One of the very hard to get resistors is off spec :-( It's a 7M 1W 1% that measures 7.35M.
Doesn't seem like much, but it's the first in the range divider network. I put about 136M in parallel to knock it into spec. I had to string together a few 22M and 10M to get the right vallue.
I'm still acting strangely though. I can calibrate it on the 50V range so it's dead on at 10V and 50V, but it's off by a tick or so at 20, 30 and 40. A non-lenear response in other words. All the other resistors seem to be within spec and I tried swapping out the tubes. Maybe the meter is off a little ?
Not a big deal as it's accurate enough for my needs. I'm just curious what might cause this problem. Perhaps one of the ceramic caps is a little leaky ?