Interesting. In some ways it could be drift in the AC line since your power is 230VAC at 50Hz. If there are any voltage swells which do occur that may be it. Doubt it. though. Just a thought. If that is the case a step-down transformer may not be the ideal. There are regulated power supplies you can get that will take that 230VAC @ 50Hz and convert it to a 115/120 VAC at 60 Hz. They do exist as we use that type of equipment for testing products for other countries other than USA.
What the issue is that the volume slowly increases over time, that is odd. Two things come to mind what could cause this other than the input power.
1. You already have a compromised preamp tube, could be in V3 (tone stack driver but only affects CH3 and CH4). V5 is used with the FX loop. Both of these positions have a cathode follower circuit. That is not easy on some preamp tubes. Volume swells is the symptom of this when a tube is not suitable in those two locations. Normally the swell is in the opposite direction, it drops. However, if you dialed in the amp to compensate for the issue, chance that it may increase in volume is possible.
2. The send level control is moving on you for some odd reason. I doubt vibrations would result in its movement. Could probably rule that out unless it has a small hand or big hand changing it when you are not noticing.
Are you using some sort of volume pedal? That I can see being a cause if you are not setting it at max travel. Some of those can self-adjust or move to due to the weight of the foot lever.
Just shot-gunning suggestions as this is not normal behavior. The only other thing that comes to mind on a rising volume is when the power tubes bias begins to drift. Since that is derived from the power transformer, it does not have a well regulated supply. So, if the 50Hz line frequency is too slow to maintain sufficient voltage on that circuit, it will drift and could become lower in magnitude (more positive) which will shift the bias on the power tubes making them louder until you reach the critical voltage when they red-plate and blow the fuse. The bias circuit does float to some degree based on loading of the power transformer. How quick does it recover as it is a half wave rectified circuit is hard to say.
