Generator keeps cutting out mid-cycle - faulty AVR or something else?

by Heather Child · 1 month ago 26 views 4 replies
Heather Child
Heather Child
Active Member
14 posts
Joined Jul 2025
1 month ago
#22202

Had this exact drama with my Honda EU22i last winter when the static caravan's immersion heater decided to throw a tantrum — turns out it wasn't the AVR at all, it was the brushes on the alternator worn down to basically nothing, which causes exactly that "cuts out mid-cycle" behaviour people keep misdiagnosing as AVR failure.

Few things worth checking before you shell out on a new AVR:

  • Load surge at startup — some appliances (especially anything with a motor or compressor) pull 3-4x their rated wattage on startup, which trips the overload protection rather than the AVR itself
  • Fuel flow restriction — a partially blocked carb jet will let it idle fine but starve under load
  • Brushes and slip rings — dead cheap to replace and massively overlooked
  • Capacitor on brushless units — if yours is a brushless design, a failed run capacitor mimics AVR death almost perfectly

The AVR itself going faulty is possible obviously, but it's genuinely the last thing I'd replace rather than the first — they're not exactly cheap and fitting one to find it made zero difference is a special kind of miserable.

What generator model are you running and what's it powering when it cuts? Makes a massive difference to the diagnosis — my EV charging setup has killed more than one genny that was technically rated for the load but couldn't handle sustained output without overheating.

Anyone else here had the capacitor issue specifically? Feels underdiagnosed on this forum and I'd love to know if it's a pattern.

YEL_Marine
YEL_Marine
Active Member
12 posts
Joined Jun 2025
1 month ago
#22237

YEL_Marine | Posts: 847

@HeatherChild — worth also checking whether it's actually a governor response issue rather than electrical. On the EU22i specifically, if the inverter board is seeing voltage fluctuations it'll shut down as a protective measure, which looks identical to an AVR fault from the outside.

Quick test: run it under no load for ten minutes and watch whether it cuts out. If it stays happy, your problem is load-related — sudden demand spikes (immersion elements are notorious for this) overwhelming the governor's ability to maintain revs before the inverter board panics.

A dirty carburettor needle valve can cause exactly this — the engine stumbles under load, revs drop momentarily, and the inverter interprets that as a fault condition.

Clean the carb first. It's free and takes twenty minutes.

Renogy_Nerd
Renogy_Nerd
Active Member
40 posts
thumb_up 24 likes
Joined Jul 2023
1 month ago
#22247

@HeatherChild mine did exactly this in the shepherd's hut last February — ended up being a partially blocked carb jet causing the revs to hunt under load, which the generator's brain interpreted as "time for a dramatic exit."

Clean the carb before you start replacing expensive bits.

Sussex Boater
Sussex Boater
Active Member
26 posts
thumb_up 7 likes
Joined Feb 2024
1 month ago
#22276

SussexBoater | Posts: 312

Had the same saga on the boat — turned out to be a loose earthing strap vibrating itself free over a full season of tidal movement, which the Honda's overload protection then interpreted as a fault and threw a wobbler every 20 minutes like clockwork. Worth wiggling everything before you condemn the AVR, because replacement parts for these aren't exactly Fogstar prices. @YEL_Marine raises a fair point on the governor too — if the RPM dips under load you'll get the same cut-out symptom but it'll feel different; listen for the engine note dropping just before it trips.

SmartSolarMaster
SmartSolarMaster
Active Member
18 posts
thumb_up 8 likes
Joined Dec 2023
1 month ago
#22322

SmartSolarMaster | Posts: 203

Interesting thread — I've been keeping an eye on this as I run a Honda EU22i alongside my Victron MultiPlus for the garden office.

Before assuming AVR or alternator, have you checked the overload protection circuit? A few questions worth working through methodically:

  • What's the actual load when it cuts out — have you measured it with a clamp meter?
  • Does it cut immediately or after a few minutes of running?
  • Any smell or discolouration near the control panel?

The reason I ask: my setup cuts unexpectedly when the Victron's charger ramps up simultaneously with a compressor starting — that combined inrush briefly spikes the generator's perceived load. Could be something similar rather than a mechanical fault entirely. Worth logging the behaviour pattern before pulling anything apart.

Log in to join the discussion.

Log In to Reply