Advantest R4131A Spectrum Analyser - bad power supply caps
Recently, I bought an Advantest R4131A spectrum analyser (at a good price) from the Philippines from a seller of automation surplus with the fault that this unit would not power on.
I once thought that something in the power supply was shorted, but upon opening this unit, I found nearly all of the electrolytic capacitors (Nichicon PF and VT series) were bulging and/or leaking
and they were all replaced, along with reparing a corroded lead on the transformer coupling the PWM pulses to the 2SK794 primary switcher with the 150R (+15V) and 220R (-15V) load resistors being replaced with 5W units along with the PCB traces near the scorched part being repaired accordingly.
Fortunately, I was aware of bad batches of certain series of Nichicon, Rubycon, and possibly Panasonic and Chemi-Con units from the late-80s to the early-90s, which was around the time the spectrum analyser was made (1988).
When testing this unit with the fan connected and a 3.1A load on the 5V rail, there was no regulation of the -15V rail (it was a bit high
), despite adjusting RV3 through its range; so this pot as well as VR2 (+5V) and VR4 (+15V) were all replaced to no avail.
A zener diode with its cathode connected to the Emitter of Q1 (2SC2655) was found to be shorted (I found the good zener on the +15V rail was rated at 6.2V) and both of them were replaced, but the -15V regulation problem remains.
I checked the voltages on Q1 while varying RV3 through its range; the Base voltage varied but there was no voltage variation on the Collector and Emitter.
The penny dropped when I removed Q1 and tested it in a semiconductor analyser which indicated that no component was detected
, and after replacing this transistor (which became open circuited), regulation for the -15V rail was restored.
RV2 was adjusted for +5V with a 3.1A load, along with RV3 and RV4 being adjusted for -15V/+15V respectively, as well as fine adjustment of the +15V/-15V voltage rails after the power supply was connected to the spectrum analyser boards which were all connected correctly (and the +5V rail did have a good regulation characteristic).
Finally as a good measure, I replaced all of the electrolytic capacitors in the monitor assembly with the Horizontal yoke coupling capacitor being replaced with a polyester unit, along with minor adjustments to the display controls.
At first, it was a bit of a gamble, but it did pay off
.
I once thought that something in the power supply was shorted, but upon opening this unit, I found nearly all of the electrolytic capacitors (Nichicon PF and VT series) were bulging and/or leaking
and they were all replaced, along with reparing a corroded lead on the transformer coupling the PWM pulses to the 2SK794 primary switcher with the 150R (+15V) and 220R (-15V) load resistors being replaced with 5W units along with the PCB traces near the scorched part being repaired accordingly.Fortunately, I was aware of bad batches of certain series of Nichicon, Rubycon, and possibly Panasonic and Chemi-Con units from the late-80s to the early-90s, which was around the time the spectrum analyser was made (1988).
When testing this unit with the fan connected and a 3.1A load on the 5V rail, there was no regulation of the -15V rail (it was a bit high
), despite adjusting RV3 through its range; so this pot as well as VR2 (+5V) and VR4 (+15V) were all replaced to no avail.A zener diode with its cathode connected to the Emitter of Q1 (2SC2655) was found to be shorted (I found the good zener on the +15V rail was rated at 6.2V) and both of them were replaced, but the -15V regulation problem remains.
I checked the voltages on Q1 while varying RV3 through its range; the Base voltage varied but there was no voltage variation on the Collector and Emitter.
The penny dropped when I removed Q1 and tested it in a semiconductor analyser which indicated that no component was detected
, and after replacing this transistor (which became open circuited), regulation for the -15V rail was restored.RV2 was adjusted for +5V with a 3.1A load, along with RV3 and RV4 being adjusted for -15V/+15V respectively, as well as fine adjustment of the +15V/-15V voltage rails after the power supply was connected to the spectrum analyser boards which were all connected correctly (and the +5V rail did have a good regulation characteristic).
Finally as a good measure, I replaced all of the electrolytic capacitors in the monitor assembly with the Horizontal yoke coupling capacitor being replaced with a polyester unit, along with minor adjustments to the display controls.
At first, it was a bit of a gamble, but it did pay off
.
