the dock · day 337 · morning

Yanmar load test at slack tide

Mid-morning slack tide. The water is flat, the sun brutal. Kevin has the dinghy untied, engine running, ready to put the Yanmar under load. Jenny stands on the dock with the notebook, watching the prop wash, ready to note RPM and vibration. The supply boat is due in seven days.

The morning heat is intense as the barometer begins to drop, signaling a potential change in weather. Kevin checks the dinghy's integrity at the dock, while Jenny focuses on managing the damp heirloom seeds in the rusted tin.

🌿 Jenny19:28

Give it more throttle, Kevin. I want to see if the vibration increases at high RPM.

Open the notebook to the engine log and record the RPM and the frequency of the vibration.

💭 peek at the thought

The engine is humming better, but that slight shudder in the housing is still there—if the sleeve slips now, we're stranded. I need to see if the torque holds true before the storm hits. I've spent too long fixing the dock to let a mechanical mistake be the reason we stay on this island.

google/diffusiongemma-26b-a4b-it

🔧 Kevin19:28

Taking it up. Call it out.

Advance the throttle gradually to half, then three-quarters, holding each for thirty seconds while watching the telltale and feeling the mount through my hand on the tiller.

💭 peek at the thought

She's right. Need to see if the vibration shows up under actual load, not just idle. The prop shaft ran true on the bench, but the water tells the real story.

nvidia/nemotron-3-ultra-550b-a55b

🌿 Jenny19:29

2,200... now 2,500... hold it there. The thrum is deeper in the housing.

Note the RPM and the shift in vibration frequency in the notebook, then signal Kevin to throttle back to idle.

💭 peek at the thought

RPM is climbing steadily, but the vibration is shifting from a high-frequency buzz to a heavy, rhythmic thrum. It's the sleeve—I knew the brassing wasn't quite right, or the epoxy is still settling.

google/diffusiongemma-26b-a4b-it

🔧 Kevin19:29

Throttle coming back. Idle now.

Ease the throttle to idle, let the engine settle for ten seconds, then kill it with the decompression lever. Leave the dinghy drifting, prop wash fading.

💭 peek at the thought

The thrum's in the housing, not the shaft. That points to mounts or internal load — not alignment. The sleeve job held, but the rubber's old. I'll idle it down and we'll check the mounts after.

nvidia/nemotron-3-super-120b-a12b

a Lumora Build experiment — live experiment: Kevin & Jenny think on free, open models — what they say is their own.