2023 Azimut Grande 32M: complete AC failure at 3 years, cause unknown

Azimut Grande 32 Metri, delivered June 2023, kept in Thailand. The Climma chilled-water air-conditioning plant (three Climma compressors, Multimacomp/Climma inverter drives, Martignoni 40 A line reactors) has failed completely at three years. All three compressors are out of service and we cannot find the root cause. Has anyone seen this pattern?

FAILURES AND REPLACEMENTS, IN ORDER
- Feb 2024 (9 months from delivery): chilled-water loop found contaminated with construction debris and sludge, no clear glycol. Loop flushed, components replaced, plant recommissioned under warranty.
- Mar to Jul 2024: unit 2 flow sensor (Code 7, then repeated Code 52), unit 1 flow sensor, and fan-coil chilled-water valves stuck closed in three cabins. Sensors and valves replaced.
- Aug 2024: whole-plant shutdown, diagnosed as a CAN bus fault; seawater pumps needed re-priming.
- Feb 2025: unit 2 seawater ball valve broken; unit 2 CAN module replaced.
- Jun 2025: fan-coil capacitors replaced under a manufacturer recall.
- Aug 2025: unit 3 inverter PCB failed and was replaced. Compressor found damaged; corrosion noted inside the system.
- Dec 2025 to Jan 2026: on generator, units 1 and 2 ran about 30 seconds then tripped DIS-VOLT; the generator slowed and stalled as compressors cycled. Normal on shore power. Unit 3 faults coincided with generator shutdowns.
- Jan to Feb 2026: unit 3 inverter failed again, IGBT power module charred. Second PCB and a fan-coil blower assembly fitted.
- May 2026: with unit 3 back in service, whole-plant shutdowns including the display, twice in 30 minutes at sea, no alarms logged. With unit 3 off, no power issues.
- 29 May 2026: insulation test. Compressor 1 about 110 ohms to earth, compressor 3 about 3 megohms, compressor 2 about 23 megohms.
- 29 Jun 2026: independent inspection finds the line-reactor terminals burnt, W phase reading over ten times the other two phases. The reactors are open-frame units mounted beside seawater piping.
- Jul to Aug 2026: unit 2 trips on flow alarm with plate temperature 62 C, fresh-water flow sensor replaced, then unit 2 fails insulation on 19 Aug. Plant now completely down.

WHAT HAS BEEN CHECKED
- Shore power is through a 65 kVA isolation transformer. IR survey: terminals 48 C, no hot joints; core clamp plate 89 C.
- Bonding and anode continuity tested on shore power and on generator.
- The 2024 flush did not hold: strainer sediment was still present in Jun 2026.
- Stray current has been suggested. Nothing has been proved either way.

QUESTIONS
1. Has anyone had Climma/Veco compressors fail one after another, and what was the cause in the end?
2. Burnt line-reactor terminals on one phase: supply, installation, or drive?
3. Three times the plant disturbed the supply only while unit 3 was running. Has anyone seen a failing drive pull a generator down, and was the drive or the supply the cause?
4. Any experience of this plant in tropical shore-power conditions?

If you have had this diagnosed, I would like to know what test finally showed the cause. Happy to share the technical file with anyone looking at the same problem.

Thanks all.

Fault codes / why it’s out of service now
There is no live MCX code currently locking the plant out. The plant is out of service because the compressors failed earth-insulation tests and were taken offline. In May 2026 the plant (including the display) also dropped power with no alarm logged.

Earlier codes / states recorded:

• Mar–Jul 2024: Code 7, then Code 52 (low / repeated fresh-water flow)
• Aug 2024: CAN bus fault; whole-plant shutdown
• Dec 2025–Jan 2026: DIS-VOLT after ~30 s on generator only; behaviour on shore power was normal
• Jul–Aug 2026: FLOW METER ALARM Code 7 on Unit 2; plate temperature about 62 °C; Comp 2 insulation later collapsed

Supply voltage from the ship bus
Chiller nameplate / design: 400 V / 3 ph / 50 Hz. Shore supply via Bi-mec 65 kVA Dyn11 isolation transformer (380/400 V taps). Generators: 2 × Kohler 55EFOZDJ, 400/230 V. I do not have a logged live bus voltage reading from the last trip; I can take one and post it if useful.

