 954 343 361 375 356
391 336 373 362 375 952
350 357 368 952 328 384
346 386 347 388 343 387
345 386 346 386 346 386
345 418 318 408 322 410
321 412 320 411 320 411
321 415 318 412 319 416
320 413 320 413 320 413
292 442 316 414 323 420
312 417 316 418 314 417
316 415 313 419 287 445
312 422 310 422 308 423
309 429 307 424 303 419
310 437 295 424 308 423
308 423 310 447 285 426
310 446 286 447 287 446
286 446 285 446 285 447
288 447 284 449 294 1009
286 1030 273 446 288 1025
276 445 286 446 286 448
285 446 4553 138057-pulse
Run that through decode though and I get weird results:
./decode off_daikin_new off_daikin_new.decoded
result:
HHµÒb0C80^T000<87>^T00^R000^M0000^[00^V00ò00000000,0000000^Z000^F^Y000#0000^R^@^D00^E¹"0000 0 000000000000000000000000000g00000000000e0fgfeehdYfYh\ffe^Q]
~
I tried removing the “Running as regular user Pi line” at the start but no real difference.
before you run it through decode you want to remove all NON-numeric data except spaces
Done that, still bad output. All that program does is round values and convert to binary so I’m struggling with what’s going wrong.
Oh yeah you do need to adjust the ranges he has in his. I would make the LONG around 900 and seperator 20000
Improved:
sudo mode2 -d /dev/lirc0 -m > off_daikin_new
Output (non-numeric removed):
9845 9768 9843 9759 4634 2496
359 344 389 937 364 939
400 297 401 941 364 342
387 344 390 345 346 404
378 935 391 313 426 312
383 941 359 342 387 345
384 364 409 904 362 345
376 356 385 947 359 344
389 343 386 348 377 357
385 345 390 343 389 343
377 356 388 939 362 348
388 345 387 347 400 334
386 345 351 381 387 346
385 940 369 342 425 289
408 345 390 343 386 346
383 349 386 345 386 950
360 342 391 350 379 404
329 940 363 952 340 351
379 354 409 321 381 941
360 368 362 360 380 362
363 350 348 980 377 918
354 953 330 967 366 942
320 387 377 20296
352 924 385 344 380 352
417 317 381 339 394 908
381 350 394 913 414 308
388 351 383 350 399 335
376 356 381 351 382 351
350 385 380 353 381 353
375 355 376 356 369 366
342 389 376 357 348 387
349 383 397 345 359 369
381 352 368 364 372 360
344 386 346 388 344 393
337 388 343 398 338 382
354 401 318 416 318 409
321 414 314 419 316 412
320 415 357 373 318 410
320 414 316 415 316 419
315 420 361 399 308 390
346 385 319 405 314 415
315 417 313 420 313 991
310 1016 287 423 311 1017
284 422 310 423 310 423
310 438 4577 135115
./decode off_daikin_new off_daikin_new.decoded
Decoded output:
½»b01101000010010001001000000001000000010^A0000010001100010000111110(1000010100000000000000000000000000000000000000000000000011g1d000)]
Still wrapping my head a little around how the values interact, I get the math but not strong in C so figuring out what to tweak is difficult.
@hajak I’m at a standstill with this unfortunately. I can capture all the raw codes, starting from a standard baseline config from which I create one new file per button press for, but running the rounding program doesn’t convert to binary. I’d be willing to pay to help get this in at this point if you are interested.
if you get me the raw recordings with labels of the STATE for each one. Need to know what each setting is for each one, MODE, TEMP, FAN, ETC.
I will see what i can do. And not really into this for the money so dont worry about that.
Thanks! Do I need to run through all the iterations based on mode or can I just set it to one config i.e. “Cool, Fan low, 75, no swing, no eco” and then do each command/state per file from that baseline?
You just need a baseline to work from.
Cool, Fan Low, 75
Cool,Fan Med, 75
Heat, Fan Low, 75
etc.
Please provide 4 temperature ones.
I’ve gotten the first round of raw codes ready so we can see if they work for you.
Files ordered by change date, I started with “cool_75f_swing-off_fan-low_on-off” that is I had the remote at cool, 75f, no swing, low fan, no aux modes (powerful, quiet, eco, etc.) enabled and then pressed the on/off button and recorded it to the file.
I then created each new file based on a single button push, the change is in the file name i.e. dry_75f* means I used the base config above and set the mode to “dry”, and so on.
It’s a pretty tiny zip file - how do I get it to you? I don’t see a way to attach files here.
Your unit has “powerful, quiet, eco” modes? Is this a Daikin Ceiling Cassette unit? This sounds more like the wall units that use the standard Daikin module. Have you tested that module?
I have, that it didn’t work is what led me to discover your work on BRC remote support.
It’s a 17 series, indoor model FTXB12AVXJU Outdoor RXB12AXVJU. Operations manual: (pictured on 1st page): https://backend.daikincomfort.com/docs/default-source/product-documents/residential/manuals/operationmanuals/operation-manual_17series.pdf?sfvrsn=5d422a26_8
Hmm yeah this remote is much closer to the original remote. The ceiling cassettes that use the remote i built this for do not have those extra modes. I will look at it but it is probably not something i want to add to this module since it is probably a completely different protocol
Also Daikin really could use some unification in their product lines, good lord they seem to be all over the place. Also why I did not bother with their Wifi interface because it looked like a closed system and the IR I could atleast write some code to work with it.
I understand. Appreciate any guidance or insight from the files you can spare.
@hagak I’ve discovered a few interesting things:
The Daikin protocol used with these remotes is 128. I don’t believe that has been upstreamed to ESPHome but it is in the ESP8286 IRRemote library. You can see it referenced in here Starting line 552 you’ll find the correct frequency and bits for this protocol along with the supported modes. It is further broken out in the class reference: IRremoteESP8266: IRDaikin128 Class Reference
I’m fooling around with the library with custom components/libraries in ESPHome but I’m curious if the above is useful to you as well. I haven’t gotten it working yet but that may be more to my unfamiliarity with ESPHome config syntax.
Edit: Well, well. I’ve managed to get the unit to turn on and off. ATM when I try to change the temp. it turns the unit off instead but that’s just something in the code I hacked together. I’m just delighted that much worked, and also confirmed I’m getting enough range from my IR led. More to come 