command list cleanup

[segaura]

Hallo.
While working on creating a DBC database with the HPSU commands, I've found some maybe quirks in commands_hpsu.json.

Easiest to fix are these on "screed section":

• these two commands are the same
  t_screed_day5 61 00 FA 0B BD 00 00
t_screed_day7 61 00 FA 0B BD 00 00
• and same for these two
  t_screed_day21 61 00 FA 0B CD 00 00
t_screed_day22 61 00 FA 0B CD 00 00

but looking at the progression of the values and also observing that 0x0BBD = 3005 and 0x0BCD = 3021 the correct ones should be

t_screed_day7 61 00 FA 0B BE 00 00
t_screed_day22 61 00 FA 0B CE 00 00

Second group are these that are almost the same

t_flow_cooling 31 00 FA 03 DD 00 00
t_flow_cool 61 00 FA 03 DD 00 00

the first receives a 180#32 10 FA 03 DD 00 BE as answer and the second receives a 300#32 10 FA 03 DD 00 BE: is it correct to have them both listed?
I am considering of dropping the second one.

Third group is

quiet_mode 31 00 FA 06 96 00 00
aux_fct 31 00 FA 06 96 00 00

these are also listed as the same command, but after some dump and click I think that this is the correct value

aux_fct 31 00 FA 06 71 00 00

Last group is this, all three command are the same

outside_conf 31 00 FA 09 61 00 00
storage_conf 31 00 FA 09 61 00 00
pres_conf 31 00 FA 09 61 00 00

and this is correct, they are the same command, but the problem is with the values because they work on one or two bits each and not on one or two bytes like pyHPSU think.

Naming o the bit for outside_conf, s the bits for storage_conf and p the bit for pres_conf the last two bytes (data section 6th and 7th byte) are like this

0p10100000000sso

so, for instance

6804 = 0110100000000100

corresponds to

• outside_conf off (0)
• storage_conf sensor (4)
• pres_conf on (0x6800 = 26624)

and

2803 = 0010100000000011

corresponds to

• outside_conf on (1)
• storage_conf thermostat (2)
• pres_conf off (0x2800 = 10240)

issuing those three commands correctly requires reading the current value, altering the correct bits and sending it back as new value.
This is possible but I have no evidence that this ic currently managed correctly by pyHPSU.
Obviously these three are "less important" commands to control HPSU via pyHPSU so they will probably remain like they are for a while.
They are rather important for backup/restore feature, though, so they deserve some attention, in my opinion.

I have "fixed" first three group in my current development branch, if no objections emerge I can include that fixes in my next PR.

[svde]

t_screed_day7 61 00 FA 0B BE 00 00

Which is then the same as: t_screed_day6 61 00 FA 0B BE 00 00 So maybe this is better? t_screed_day7 61 00 FA 0B BF 00 00 with kind regards, / met vriendelijke groet, Stefan van der Eijk

…

On Sun, Feb 21, 2021 at 12:11 AM segaura ***@***.***> wrote: Hallo. While working on creating a DBC database with the HPSU commands, I've found some maybe quirks in commands_hpsu.json. *Easiest to fix* are these on "screed section": - these two commands are the same t_screed_day5 61 00 FA 0B BD 00 00 t_screed_day7 61 00 FA 0B BD 00 00 - and same for these two t_screed_day21 61 00 FA 0B CD 00 00 t_screed_day22 61 00 FA 0B CD 00 00 but looking at the progression of the values and also observing that 0x0BBD = 3005 and 0x0BCD = 3021 the correct ones should be t_screed_day7 61 00 FA 0B BE 00 00 t_screed_day22 61 00 FA 0B CE 00 00 *Second group* are these that are almost the same t_flow_cooling 31 00 FA 03 DD 00 00 t_flow_cool 61 00 FA 03 DD 00 00 the first receives a 180#32 10 FA 03 DD 00 BE as answer and the second receives a 300#32 10 FA 03 DD 00 BE: is it correct to have them both listed? I am considering of dropping the second one. *Third group* is quiet_mode 31 00 FA 06 96 00 00 aux_fct 31 00 FA 06 96 00 00 these are also listed as the same command, but after some dump and click I think that this is the correct value aux_fct 31 00 FA 06 71 00 00 *Last group* is this, all three command are the same outside_conf 31 00 FA 09 61 00 00 storage_conf 31 00 FA 09 61 00 00 pres_conf 31 00 FA 09 61 00 00 and this is correct, they are the same command, but the problem is with the values because they work on one or two bits each and not on one or two bytes like pyHPSU think. Naming o the bit for outside_conf, s the bits for storage_conf and p the bit for pres_conf the last two bytes (data section 6th and 7th byte) are like this 0p10100000000sso so, for instance 6804 = 0110100000000100 corresponds to - outside_conf off (0) - storage_conf sensor (4) - pres_conf on (0x6800 = 26624) and 2803 = 0010100000000011 corresponds to - outside_conf on (1) - storage_conf thermostat (2) - pres_conf off (0x2800 = 10240) issuing those three commands correctly requires reading the current value, altering the correct bits and sending it back as new value. This is possible but I have no evidence that this ic currently managed correctly by pyHPSU. Obviously these three are "less important" commands to control HPSU via pyHPSU so they will probably remain like they are for a while. They are rather important for backup/restore feature, though, so they deserve some attention, in my opinion. I have "fixed" first three group in my current development branch, if no objections emerge I can include that fixes in my next PR. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#41>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAELZNT6IXBVMYJWO5WS5GDTAA6QRANCNFSM4X6K37ZQ> .

[segaura]

Sure, my typo, you are right, thank you.

Do someone knows if some pyHPSU code deals with single bits (e.g. bit masks) like needed for the *_conf commands?

[Spanni26]

I don't think so. pyHPSU is not very intelligent. I spits commands to tha can bus and waits for (more ore less) fixed answers. I always wan't pyHPSU to build the commands by pieces (like discribed by the elster telegram http://juerg5524.ch/data/Telegramm-Aufbau.txt ), but I don't find the time to do it so. Am 21.02.2021 um 21:30 schrieb segaura:

…

Sure, my typo, you are right, thank you. Do someone knows if some pyHPSU code deals with single bits (e.g. bit masks) like needed for the |*_conf| commands? — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#41 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AC2I5HCORZAX64D2A3W2DZLTAFUOJANCNFSM4X6K37ZQ>.