I guess it's my turn to (try to) calibrate my 2440 (and 2465...)
I was thinking about doing the same and using a power supply and whatever else I can throw together hoping for some sort of accuracy. My 3.5 digit Tektronix DM502 and Fluke 77 also need cal but I think they're the most accurate that I have at this point, and I don't have an arbitrary signal generator - I'll have to depend on crystal oscillators for this. And to build something that has the bandwidth to not have its own losses. My power supply is an analog homemade unit, as well.
But anyway it looks like I will have to do additional RC filtering, the noise on my bench power supplies are horrendous. Perhaps this is one of the reasons for a failure? If it needs to be within 2%, the 0.2V range needs to be within 4mV including noise... 4mV is within the noise of my equipment! How much of an RC constant were you using? I'll need to look for low leakage capacitors to ensure I don't have another voltage divider here adding additional error.
Maybe I will need to get better equipment before attempting calibration.
I definitely know my 2465 needs attenuator var linearity calibration, when I turn the var knobs, the trace shifts by a little. Not sure if this is a potentiometer cal or it's an EAROM calibration but suspecting the latter. A Youtube video on calibrating a 2201's analog section sort of sparked some interest. Though concepts are the same, unfortunately the nitty gritty doesn't apply to either of my scopes... I figure I should cal the 2440 first as if I screw up, I can remove the battery, and I'm no worse off than when I started.