Columbia C301

 This old family set from the early 1950s arrived in late June 2021 for an overhaul.

 

 

 
 

 

 Click the label to see the service details of the identical chassis used in the Marconi T32A. The label has a "Sold" and customers name "Pugh" with what looks like a house name which by its shortened form ".. urndy" must be local to the radio shop.

The set was probably made around 1952 and looks complete, very dusty and with a broken dial cord. Not too much chassis rusting and apparently (from their condition) original valves, RF amp/Frequency changer X150, IF amp W150, Det/AVC/Audio amp DH150, Audio output and U150 rectifier.

The datasheet tells me the dial cord is 58.5 inches long and has two turns around the tuning spindle and nothing else.

The 8 inch speaker is low impedance and handles the output of a decent 4 watts. Apart from a little missing veneeer which can be fixed, the case is quite presentable just needing a good clean and polish.

Of course there are likely to be a few condensers in poor condition, maybe a low emission valve or two and perhaps a new mains lead needed, but first I decided to swap its old condensers which proved to be miles too high in value indicating to me that all would be leaky.

 
 

 

 Condenser

C15

C27

C17

C28

C26

C25

C19

C21

 C22 ***

 TEST

 Marked

 0.1uF

0.005uF

0.05uF

0.1uF

0.01uF

0.25uF

0.01uF

0.25uF

 0.01uF

 0.15uF

 Measured

 393nF

 11nF

 169nF

364nF

177nF

515nF

40nF

819nF

 18.7nF

 151.5nF

 High by

 393%

 220%

 338%

 364%

 1770%

 206%

 400%

 327%

87%

 1%

 Re-measure

 160nF

 30nF

 76nF

 159nF

 2nF**

 353nF

 1.5nF**

 393nF

 51.9nF

 175nF

 Voltage

 74V

 15V*

 77V

 108V

 50V*

 45V

 165V

 145V

 31.7V

 1mV

 Leakage

 0.62mA

 0.13mA

 0.64mA

 0.9mA

 0.42mA

 0.38mA
 1.34mA

 1.21mA

 0.26mA

 8pA

Equiv 

 366K

 2.3M

 347K

 213K

 600K

 689K

 98K

 128K

 1M

 36G

 Key: ...* figure increasing by 1 volt per 5 secs; ** reading failed with max figure cycling from 0, *** Hunts plastic type

"Equiv" equates to a resistor wired across the condenser. In most cases, with reference to the circuit below, because of the function of the condenser, the leakage isn't important. C17 is critical and C26 also, particularly as the latter showed increasing leakage with time.

 

The marked values are as shown with the measured values in nF where 0.1uF=100nF. All were marked 350 volt working and all bar C22 were wax covered types made by Hunts. All are referenced by the Marconiphone T32A chassis drawing below. The 1st check was using a digital testmeter, 2nd check I used an old auto-ranging capacitor tester and for the leakage test I applied 300V across the condenser and a 120Kohm resistor in series, measuring the voltage across the resistor with a digital voltmeter. The higher the resistor voltage the higher the leakage current. In one case, that of C22 I additionally used a Peak LCR tester which gave me a reading of 17.48nF.

The column marked TEST shows test results for a new 150nF +/-10% plastic capacitor which read 175nF on the 2nd tester but 151.1nF on the 1st tester.

My guess is a faulty capacitor or two in the 2nd tester although this is excellent for reading small capacitors such as 33pF.
 

 After replacing the old condensers I spotted another Hunts plastic one across the tone control switch which is clearly C24 but this isn't critical. Checking all the resistors in-situ was the next step and during this exercise I spotted a short across R20, 150 ohms. Unfortunately this turned out to be a heater-cathode short in the N120 output valve. As you can see below, and checking their function in the circuit, none of the resistors really needs changing. A couple, R6 & R11 cannot be checked properly because they are in parallel with other resistors. The bad N150 is a loctal B8A having a rated output of over 4W, with an EL41 equivalent, which according to my listing I have several... but before powering the chassis I connected a variable HT supply to the pair of main HT condensers and slowly increased the voltage whilst monitoring the HT current. Once it had dropped to several mA at 300 volts I fitted in a replacement EL41, wired temporary mains connections plus a grounding wire, plugged in and switched on. The loudspeaker, still in its case but wired to the output via leads, came on and with a short wire I could tune medium and long wave broadcasts. Signals were not too strong and the next step will be to set the IF to its correct frequency of 470KHz but, because the dial is fitted to the case I'll need to mark the backing plate in order to align the three wavebands.

 

 Resistor

 R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

 Marked

 33K

22K

1M

47K

 150

10M

27K

2.2M

330K

270

470K

68K

150K

3.3K

10M

680K

 220K

1.2K

1.2K

150

 Reads

 38K

 26K

 1.2M

 52K

 163

 2.5M

 32K

 2.2M

 384K

 311

 396K

 84K

 152K

 3.9K

 12M

 732K

 230K

 1.3K

1.3K

160

 High

 15%

 18%

 20%

 11%

 9%

 -

 19%

 0%

 16%

 9%

 -

 24%

 1.3%

 18%

 20%

 8%

 5%

 8%

 8%

 7%

 

 The service data for the Marconiphone T32 uses the same chassis as the Columbia C301 but this tells one nothing about aligning the receiver other than the IF is 470KHz so one must assume that nothing special needs to be done. The initial check is to look at the IF response. The IF transformer (there's only one) has what appears to have access points at front and back for setting the dust cores. There are three RF trimmers, TC1 for shortwave aerial tuning, TC2 for mediumwave aerial tuning and TC3 for the mediumwave oscillator. All six aerial and oscillator coils have adjustable dust cores meaning, all told that only medium wave alignment can be set up properly, leaving long waves and short waves to their own devices. In fact the only adjustments needed are setting Radio 4 to close to 198KHz on the dial and to average out short wave reception, leaving a decent aerial to deal with reception.

 

 Tuning the receiver was fairly straightforward. I located the trimmers etc and found the IF cores were moveable. Initially the receiver was deaf and the IF responded to about 478KHz but retuned to 470KHz readily. I used my HP signal generator which you can see behind the receiver case but switched to the Wavetek on the right to check long waves because the HP doesn't go much below 450KHz. I marked a paper scale with key frequencies and everything peaked up nicely. The tuning doesn't have a fine tune position unlike some sets so tuning across short waves is a little tricky. I used my audio wattmeter but, because the set was so sensitive, I had to put a handy 36 ohm resistor in series with the speaker to balance the volume with the meter reading. There aren't many broadcasts on medium waves and long waves during the day, but in the evening the former will be jam packed with stations. All that now remains is to finish tidying the case then sort out a decent mains cable or IEC socket then fit the chassis back.

 

 

 

 

 By moving the mains tapping panel (which will never be used) to a convenient space to one side I was able the snugly fit an IEC connector (complete with a mains suppression capacitor and 1M leak) in its place so the the existing hole in the rear cover will now accommodate a standard PC style lead.

Reassembled, the Columbia works really well. The audio from Radio 4 longwave is crystal clear. Medium and short waves are plagued with local noise which fades away nicely as stations are tuned in. 

 

 Return to Reception