I got kind of busy last week. ?In replacing R3 in the AC-4, I decided to isolate C3 and C4 from the circuit and use a DC PS to condition both of these electrolytic capacitors, starting with a resistance measurement with the VTVM (1.5 VDC); individually, within 30 minutes they were up to 50 Mohm. ?Then overnight, they were up to 150 Mohm and climbing. ?I noticed C4b would drop a little to 75 Mohm then increase back to 150 Mohm; I interpret this as a clump of non-ionized electrolyte being uncovered and taking ?time to ionize the solution with a consistent open dielectric gap; nice that all of that is automatic, just applying a potential difference with the correct polarity. ?C4b had not been conditioned otherwise, all others had been. ?This AC-4 appeared to be completely factory with their practice of laying a component lead or wire against a terminal, then applying A LOT of solder; no good mechanical joint was created before soldering. ?Solder must have been really cheap then. ?The good part was it was really easy to unsolder all the connections. ?C4 was found to be a shiny aluminum can with C4b a 100 uF in lieu of 80 uF. ?The rest of the electrolytics were covered with canary yellow cardboard: all Mallory. ?C5 a and b were found to be 40 uF in lieu of 20 uF.
Leakage current measured for C3 and C4 follows.
C4a: 3.4 uADC constant. ?Capacitance measured 122.5 uF.
C4b: 4.1-4.4 uADC hopping around. ?Capacitance measured 124.2 uF.
C3: 3.4 uADC. ?Capacitance measured 121.8 uF.
With these caps about 20% greater than nominal capacitance ?(nothing surprising), the average DC voltage rises a little higher due to longer RC time constant of the RC filters.
Ambient temperature for slow conditioning up to rated WVDC was 78 deg F; the base of the can was a few degrees higher (handy IR thermometer).
With input voltage set to 120 VRMS, the following UNLOADED outputs were as follows.
650 VDC nominal, 720 VDC measured.
250 VDC nominal, 295 VDC measured.
BIAS -46.5 to -100 VDC, steady.
Filament voltage measured 14.87 VRMS.
I expect all these voltages will decrease much closer to nominal when loaded with the transceiver or transmitter. ?I also expect a significant rise in input line voltage will produce a significant rise in output voltages, loaded or unloaded. ?Probably the best thing to do is use an auto-transformer with the AC-4, as these are clearly unregulated power supplies. ?
I¡¯m cleaning up and repairing a DC-4. ?The input voltage must be 12.6 VDC, as 13.8 VDC has shown to produce much higher than desired output voltages, with little to no sagging of 12.6 VDC filament voltage. ?Perhaps an AC-DC switching power supply, and the DC-DC power supply (DC-4) is a better solution in order to regulate the input voltages to these radios. ?Using late technology, modest priced good quality and noise-free switchers that adjust to 12.6 volts DC are available. ?I¡¯m going to investigate this scheme. ?There are many attributes to using this power supply scheme.
One other note: A sign of a good functioning electrolytic capacitor is observed after conditioning, or regular use. ?With the capacitor isolated from circuit, discharge the cap with a homemade shorting probe and approximately 10 kohm resistor in series (keeps the v= Ldi/dt air breakdown sparks from occurring). ?Count to 20 connected to the cap. Have your voltmeter with lower voltage scale ready and attach it to the leads immediately after removing the shorting probe. ?The DC ?voltage should immediately and continually rise from zero on its own to some value being a reasonable percentage of the rated working voltage. ?A fast rate of immediate rise indicates you should EXPECT it to be in good operating condition.
73,
Michael, N4KZO
