Yet Another Unidentified Radio
This radio came from the estate of an AT&T engineer.  It's a regenerative detector followed by two transformer-coupled audio stages.  The tubes are type 30's and the five-pin plug-in coil covers 2.5 to 5 MHz.  The coil form has a machined slot for each winding.
Everything about this set is professional. Top quality components, careful assembly, and good sheetmetal suggest it's not homebrew or a one-off.  The filament switch places resistors between the 3-volt A battery and the 2-volt filaments.
The panel is engraved, and the battery wires are carefully identified.  There is a serial number and other cryptic data, but no indication of manufacturer.  Yes, the sideways National dial is original.  There's paint inside it's mounting holes.

Please let me know if you have any information.

Idle Speculation:
 



This radio reminds me a lot of the CMS.  Compare it's physical layout to the CMS transmitter above.  While the CMS receiver has internal tubes and band-switched coils, it uses triode-connected 3S4 tubes, a reasonable equivalent to the 30's.  Both sets have two banana-jack antenna terminals.  Terminal 1 is connected to a primary winding on the coil, and terminal 2 is connected to the detector grid via a gimmick capacitor.  Both receivers are simple regen's followed by two audio stages, so the fact that the two circuits are nearly identical may only be a coincidence.

I consulted with the gentlemen on the Boatanchors and Milsurplus email reflectors, hoping someone could identify this radio.  There's been no ID, but there were a number of useful observations:

A number of people suggested it was a well-done homebrew or even a kit.  There's too many expensive components for a kit, and while it could be homebrew or a factory G-job (Government jobs took priority.   (-:    ),  it looks and feels like a model-shop piece to me.  One person pointed out that the greenish gray paint looks like a phone-company color.

WB4UIV points out,  "The front panel is stamped with metal stamps hit with a hammer.  The lettering is not absolutely straight."   He further suggests that "MN-R3RB-61" might be a Model Number indicating a 3 tube Regen, and "YN-3405-1" might be a date code.  1934 seems a little early.  The resistors seem newer.  Does anyone know the introduction date for the Stancor A-63-C transformers?

Hue Miller's conjectures:  "Actually i think the fellow who said shop-model or shop built was likely correct, no homebrew or kit would have a stamped serial number.

YN= "year number", i.e. date code
MN= "model number" or what the radio was named

Designed to fit in a small case, or else power lead would have come out the back., and also to be transported "knocked down" or with tubes and coil removed, else there would have been a front panel screening tubes and coil from the operator. Otherwise, for not - miniaturized receivers, not having a  front panel is poor building practice, and the builder of this receiver surely
knew that.  Prototype or demonstration model or limited production for what purpose, though?"

I concur with Hue's observations.  With the tubes and coil removed, the radio takes on the "loaf-of-bread" form factor of an agent set.  That's so you can carry it around in the shopping basket.   So, is it possible that this was a developmental set that was the daddy of the CMS?  Even if we knew that for sure, we still don't know the whys and wherefores of the CMS.  I guess the whole thing will remain a riddle inside a mystery wrapped in an enigma.   (With apologies to Winston Churchill.)