[NostalgiaRadioTime]

Thanks Mike and Terry for the suggestions. Spent another evening gathering data. I started with measuring the shadow meter. I got 1400 ohms across the coil (reading across the pins with it disconnected). That seems within reason.

The voltage chart on the Philco service bulletin for the 38-4 / 38-5 is sparse. Mostly just plate voltages with some screen grids. Here's what I got: (all readings taken to chassis ground)

6F6 Plates read 274 VDC on each. For the screen grids I get 264 VDC on one and 285 VDC on the other

6K5 Plate reads 174 VDC

6J5 I get ZERO volts on the plate and on the screen grid. Just the 6.3 heater voltage

6K7 I get 267 VDC on the plate and 86 volts on the screen grid

6A8 I get 249 VDC on the plate / 86 VDC on the screen grid / 152 VDC on the OSC plate (as it's called on the voltage chart)

6U7 I get 276 VDC on the plate / 100 VDC on the screen grid

I also get 274 VDC on the primary's of the 1st and 2nd IF's

BUT, I get negative (-) 18 VDC on the negative side of the electrolytics (64, 64A, 66). This is from the center tap of the transformer to the negative common of the 3 electrolytics.

I measured the resistance of the transformer at the socket pins in the transformer for the 5Y4 rectifier and from the center tap of the transformer to the plates of the 5Y4 (transformer secondary windings) I get 106 ohms on one side, and 96 ohms on the other.

Based on this data, what could be killing the voltage to the 6J5 tube?