Spent the evening working on the first of the two 191s. The case is beat up but everything inside looked okay after I cleaned out the spider webs. I worked through the calibration procedure and it is well within specification now from .75 to 100 MHz. The two regulated voltage supplies didn't need to be adjusted, both being within one percent of spec.
The only tedious part of calibration is tuning seven LC networks. I have to say it is a lot easier to do using a frequency counter instead of following the procedure described in the manual, which involves zero beating the 191 output against a time marker generator.
The amplitude was constant within a couple of percent across the band, though I think my measurement scheme was a little suspect. I also put it on the spectrum analyzer. The 191 does not, as one would expect, generate a pure sine wave. There are three harmonics down 60, 70, and 80 dB. For a design based on a series of LC tanks, it works pretty damn well.
It is late here, so I will post photos tomorrow after I clean up the case, blow the rest of the dust out of the interior, and let it burn in for a couple of hours.