so this has been bugging me for a couple weeks now. my 7300 is only putting out maybe 30 watts on a good day when it should be doing 100, and its not a wattmeter issue because ive checked it on two different meters and both agree. the ALC looks totally normal on the scope display which is what's really throwing me off, you'd expect to see that pegged if something was clipping it back but it's just sitting there looking fine.

i already swapped the coax, different antenna, checked the power supply voltage which is rock solid at 13.8 under load. no fan errors, no error codes, runs cool. TX on all bands is affected, not just one, which kind of rules out a bandpass filter issue i think. audio sounds clean on the modulation monitor, just... low power.

starting to wonder if its a finals issue but i dont really want to crack it open until i have a better idea of what im looking for. anyone dealt with this specific symptom on the 7300 or similar Icom stuff?

yeah that ALC being normal while power is low is the weird part. on most of the Icom stuff i've worked on, if a final transistor is partially failing you can sometimes still get relatively clean output just at reduced level, the ALC doesn't always misbehave until the device is really cooked. the 7300 uses a pair of RD70HVF1 finals if I remember right, might be worth pulling the board and doing a static check on them if you're comfortable with that. also check the bias trimmer — there's a VR on the PA board and if that's drifted it can definitely cause reduced output without throwing obvious symptoms.

before you go that deep though, have you tried the internal ATU bypassed and confirmed it's not somehow introducing loss? long shot but worth ruling out.

not a 7300 specifically but i had basically this same situation with an older 746 Pro a few years back. turned out one of the two finals was completely dead and the other was limping along at reduced efficiency. output was low and inconsistent, ALC looked okay, no smoke no drama. radio just quietly gave up half its power and kept on trucking. point being dont assume both finals are okay just because theres no obvious failure — one dead one can hide behind the other working one depending on how theyre configured.