RECEIVER RCA BC-348-R
RESTORATION (March 2007 – October 2009)
(DIARY)
Restored unit Belmont BC-348-R built 1943 adainst order 31414-WF-43 and with serial number 612.Will be restored to its original state except for the dynamotor.
1. Initial conditionThe receiver had been repainted and was a bit too shiny in some spots, but it was very good inside, almost like new! It has the older type of tubes with the top connectors and not the later "single ended" design. The dynamotor had been replaced by a 220 V power supply, which I find to be an acceptable change for practical reasons! No other modifications were found, but the rotary switch for AVC-OFF-MVC was faulty and replaced with a DPDT switch. The output tube was a 6Y6G instead of a 6K6.
As usual the power supply was badly designed, so the first action was to redesign it. It had no safety grounding and no fuses (when will radio amateurs learn about safety???), and when tested with a dummy load of 70 mA it gave 350!!! V out, the poor receiver is designed for around 220 V. The filtering was very bad, no choke or resistor, only a condenser of 16 uF! The transformer was fine and it fed a nice, classical 5Y3GT. The filament chain had been rewired for 6,3 V instead of the original 24 V and worked well.
Restoration
I will continue reporting on the restoration here on my site.
Report 1 March 12th: The power supply and the AVC-OFF-MVC switch
The first action was of course to fix the power supply. Safety ground and fuses were added and the filtering was improved by adding two power resistors and one more capacitor to form double RC chains. This also decreased the voltage to 220 V at about 50 mA. I will later add some dummy load to keep the voltage down until the filaments are warm, probably a zener diode arrangement. The blown capacitor was removed, I found out it had been designed for the 24 V dynamotor input but was now connected across the 220 V HT, no wonder it had blown!!
I made a quick listening check while monitoring HT voltage and current, the receiver worked!
I heard some BC stations in the 49 m band in a short test. The anode current was a bit too high 70 - 80 mA, and I found it
to be 40-50 mA even without the valves. I will now check all the components, probably I will have to replace some leaky
capacitors.A correct type of three way rotary switch for the AVC-OFF-MVC function was substituted for the DPDT switch repair, see photo. For safety reasons the mains was not routed over this as the 24 V was in the original, but a real mains switch, rotary, is inserted instead of the DIAL LIGHTS control.
The receiver is now very close to its original state and I will now check the components and then the function and alignment. The exploded capacitor has blown some wax over the interior so some cleaning is needed. When the receiver is working I will repaint the panel and the case with the right shade of wrinkle paint.
Report 2 March 14th: More on the power supply, resistance and voltage measurements
The power supply gives too high voltage, 350 V, during some 10 seconds until the filaments are hot. Will be cured with a zener diode chain of 4x56 V, 5W each diode, across the HT, this will load the output with 30 - 40 mA as long as the voltage tries to increase above 240 V, but will not draw any current when the valves are warm and the voltage is around 220 V.
Final power supply schematic.
The resistance between +HT and chassis was measured to be 4,6 kohm, a much too low value and responsible for the high current. By breaking up the anode chain in some strategic places I soon found that capacitor 11-6 had a short, grounding resistor 57-3/first IF. As the resistor is 4,7 kohm this was the cause of the low resistance between +HT and chassis. The capacitor was changed to a vintage Aerovox 0,05 uF 400 VDCW. The resistance to chassis was now 27 kohm which is quite resonable. How the receiver could give any response with such a fault is a riddle, it should be quite dead really!
The HT current now was 45 mA at 215 - 225 V and the receiver was quite lively on all bands. Now I checked all the resistances and voltages at the valve pins and compared with the tables in the manual. Both measurements showed a couple of serious faults: The screen grid resistance and voltage on the mixer and also on the first IF was 0! The screen grid voltages on the first and second RF stages as well as on the second and third IF was only 50 V instead of more then 80 V. The fault was the decoupling capacitor of the mixer screen grid 11-5 which was shorted. I changed it to an Aerovox 0,05 uF 250 VDCW. The screen grid resistor on IF 3 was 480 kohm instead of 180 kohm, it was changed.
Report 3 March 15th: Change of output valve, resoldering of the heater chain
A batch of NOS spare valves were bought at Radio Daze in New York, www.radiodaze.com, and on eBay. They cost only $4 as an average and a few dollars postage. The valve 6Y6G was changed to the original 6K6(GT) which takes only 0,3 A LT current compared to 1,4 A for 6Y6. The dynamotor fed a negativ grid voltage to the valve in the original circuit, this has to be substituted with a cathode voltage. As I do not need much power I decided to use the lowest voltages in the data sheet, 100 V anode and screen grid voltage and 10 V on the cathode, current 10 mA. I got this through a 10 kohm 2W resistor for the anode/screen grid and a 1 kohm resistor decoupled with 100 uF for the cathode. The entire receiver now took only 32 mA at 220 V!
The filament chain was badly soldered when changed from 24 to 6,3 V, I changed some wires and went over the solder joints.
Listening tests gave a very good result on the 49 and 31 m bands. The AVC voltage varied from 0 to -10 V with the signal strength and the MVC was 33 to 0 V with the volume control, very good. The BFO and crystal filter worked fine. The receiver is now in place in the collection and I will make listening tests a week or two. The next steps will then be to plot the IF curve and measure and perhaps trim the sensitivity on all bands.
Report 4 April 4th: Testing and fine alignment of the IF
A 1 mV ptp signal from a signal generator tuned to the receiver intermediate frequency 915 kHz was connected to the mixer grid with a 100 nF capacitor. The signal was modulated to 30 % with 400 Hz. The frequency was varied and the output was measured at the headphone output.
First the frequency of the crystal filter was checked to be 915.00 kHz. All IF circuits were then aligned to that frequency, the output increased nearly 20%.
With the crystal filter OUT the bandwidth was measured to around 5.6 kHz (+- 2.8 kHz) @ - 3 dB and around 14 kHz (+- 7 kHz) @ - 20 dB.
With the crystal filter IN the bandwidth was 400 Hz (+- 200 Hz) @ - 3 dB and 3 kHz (+- 1.5 kHz) @ - 20 dB (specified to +- 2 kHz).
With 1 mV ptp input the output was 12 V ptp, which is 1 mW in 20 kohm, the amplification in the mixer/IF/AF stages being around 12000.
Report 5 March 28th: Check and alignment of the frequency accuracy (the oscillator alignment)
A signal generator and counter was connected to the aerial input over a 100 pF capacitor as dummy antenna. The frequency was measured at the dial end points on all bands. All the four upper bands 3 – 6 are better then the specified tolerance (0.5 %). Band 1 has 1.4 % fault at 500 kHz, will be aligned, and band 2 is very inaccurate (up to 5 %!), will be examined and corrected later.
My BC-348-R in its collector environment! It remains to repair band 2 and to repaint the front panel.
Report 6 October 1st 2009: Repainting the front panel
The front panel was removed and the text masked with tape. The outer layer of paint was removed with a chemical paint remover,
applied in very thin coats. The underlying wrinkle paint structure was thereby saved. Finally the panel was sprayed with
two thin coats of mat black paint.
The plastic identification plate was original but in bad shape with paint and pieces missing. The paint on the back of the plastic sheet was retouched and the plastic sheet was then glued to a thin white aluminum plate. Missing pieces were replaced. The identification plate now looked fairly good and was riveted in place.
Two service stamps to the left of the plate were taped during the paint process and thus saved.
My restored BC-348-R receiver
More on WWII receivers
Top of page
Home
Updated October 23rd 2009.