The 1927 Silver-Marshall "Improved Shielded Laboratory Model Receiver" superheterodyne radio is a well-designed and well-crafted receiver that performs with excellent reception and great audio sound.  The front end of the set's circuit consists of a separately tuned oscillator and detector circuit whose tuning dials turn in opposite directions of each other when tuning in a station.  Both circuits incorporate a regenerative circuit which is essential for the oscillator but not for the detector.  Regeneration in the detector is adjustable and is used as an aid in tuning in distant stations.  The plug in coil that's used in both the oscillator circuit and in the detector circuit is the same in design and consists of three sections of winding with the third section (feedback winding) being hinged and adjustable.  The I.F. section, of the receiver, consists of an enclosed brass box assembly which houses the 1st., 2nd., and 3rd. IF amplifier stages and the 2nd. detector stage.  Each I.F. coil is wound on a large wooden dowel and coated with paraffin to protect against moisture.  The audio output consists of two, transformer coupled, audio stages with a 71A amplifier tube driving the speaker.  The front panel is a sheet of metal, in which the wood grain and central artistic design are printed on, and the steel chassis is painted flat brown.  I acquired this receiver, in November of 2005, from fellow collector Phil Squire through an abay auction.

Tube Line Up:
01A...1st. Det.
01A...Oscillator
01A...1st. I.F. Amplifier
01A...2nd. I.F. Amplifier
01A...3rd. I.F. Amplifier
01A...2nd. Detector
01A...1st Audio
01A or 12A or 71A...Audio Output

Power Source:
Battery...+6 Volts
Battery...+45 Volts
Battery...+90 Volts
Battery...+135 Volts
Battery...-4.5 Volts
Battery...-22.5 Volts

Frequency Range:
BC...550 kHz to 1400 kHz
I.F.Freq...112 kHz

Detentions:
Height...10.5 inches
Width...29 inches
Depth...15.5 inches

Click Here to view the factory building on Terraserver.  The factory is the building just above the red dot.

    The schematics and Information for the model 440 I.F. amplifier unit can be found at the Nostalgia Air web site by clicking on the link below.

Click to this website to a see a mid 1920s Silver-Marshall radio catalog which shows a photograph of the factory building.

Schematic and information

    The dial to the left controls the detector, the dial to the right controls the oscillator, the Gain control regulates the -4.5 volts supplied to the three I.F. amplifiers, the Sensitive control regulates the detector's regeneration and the Filament control regulates all of the tubes filament voltages.

    Silver-Marshall could not identify the model 440 as having an I.F. circuit which could be used in a superheterodyne receiver.  To do so would be in violation of RCA's superheterodyne patents.  To legally get around this, Silver-Marshall described the 440's circuit as a three stage, long wave, amplifier and detector that was specifically designed to receive station NAA, on 112 K.C., when the 440 was hooked up as a fixed frequency, TRF, receiver.  Most customers that purchased and used the 440 were well aware of it's true purpose.

    Here's a photo, featured on page 1020 of Radio News, March 1928, of a young girl tuning an "Improved Shielded Laboratory Model" receiver.  Upon close examination, it's shown that the radio cabinet used is too tall for the radio and a one inch gap exists between the front panel's decorative upper bar and the lid.

    The overall performance for this radio is excellent, a real DXer, and the local stations sound as good as though they were being received by a 1930s AC radio.  I've been able to achieve this caliber of performance only by doing a complete electrical restoration on this radio.  The first thing I did was to replace the original gain control.  There was an open in the windings of the 200 ohm filament rheostat resulting in partial control of the sets gain.  After I replaced it, I examined the detector and oscillator stages, which seemed to be in good shape, but I chose to go ahead and clean the tube socket contacts, resolder solder joints and tighten a few bolted connectors.  The I.F. stages were another story.  When I first powered up the receiver, one I.F. tube had an intermittent filament connection that caused gain to drop at times.  The radio would also go into a locked, runaway, oscillation whenever the 1st detector's regeneration was run high or if the set was physically bumped or tapped.  The cause of this problem was found when I carefully opened up the "442" I.F. box and found that the fixed capacitors on three of the four I.F. coils had cracked solder joints due to the type of solder that was used and age.  My 35 watt soldering iron wasn't sufficient enough to solder the connections because the mica capacitors, with their metal edges, distributed the heat away, too quickly, from the solder joints for the solder to melt.  I resorted to using my micro butane torch to make the required solder repairs, which it accomplished perfectly.  After I put the cover back on and mounted I.F. assembly back onto the receiver, normal performance was achieved.  After I ran the set for some time, I determined that it still didn't quite perform as well as other superheats in my collection.
   After carefully studying the circuit diagram at some length time, I determined that there were two original design flaws that prevented this radio from ever achieving its full potential in performance.  The first flaw was in the circuit of the 2nd detector. It used a negative grid bias setup and Silver-Marshall chose this for two reasons.  First, there wasn't much room to install a grid capacitor and resistor in the 2nd. detectors small chamber, and second, since the I.F. assembly was designed to be permanently sealed in a copper box, maintenance for the grid's capacitor and resistor was near impossible.  Using the negative bias circuit required less components and no maintenance.  Even though the 2nd. detector was sufficiently sensitive in it's original design; the audio quality was very poor.  I reopened the I.F. assembly and installed a .002 uF silver mica capacitor with a 1 Meg. metal film resistor into the grid circuit and, since I had the assembly opened, took to the liberty of replacing two 0.05 uF bypass capacitors, a 0.003 uF detector plate bypass capacitor, and I resoldered all of the solder joints.  Once the I.F. assembly was put back together and installed, I grounded the -4.5 volt wire that connected to the secondary winding of the 4^th I.F. coil.  When I had the set powered up and playing, there was a tremendous improvement in the sound quality.
   The other design flaw I corrected was in the wiring of the filament control's circuit.  All of the tube filaments were controlled by the filament control, which worked ok, but every time the control was adjusted, the change of filament voltage would cause the detector and oscillator to slightly shift their frequencies.  This would require the retuning of the radio every time the gain was adjusted.  Also, when the filament control was adjusted down because of a strong local station, the audio output would sound poor and distorted.  To fix this problem, I mounted a 1920s filament rheostat underneath the chassis, using one of the existing chassis bolts to secure it, and I rewired the 1st. detector, oscillator, and two audio output tube filament lines to the rheostat.  I adjusted the rheostat so that with 6 volts supplying the filament voltage, the selected tubes would receive 4.75 volts each, which proved very satisfactory in each of the tubes performance.  At this point, only the three I.F. amplifier tubes and 2nd. detector tube were adjustable.  With this setup, the receiver now performs exactly as one would expect it should.  It has complete control in the 1st. detector's regeneration control, full control in the filament control with out any negative side effects and full quality audio output.  The set is now ready for serious competition in an antique radio contest and in general home use.