Also, I forgot to mention that we checked the level of the machine and it is perfectly level and the tub is centered.

Thanks much for your reply. We will try running it with the lid up again but we have already established that the leak happens during recirculation and only on SOME small loads. We have never had a leak with a full load. So we're still perplexed.

Dadoes, here is what we found today. We lifted the machine top as you suggested (thanks) and found that during recirculation, some of the water was hitting the top of the inner part of the tub and splashing back out. The recirculation hose is properly seated, the ring around the top of the tub was properly seated, and we couldn't find anything else wrong. So as we understand it, the tub "floats" a bit after a certain point of filling? Does this explain why this leak was only happening on some small loads; if a load was just a little bit bigger the tub wouldn't have floated up yet when the recirculation begins? On a fuller load there was never a problem. Don't know why this suddenly started happening though. Are there any adjustments that might fix this problem? For now, we fixed it by putting a scrap of foam foundation sealer over the top of the recirculation hose so there is more of a seal between it and the ring that sits on top of it, and that has kept the water from splashing back out.

The basket does not (or shouldn't) float during recirculation. The water level in the tub is purposely kept below the float level. Floating disengages the basket from the drive cog so the agitator can oscillate separately ... the basket rotates at 25 RPM during recirculation so it must engage the drive cog for the duration (otherwise it would free-wheel, not be driven by the motor to rotate properly). You can visually observe when the basket floats, it shifts upward maybe 3/8", narrowing the gap between the basket rim and tub cover. The control board shouldn't allow recirculation to run if it senses the basket is floated. It will skip the recirculation phase if powered off and a cycle restarted after the machine is filled and agitation started. Part of the automatic water level process is determining when the basket has floated by pulsing the motor during fill (after recirculation) to sense the drive load and rotational inertia.

Thanks, Dadoes. Here is what we have just done: first, a full load of wash. We measured the gap between the rim and cover at various points: at start: 12 mm; at start of recirculation: 12 mm; at start of agitation: 4 mm (just as you described). But then we did a small load and measured the gap again. At start: 12 mm, at start of recirculation and during agitation: 0-4 mm (the gap was a little uneven around the tub, to the point where it was rubbing a bit at some spots). The basket seemed to be rotating normally during the recirculation. So the basket has floated up by the time recirculation has started, but only during the small load. The recirculation water was spraying out in a very thin sheet because the gap was so small and this time some leaked on the floor again despite our makeshift foam gasket. Also, it looked like there was way too much water in the tub for the small load by the time agitation started (using the automatic water level setting). We're going to check the pressure sensor tube to make sure there isn't water trapped in it, but other than that, we don't know what else to do. Do you have any suggestions?

A little water may get into the air dome and tube at the tub-end when it fills but should drain completely, unless the air dome has gotten clogged and can't fully drain. Could be the tube has a crack or break and is leaking pressure ... thus filling too high. However, that should result in overfilling on all water levels. Do the other levels also seem too high? There's a flood protection function that turns the pump on if the tub level exceeds the highest level by a margin ... but of course that depends on accurate pressure reading from the transducer. The pressure transducer may have gone bad, off-spec, if the air dome and pressure tube prove to be OK. The transducer is integral to the board and has no adjustment so the board must be replaced if that's the problem.

Automatic vs. manual selection doesn't affect the fill levels. Automatic sensing settles on one of the five levels, same as a manual selection.

Two other things I can come up with that could affect the water level and/or the basket's float characteristics on a small load vs. large --
1 - Weight of the basket. There are liquid-filled gyroscopic balance rings integral to the rim and base of the basket. The weight of the basket, including the fluid, must be within a specification range for proper operation. The basket must be replaced if fluid leaks out of either ring, which is rare and would also cause spin-balancing issues.
2 - Wrong machine size is set. F&P agitator-style washers on the Aus, NZ, and other world markets are sold in several capacity sizes. US models are all the largest size but the control board must still be set for the size of the machine. Wrong size setting makes for wrong water levels and other operational characteristics. Size is set via a "hidden" key pattern but I suppose it's possible for someone to accidentally change it, except that the wrong sizes would be for smaller capacity and presumably lower levels. But, check it anyway. With Power off, press/hold Wash Temp Up, then press Power at the same time. Should get four beeps and the panel turns on. Spin Hold light indicates the correct size. Press Spin Up to change if necessary, then Power to confirm & exit.

Dadoes, thanks so much for all of your time on this!!!! Today's report: the water level on a small load looks right today. Maybe the level was off yesterday because we started and stopped the machine several times during that small load. Today we started a small load with no laundry in it, and the basket floated up immediately. We then added the laundry and confirmed that the water level was right (but of course the recirculating water sprayed all over the place). I did a large load yesterday and the water level was correct.

I don't think we've had any trouble with the spin-balancing, but how would you check the gyroscopic balance rings? And if that was the problem, wouldn't it affect all load sizes?

We checked the machine size and it was set correctly.

"Today we started a small load with no laundry in it, and the basket floated up immediately. We then added the laundry ..."

Is your normal procedure to start the machine and load the clothes after it has filled? Hopefully not, as that may prevent auto water level sensing from working properly. Operation for manual water level selection is not affected.

Another possibility came to mind ... incomplete draining (or dripping/leaking from the inlet valves) would leave excess water in the tub which could cause too-high a fill level for Eco Active ... except not really because the water level sensor should control the fill accordingly.

Or ... if the basket sometimes isn't dropping from float upon draining, which can be caused by residue/mineral deposits on the drive shaft/hub area. This situation should trigger a fault code since the machine can't spin with the basket in float position.

High-sudsing detergent could possibly have an effect on premature basket float but I've not seen that happen in my usage experience.

Proper method to check the balance rings is by weighing the basket to confirm it's within spec -- 23.10 lbs +/- 9.7 oz. Leakage of the balance ring fluid would require some sort of gross damage to crack the ring(s) ... or deliberately drilling the rings to drain the fluid (such as if someone hears the fluid sloshing when jostling the basket and thinks it's retaining water where it shouldn't).

No we don't usually load clothes that way! It was just something we wanted to look at. The basket always drops after draining and we haven't done anything that would damage the rings. And we don't use high sudsing detergent.

I'm out of suggestions at this point, without having access to directly observe/examine the machine.

Thanks for your time!