register availability across models.

[divenal]

In commit ddafa2a Dewet added holding registers up in the 4000 range. These aren't mentioned in the giv_tcp version. Any idea what those are about, or where they came from ?

The block starting IR(120) isn't currently used - is that reserved for expansion for batteries in the future ?

The giv_tcp code has some more input registers starting IR(180) and IR(240) which we don't yet have.

More generally, do you know much about the availability of registers across the different hardware models / firmwares? Is there a simple heuristic we could use such that, given the model from HR(0) and the firmware from HR(21) ? Doing a probe for possible register sets is a bit time consuming since we typically have to allow for timeouts and send repeats.

I'm guessing any inverter use a strict subset of the total defined register set ? ie whenever registers are added, they don't tend to repurpose an existing register number on different hardware?

[hoggyhoggy]

Yes I do know what they are but I can't share the doc because of an NDA I'm under. All I will say is that those ones he has added are the only ones in the 4K range I'm aware of existing.

The 180-240 range are mostly for the newer FW's with extra timeslots:
I think i mapped this as below based off the old Inverter.py -

163, # REBOOT_INVERTOR
242, # CHARGE_TARGET_SOC_1
243, # CHARGE_SLOT_2_START
244, # CHARGE_SLOT_2_END
245, # CHARGE_TARGET_SOC_2
246, # CHARGE_SLOT_3_START
247, # CHARGE_SLOT_3_END
248, # CHARGE_TARGET_SOC_3
249, # CHARGE_SLOT_4_START
250, # CHARGE_SLOT_4_END
251, # CHARGE_TARGET_SOC_4
252, # CHARGE_SLOT_5_START
253, # CHARGE_SLOT_5_END
254, # CHARGE_TARGET_SOC_5
255, # CHARGE_SLOT_6_START
256, # CHARGE_SLOT_6_END
257, # CHARGE_TARGET_SOC_6
258, # CHARGE_SLOT_7_START
259, # CHARGE_SLOT_7_END
260, # CHARGE_TARGET_SOC_7
261, # CHARGE_SLOT_8_START
262, # CHARGE_SLOT_8_END
263, # CHARGE_TARGET_SOC_8
264, # CHARGE_SLOT_9_START
265, # CHARGE_SLOT_9_END
266, # CHARGE_TARGET_SOC_9
267, # CHARGE_SLOT_10_START
268, # CHARGE_SLOT_10_END
269, # CHARGE_TARGET_SOC_10
272, # DISCHARGE_TARGET_SOC_1
275, # DISCHARGE_TARGET_SOC_2
276, # DISCHARGE_SLOT_3_START
277, # DISCHARGE_SLOT_3_END
278, # DISCHARGE_TARGET_SOC_3
279, # DISCHARGE_SLOT_4_START
280, # DISCHARGE_SLOT_4_END
281, # DISCHARGE_TARGET_SOC_4
282, # DISCHARGE_SLOT_5_START
283, # DISCHARGE_SLOT_5_END
283, # DISCHARGE_TARGET_SOC_5
285, # DISCHARGE_SLOT_6_START
286, # DISCHARGE_SLOT_6_END
287, # DISCHARGE_TARGET_SOC_6
288, # DISCHARGE_SLOT_7_START
289, # DISCHARGE_SLOT_7_END
290, # DISCHARGE_TARGET_SOC_7
291, # DISCHARGE_SLOT_8_START
292, # DISCHARGE_SLOT_8_END
293, # DISCHARGE_TARGET_SOC_8
294, # DISCHARGE_SLOT_9_START
295, # DISCHARGE_SLOT_9_END
296, # DISCHARGE_TARGET_SOC_9
297, # DISCHARGE_SLOT_10_START
298, # DISCHARGE_SLOT_10_END
299, # DISCHARGE_TARGET_SOC_10
305, # PV_INPUT_MODE
311, # LOCAL_CONTROL_MODE
318, # BATTERY_PAUSE_MODE
319, # BATTERY_PAUSE_SLOT_START
320 # BATTERY_PAUSE_SLOT_END

[hoggyhoggy]

As for the "re-use of old registers" I don't THINK that's happened. I do know that the move from "old" to "new fast response" firmware on Gen1 requires the inverter registers to be wiped before being rebuilt on the new FW. HOWEVER that being said I don't recall seeing GivTCP accounting for any of that (it still works as is I think?) other than whatever this is for: https://github.com/britkat1980/giv_tcp/blob/main/givenergy_modbus/model/inverter.py#L356C5-L357C57
Which i've never cross-referenced against the standard registers to confirm what, if any, have moved.

[hoggyhoggy]

Newer stuff - 3 Phase, the Gateway, the CEMS unit - I have no visibility of registers for. I can attempt to ask for them from GE (provided they still want to talk to me) but that may have to Britkat to advise on how they map out.

[divenal]

Yes I do know what they are but I can't share the doc because of an NDA

Does the NDA also prevent you from publishing code which makes use of the secret information. Because that could be quite a handicap...

[hoggyhoggy]

I'd hope not given Britkat signed the same one presumably. It does make life more awkward though.