12-27-2017, 04:31 PM
This past Summer, I rescued a 42-380 from a neighbor's garage where it had been stored for several decades. The elements were not kind to the radio, but I thought it would be a fun project. Not being much for woodwork, I paid a pretty sum to have the woodwork restored and took the radio to a local repairman to restore the radio. Though I have a decent electronics lab in my basement, I've been busy with work and figured someone with experience repairing these radios would be able to do a better job than I could.
Several months later, the radio was finished. A quick check sounded fine at the repair store, so I paid the man and took it home. After setting it up in my living room, I started playing with it. Sure enough, there was an annoying 120 Hz hum present throughout the AM broadcast band, especially bad on the lower end. The hum is present with or without a station being tuned, although it is more noticeable when some stations are tuned than others. I contacted the repairman and he was clueless about the cause. Given the hassle of transporting the radio back to his repair shop again, I decided I would attempt the repair myself.
The first thing I did was go through the full alignment procedure specified in the Philco 1942 RMS Year Book. It was clear my repairman had not performed this alignment given how far off some of the controls were. When I finished, the dial was considerably more accurate and the sensitivity was also noticeably improved, but unfortunately, the 120 Hz sum was still present.
Looking through the schematic, I note the only place where 120 Hz is generated is the rectifier circuit consisting of a 6X5G tube connected as a full-wave bridge rectifier. The output is low-pass filtered by a PI-filter consisting of C61, L59, and C20A. C61 was a 475V 18uF electrolytic capacitor. L59 is the field coil in the speaker. C20A was a 475V 8uF electrolytic capacitor. A look inside the radio revealed that the electrolytic capacitors have been replaced as follows:
C61: 22uF at 450V
C20A: 22uF at 450V
I also measured the inductance of the field coil at 10.8 H at 120 Hz with a series resistance of 6.5 kOhms. The schematic says it should be closer to 1.1 kOhms which means I need to have a closer look at the speaker to see what could explain this large discrepancy.
Closer inspection of C20 and C20A revealed that one of the capacitors was dented due to the overzealous tightening of a retainer clip on the part of my repairman. I decided to remove all electrolytic capacitors and test them further:
C20: 21.2 uF, ESR of 2.48 Ohms, leakage current of 0.45 mA at 400 Volts.
C20A: 22.1 uF, ESR of 1.96 Ohms, leakage current of 0.55 mA at 400 Volts (this one is dented).
C61: 21.6 uF, ESR of 2.3 Ohms, leakage current of 0.32 mA at 400 Volts.
According to the Nichicon specification, I should have seen leakage currents of under about 0.9 mA, so I guess those values are within specification. The ESR values seem high to me, though I'm accustomed to lower voltage applications so perhaps they are reasonable given the ~200 Volts they have across them during normal operation.
Next, I did some searching on the forum to see if I could find any other ideas to try. I found a note about R60 -- a 15-30-146 Ohm compound resistor used for biasing. Sure enough it had already been replaced by my repair guy and was reading correctly when I checked it. I spot checked a few other original resistors for correctness and found them to be within 20% of their rated value, drifting high.
The wiring inside the chassis looks very questionable from a modern perspective where we work with fast-changing signals, but I'm assuming outside of avoiding routing the power lines close to any RF input stages, it should be fine. Some of the original wiring was replaced by my repair guy and I can only assume he routed it in the same manner as he found it. I'm a bit worried about the placement and wiring of C20 given that it appears to provide additional power supply filtering to the IF stages and is placed next to the power supply filter capacitors.
I'm getting ready to hook up an oscilloscope to the radio and start probing around to see if I can find anything. I'll use two probes in A-B mode in order to avoid the floating ground issue that would otherwise be a problem. Meanwhile, I'd certainly appreciate any advice or suggestions anyone here can provide as I've not worked on any electronics of this vintage before and have probably neglected to try many things which would be obvious to those who have.
--
Jason
Several months later, the radio was finished. A quick check sounded fine at the repair store, so I paid the man and took it home. After setting it up in my living room, I started playing with it. Sure enough, there was an annoying 120 Hz hum present throughout the AM broadcast band, especially bad on the lower end. The hum is present with or without a station being tuned, although it is more noticeable when some stations are tuned than others. I contacted the repairman and he was clueless about the cause. Given the hassle of transporting the radio back to his repair shop again, I decided I would attempt the repair myself.
The first thing I did was go through the full alignment procedure specified in the Philco 1942 RMS Year Book. It was clear my repairman had not performed this alignment given how far off some of the controls were. When I finished, the dial was considerably more accurate and the sensitivity was also noticeably improved, but unfortunately, the 120 Hz sum was still present.
Looking through the schematic, I note the only place where 120 Hz is generated is the rectifier circuit consisting of a 6X5G tube connected as a full-wave bridge rectifier. The output is low-pass filtered by a PI-filter consisting of C61, L59, and C20A. C61 was a 475V 18uF electrolytic capacitor. L59 is the field coil in the speaker. C20A was a 475V 8uF electrolytic capacitor. A look inside the radio revealed that the electrolytic capacitors have been replaced as follows:
C61: 22uF at 450V
C20A: 22uF at 450V
I also measured the inductance of the field coil at 10.8 H at 120 Hz with a series resistance of 6.5 kOhms. The schematic says it should be closer to 1.1 kOhms which means I need to have a closer look at the speaker to see what could explain this large discrepancy.
Closer inspection of C20 and C20A revealed that one of the capacitors was dented due to the overzealous tightening of a retainer clip on the part of my repairman. I decided to remove all electrolytic capacitors and test them further:
C20: 21.2 uF, ESR of 2.48 Ohms, leakage current of 0.45 mA at 400 Volts.
C20A: 22.1 uF, ESR of 1.96 Ohms, leakage current of 0.55 mA at 400 Volts (this one is dented).
C61: 21.6 uF, ESR of 2.3 Ohms, leakage current of 0.32 mA at 400 Volts.
According to the Nichicon specification, I should have seen leakage currents of under about 0.9 mA, so I guess those values are within specification. The ESR values seem high to me, though I'm accustomed to lower voltage applications so perhaps they are reasonable given the ~200 Volts they have across them during normal operation.
Next, I did some searching on the forum to see if I could find any other ideas to try. I found a note about R60 -- a 15-30-146 Ohm compound resistor used for biasing. Sure enough it had already been replaced by my repair guy and was reading correctly when I checked it. I spot checked a few other original resistors for correctness and found them to be within 20% of their rated value, drifting high.
The wiring inside the chassis looks very questionable from a modern perspective where we work with fast-changing signals, but I'm assuming outside of avoiding routing the power lines close to any RF input stages, it should be fine. Some of the original wiring was replaced by my repair guy and I can only assume he routed it in the same manner as he found it. I'm a bit worried about the placement and wiring of C20 given that it appears to provide additional power supply filtering to the IF stages and is placed next to the power supply filter capacitors.
I'm getting ready to hook up an oscilloscope to the radio and start probing around to see if I can find anything. I'll use two probes in A-B mode in order to avoid the floating ground issue that would otherwise be a problem. Meanwhile, I'd certainly appreciate any advice or suggestions anyone here can provide as I've not worked on any electronics of this vintage before and have probably neglected to try many things which would be obvious to those who have.
--
Jason
