[rebrands]

Hi, I’m new to the “Phorum”, seeking some knowledge on my first Philco project.  I thought I’d start here rather than the other place where I spend more time.
 
First some background.  I have restored (re-capped, refinished) about 30 radios in the past eight or nine years.  I recently acquired a 48-482 that I started on about two months ago.  This thing is a beast.  My first action was to check the continuity of the power transformer, all the IF cans and coils, the values of the resistors, and check the tubes.  I found one IF can (T300) with an open coil, the FM1000 with open filament and several out of tolerance resistors.  I located and purchased a T300 and FM1000, installed the T300.  I then did a complete recap of the paper caps (a chore), and electrolytics as well as some out of tolerance resistors.  Push button contacts cleaned with DeOxit.
 
I have powered the radio up and it works well on BC and SW.  On FM there is very faint noise coming in when tuning near to a station.
 
I have run the troubleshooting procedure listed in the Nostalgia provided Rider literature.  Sections one and two checks are all within specs.  In the section three procedure I’ve run into a roadblock, and this is where I’m looking for help.
 
The introduction to section three troubleshooting says an rf AM signal generator needs to produce at least .5 volts output for this procedure.  If that isn’t possible the signal can be injected further upstream.  My vintage Simpson 315 signal generator is capable of only .5 millivolts output.  However, I tried testing with the signal injected at test point D with the .5 mv signal anyway.  I found that there was no response in the speaker at 9.1 mhz input, but did hear a response when the frequency was several khz off the 9.1 mhz frequency.  In fact there was a null position where nothing was heard and a tone heard either side of the null.  This with no modulation of the rf signal.  Voltage measured between test points A and D was 18.9 vdc at the null and 18.8 either side of the null, so no significant change.
 
I then went to steps two and three since it failed step one.  Audio signal injected at test point E and modulated rf signal at test point F yielded loud and clear signals.  Step four procedure yielded -2.5 vdc  (-2.9 vdc called for). 
 
I then used the “less than .5 volt rf” procedure with the same results.
 
So, here are the possibilities I’ve come up with. 
1) A bad FM1000 tube.  I have no way to check it unless somebody can give me the settings to use on a Hickock 6000 tester.
 2) T300 badly out of alignment. 
3) A bad mica cap somewhere – like inside T300.
4) Or, I’ve made a mistake somewhere / have a cold solder joint I've not been able to find. (Most likely?)  All caps were replaced one at a time and dressed as close to original as possible.  Two electrolytics were mounted elsewhere, under-chassis.
 
And one other odd indication: When listening on FM, if the chassis is tapped with something, like a finger, a metallic ringing is heard in the speaker.  This does not occur in BC or SW.   

Sorry if this is a little wordy.  I wanted to be as clear and complete as possible.

So, what are your thoughts?
Ralph