Memory dump: hydro board software 20007902 (EBVH16S23DJ9W / ERLA16D)

[griezla]

Follow-up to my comment in #599, as asked. Thanks for the quick answer!

Dumps (beta.27, /api/memory R reads, 64-byte blocks, all ok:true), taken 2026-11-10 11:23–11:27 local time, attached:

• dump_20007902_3000-5FFF.txt (192 blocks)
• dump_20007902_0800-0FFF.txt (32 blocks)

Unit: EBVH16S23DJ9W + ERLA16D, iu_sw_id 20007902, AS3608729-26, ESP32-WROOM-32 on X10A.

State at the time of the dump

• Compressor stopped, 3-way valve on space heating
• Outdoor (R1T) 2.3 °C
• LW setpoint (main) 33 °C
• DHW setpoint 40 °C (schedule in eco), tank 42.3 °C
• Room setpoint 21.5 °C (Madoka)

Interface settings

• Main zone LWT setpoint mode: weather dependent, curve −20 °C → 42 °C, +15 °C → 28 °C
• Tank setpoint mode: scheduled + reheat (comfort 45 °C, eco 40 °C, reheat 43 °C, hysteresis 5 °C)

I'm away from home now; this evening I'll add photos of [4-08], [4-09], [5-05], [5-09], Information → Operating hours and Energy data (and redo the dump at the same moment if that helps).
dump_20007902_0800-0FFF.txt
dump_20007902_3000-5FFF.txt

[raomin]

Thanks @griezla, the dumps were enough: 20007902 is 20007903 with each block shifted a little, and everything you listed reads where it should: the curve 28/42, the 33 °C target (33.08), the limits at 20 kW. It is mapped in the next dev build: leaving water target, power limit, meter and counters.

Tonight's photos are the check. From your dump I expect: [4-08] none, [4-09] power, limit 1 at 20 kW / 50 A; compressor hours 10939 (heating) and 6958 (hot water), pump 21313 h, 36449 starts; heat produced 115715 kWh, electricity 38378 kWh; leaving water range 18 to 50 °C. Tell me which match. No new dump needed.

Then step 1, read-only: tick the leaving water target only (your tank is scheduled, that card will say "not curve mode") and check the card shows 20007902, a range of 18 to 45 and the curve 28/42.

[raomin]

THe support is up on the Development builds. System → Firmware update → Development builds

[griezla]

Photos taken this evening (Nov 10, ~19:30). Your predictions match:

Operating hours (Information → Operating hours)

• Compressor, space heating: 10940 h (you: 10939)
• Compressor, hot water: 6959 h (you: 6958)
• Pump: 21316 h (you: 21313)
• Backup heater: 192 h heating / 364 h DHW; BUH step 2: 38 h / 352 h
• Compressor starts: not shown on this MMI as far as I can find

Energy data, since installation

• Electricity: heating 17939, DHW 20439, total 38378 kWh (you: 38378)
• Heat produced: heating 60524, DHW 55218, total 115742 kWh (you: 115715, a few hours later)

Power consumption control: mode None ([4-08] = 0); the type/limit items are not shown in this mode, so I can't read [4-09]/[5-05]/[5-09] without changing the mode (I'd rather not).

Other info (Information → About): user interface software V07.11.00 (ID20010E11), indoor micon ID0792, EEPROM AS3608729-26, outdoor ERLA16DAW1 micon ID802B, WLAN gateway 4.2.4.

Ready for step 1 (leaving water target, read-only) when the dev build is out. Thanks!

[griezla]

Step 1 done on dev-35cf15c, leaving water target ticked only, nothing written:

• status ready, software 20007902
• range 18–45 °C
• curve 28 / 42 (effective target 35 °C at the moment)

Unit meter values on MQTT match the interface: heat produced 115744 kWh (heating 60526.5 / DHW 55218.7), electricity used 38378 kWh, compressor hours 10941 / 6959, pump 21316 h, compressor starts 36453.

Great work, thanks a lot! Is there anything else you'd like me to check (e.g. a test write of the leaving water target, then restore)?

[raomin]

Thanks @griezla, that settles the reading side: 20007902 is confirmed, hours, energy and meter included. The limitation mode None is also what the board reads, so leave it as it is.

Yes to the test write, it would be the first one on this software. Two stages, and the first leaves nothing behind:

Stage 1, working memory only. Settings → Experimental → tick "Keep the targets in the unit's working memory only", save. The range on the card becomes 18–50. Apply a target 2 °C under what the unit aims for (33 if it still aims for 35). Within a minute the card and the leaving water setpoint on the unit's interface should show it, while the curve in the interface's settings stays 28/42. Then "Restore the unit's curve": the card shows 28 / 42 again and the target goes back to the curve's value. Nothing is saved in the unit in this mode; switching the unit off and on would undo it too.

Stage 2, saved in the unit. Untick the option, save, and apply the same target. After about half a minute the card says "Saved in the unit." Then "Restore the unit's curve": the target goes back at once, the unit's settings follow within five minutes (the button disappears when it is done). Then check on the interface that the curve is 42 at -20 °C and 28 at 15 °C. This stage writes the two points of your curve in the unit's settings. Your dump shows them exactly where I expect them, so I am confident, but nobody has done it on 20007902 yet. The interface may keep showing its old curve while the target is applied: what counts is the card and the setpoint.

Tell me what the card and the interface showed at each stage.

[griezla]

Both stages done on dev-35cf15c, everything as you described:

Stage 1 – working memory only

• Applied 31 (unit aimed for 33): card "Aims for 31 °C", the interface's home screen showed 31 at the leaving water, LW setpoint (main) 31.
• Restore: back to "follows its own curve, 42 °C to 28 °C, aims for 33 °C", interface 33.

Stage 2 – saved in the unit

• Unticked the option, applied 31: saved after ~30 s, interface home screen 31.
• Restore: target back to 33 at once, restoring for about 5 minutes, then done, curve 28/42.
• Interface → main zone → weather-dependent curve: 42 at −20 °C, 28 at 15 °C, shift 0 – exactly as before.

So the target write works on 20007902 in both modes, no errors. Thanks a lot for the quick support!

[raomin]

Thanks @griezla, that completes it: 20007902 is confirmed, reading and leaving water target in both modes, on the first unit with this software. It goes into beta.28 as supported. Nothing more needed from you! :)

[griezla]

Updated to beta.28, everything works: targets card shows 20007902, range 18–45, curve 28/42, and the unit meter matches the interface. Thanks!

One question. I saw #655 (installer settings on 20002203, read only). Would you be interested in the same for 20007902? I'm going through my installer settings tonight anyway, so I can provide:

• photos of the full field settings overview on the MMI (all codes with their values),
• a memory dump taken at the same time,
• before/after dumps for a few settings I plan to change: LWT modulation [8-05], max modulation [8-06], overshoot [9-04], room temperature hysteresis [9-0C], and outdoor temperature averaging [1-0A].

Tell me which ranges to dump and whether a settings export from the MMI (USB) would help more than photos. The settings I care about most are the ones above, not only the temperature ones.