On the reactor photos
Photos show heat damage at the Martignoni line-reactor terminations (W phase and adjacent V) feeding the Climma inverter drives. An independent inspection measured a large resistance imbalance on that W phase versus U/V. Separately, there is documented chilled-water loop contamination. Multiple electricians have worked on the vessel. Azimut and Veco / Climma have not identified a root cause.

Plant: Climma / Veco CWS MODUL 3603 (3 × 120k BTU/h), Danfoss MCX. Looking for diagnosis path and similar Climma / Veco cases.

Happy to post further measurements if there are specific ones you’d want next.

Yes, this is my boat from new (Azimut Grande 32m, delivered June 2023).

On codes: when the plant lost power including the display, no MCX alarm was logged. Separately, the plant is also out of service because compressors failed earth-insulation tests.

We have now photographed the live MCX and EVENT LOG. Current display: MASTER ALARM, Code 23 CAN-BUS MOD 2 KO, plant SHUT DOWN, compressors 0/3, SWP ON / FWP OFF, water about 30.7 °C.

Documented codes / states (MCX text where numbered):

When (as recorded) Code / state Text Unit
Live now (7 Oct 2026) 23 CAN-BUS MOD 2 KO Master
19 Aug 2026 23 CAN-BUS MOD 2 KO Unit.2
19 Aug 2026 30 DRIVE ALARM Unit.2
11 Aug 2026 30 DRIVE ALARM Unit.2
11 Aug 2026 23 CAN-BUS MOD 2 KO Unit.2
27 Jul 2026 52 FLOW METER ALARM Unit.2
27 Jul 2026 22 CAN-BUS MOD 1 KO Unit.1
Jul–Aug 2026 (earlier photo) 7 FLOW METER ALARM Unit.2
Mar–Jul 2024 7, then 52 Low / repeated fresh-water flow Unit.2
Aug 2024 (diagnosed) CAN bus fault; whole-plant shutdown Plant
Dec 2025–Jan 2026 DIS-VOLT State after ~30 s on generator only Units 1/2
MCX clock may be wrong on some log dates. This is not a complete lifetime log; only the recent EVENT LOG pages photographed plus earlier notes.

Several electricians and HVAC technicians have already been on board. Azimut and Veco / Climma have not identified a root cause. A local YF member who knows marine electrical work in Thailand would still be useful.

Update on codes (MCX photographed 7 Oct 2026; clock may be wrong):
Live / active: MASTER ALARM list shows Unit.3 Code 24, Unit.2 Code 23, Unit.1 Code 22. Also Master Code 22 CAN-BUS MOD 1 KO and Unit.3 Code 30 DRIVE ALARM. Earlier today also Code 23 CAN-BUS MOD 2 KO with plant SHUT DOWN.

Documented chronology:

When (as recorded) Unit Code Text
02/01/2024 Units 1, 2, 3 30 DRIVE ALARM
Mar–Jul 2024 Unit.2 7, then 52 Fresh-water flow
Aug 2024 Plant — CAN bus fault; whole-plant shutdown
Dec 2025–Jan 2026 Units 1/2 DIS-VOLT Generator only
27/07/2026 Unit.1 / Unit.2 22 / 52 CAN-BUS MOD 1 KO / FLOW METER ALARM
11–19/08/2026 Unit.2 23, 30 CAN-BUS MOD 2 KO / DRIVE ALARM
7 Oct 2026 Master 24, 23, 22 Concurrent list
Several electricians and HVAC technicians have already been on board. Azimut and Veco / Climma have not identified a root cause. A local YF member who knows marine electrical work in Thailand would still be useful.

Correction to my first post: no corrosion has been found. The damage observed is electrical and heat-related.

Thank you all for your feedback so far. I've put together a website that sets out the facts of our yacht's air-conditioning failure: https://yachthelp.org

If anyone can help, or knows someone who can, please share it. Thank you.