R-390A Performance After Rebuilds
There are two methods commonly used for measuring
AM sensitivity. I refer to these as the S+N2/N1
method and the S+N2/N2 method, where N2 is the
receiver noise in the presence of an AM carrier, and N1 is the
receiver noise with no signal present. I learned on the S+N2/N1
method, but have since become a believer in the S+N2/N2
method. The S+N2/N2 method yields higher (poorer)
values of sensitivity, but is more indicative of what the listener hears when
tuned into an AM station. The S+N2/N2 method is the one described in
all except one of the Field Depot manuals, and in that one manual, the method
can be interpreted either way. For a more detailed description, click
here.
Below are the sensitivity charts after
rebuilds. On the first couple, I did not take measurements prior to the
rebuild. The later ones show both before and after values. Note that
the scales are logarithmic, and not all scales are the same. Click on any
chart below for a better view.
All measurements below are based on the S+N2/N2 method
All measurements below are based on the S+N2/N1 method
Capehart (S/N 1543)
Collins (S/N 4174)
Motorola (S/N 6)
Collins (S/N 3198)
Stewart Warner
Imperial (blank S/N)
Rebuilt March 2002
EAC (S/N 3669)
Rebuilt December 2001
This unit had the Capt. Paul Lee product detector mod in the IF deck, but
otherwise was all EAC. A full restoration was completed, including
replacement of all paper capacitors. The Cosmos PTO allowed for easy
linearization of the PTO prior to a full alignment.
Imperial (no S/N)
Rebuilt September 2001
This receiver surprised me with both the remarkable and flat sensitivity
across all bands. This was a standard restoration for a customer.
Work included replacing all paper capacitors with Sprague Orange Drops,
rebuilding one RF and two variable IF cans in the RF deck, and replacing eight
less-than-optimal tubes. Note that I had to change the vertical scale of
the above plot because the numbers dropped below typical results.
EAC (S/N 2063)
Rebuilt June 2001
This one came into the shop with better performance than most. But you
still see some good improvement, especially in the 2-4 MC range. In
addition, the Break-In function was made to work, the remnants of a previous
modification were removed, a problem with the AGC was corrected, the 4KC filter
was replaced, sticking slug racks were made smooth, and intermittent loss of
signal (a bad tube) was repaired. One capacitor (which certainly looked
original) was the wrong installed value, and this was also corrected.
Capehart (S/N 1852)
Rebuilt Spring 2001
This receiver had a few problems. Initial rebuilt performance showed
only minor improvements from the original numbers. Measuring stage gains
traced the problem area to the RF amplifier area. All mica caps and
resistors and tubes checked OK. Reworking the solder connections in this
area seemed to be the final fix. This unit also suffered loss in the IF
deck due to wafer switches not making proper contact.
EAC R-390A (S/N 2017)
Rebuilt: March 2001
This rebuild was done for a customer. This 100% EAC '67 started out as
a checked unit from Fair Radio. Note the tremendous improvement in
performance from a "checked" unit to a fully restored one.
Stewart Warner (S/N 5157)
Rebuilt: Winter 2000/2001
The problem with this one turned out to be a burned coil in T201 (0.5 to 1.0
MC range), and a bad trimmer capacitor (8-16 MC range).
EAC R-390A (S/N 2588)
Rebuilt: October 2000
This was an EAC '67 that I brought home from Dayton (May
2000). A complete set of measurements has not been taken on this
radio. It made it into the hamshack before I completed the full round of
testing, and I have not carried it back up the hill to my workshop (yet).
Collins R-390A
Rebuilt: February 2000
This is my original R-390A. This is the one that got me started.
Measurements shown are from May 2001, over a year after the rebuild. The
initial measurements were a bit better than those shown here, so it might be
time to check the tubes for noise.
Last edited: 04/13/2003
