Good evening everyone, they gave me to repair an ARC D-52B with a faulty channel, at the output there are 35 Vdc negative.
From a series of tests carried out it has emerged that up to the output of the AM2 analog module everything works (I swapped the driver / final transistor group with the one of channel 1 that works).
In fact, even with the output of the AM2 module disconnected, the output voltage remains negative. I tried replacing the defective Q3 and Q4 transistors (both 2SB555) but without solving. After I checked the hfe of all the transistors of the broken channel but I didn't find big differences between the transistors that the manual indicates as "Matched Set", only by disconnecting one of the 2K7 - 2W resistors the output voltage goes to zero.
I have tried all transistors with both the tester and a dedicated transistor test that supplies the hfe and also with a test circuit with a load bulb and a resistor in base to bring the transistor into conduction, but found no faults.
I think the problem is the difference in gain between the original transistors and the new ones (2SB555). What should be the maximum difference between the hfe of the transistors that the manual defines as "Matched Set"?
What tests could I do?
From a series of tests carried out it has emerged that up to the output of the AM2 analog module everything works (I swapped the driver / final transistor group with the one of channel 1 that works).
In fact, even with the output of the AM2 module disconnected, the output voltage remains negative. I tried replacing the defective Q3 and Q4 transistors (both 2SB555) but without solving. After I checked the hfe of all the transistors of the broken channel but I didn't find big differences between the transistors that the manual indicates as "Matched Set", only by disconnecting one of the 2K7 - 2W resistors the output voltage goes to zero.
I have tried all transistors with both the tester and a dedicated transistor test that supplies the hfe and also with a test circuit with a load bulb and a resistor in base to bring the transistor into conduction, but found no faults.
I think the problem is the difference in gain between the original transistors and the new ones (2SB555). What should be the maximum difference between the hfe of the transistors that the manual defines as "Matched Set"?
What tests could I do?
Can you post the schematic?
Have you checked all the parts (fuses, resistors, diodes) in the output stage, and in the servo?
There aren't very many. Also the 1N914s in the clipping circuit.
Have you checked all the parts (fuses, resistors, diodes) in the output stage, and in the servo?
There aren't very many. Also the 1N914s in the clipping circuit.
Last edited:
Rayma,
the schematic can be downloaded from this address:
AUDIO-RESEARCH D52B Service Manual download, schematics, eeprom, repair info for electronics experts
the schematic can be downloaded from this address:
AUDIO-RESEARCH D52B Service Manual download, schematics, eeprom, repair info for electronics experts
Yes, I have it, but posting it here would help. What other parts have you checked so far?
The transistor beta should not be the problem.
The transistor beta should not be the problem.
Last edited:
I checked the resistances related to Q1 and Q4 (R44, R46, R48, R50, R52, R54), the resistances on the emitter of the amps (from R63 to R70) and the capacitors C26, C28 and C30.
I did not find any defective or out of tolerance resistors or capacitors.
I have no pictures of the diagram.
I did not find any defective or out of tolerance resistors or capacitors.
I have no pictures of the diagram.
I am powering the channels with an external power supply with 26 V positive and 26 V negative and with the output of the AM2 module disconnected from the circuit.
I measured the voltages (absolute value) on the 10 ohm and 237 ohm resistors of both channels:
R46 (on Q1 emitter) => 0.918 V (10 ohm resistor - 1%, channel fault)
R54 (on Q4 emitter) => 1.139 V (10 ohm resistor - 1%, channel fault)
R45 (on Q8 emitter) => 1.141 V (10 ohm resistor - 1%, channel OK)
R53 (on Q5 emitter) => 1.112 V (10 ohm resistor - 1%, channel OK)
R51 (on Q5 base) => 1.728 V (237 ohm resistor - 1%, channel OK)
R43 (on Q8 base) => 1.765 V (237 ohm resistor - 1%, channel OK)
R44 (on Q1 base) => 1.552 V (237 ohm resistor - 1%, channel fault)
R52 (on Q4 base) => 1.759 V (237 ohm resistor - 1%, channel fault)
The 1% resistances are all within tolerance.
I also tried to disconnect one pair of final transistors and then the other (reconnecting the one I had disconnected before) but nothing changes.
Basically by powering the circuit with 26 V positive and 26 V negative (with the output of the AM2 module disconnected) at the output there are 23.5 V negative. The negative output voltage disappears if I disconnect one terminal of the 2K7 - 1W resistor R50 (or R48).
I measured the voltages (absolute value) on the 10 ohm and 237 ohm resistors of both channels:
R46 (on Q1 emitter) => 0.918 V (10 ohm resistor - 1%, channel fault)
R54 (on Q4 emitter) => 1.139 V (10 ohm resistor - 1%, channel fault)
R45 (on Q8 emitter) => 1.141 V (10 ohm resistor - 1%, channel OK)
R53 (on Q5 emitter) => 1.112 V (10 ohm resistor - 1%, channel OK)
R51 (on Q5 base) => 1.728 V (237 ohm resistor - 1%, channel OK)
R43 (on Q8 base) => 1.765 V (237 ohm resistor - 1%, channel OK)
R44 (on Q1 base) => 1.552 V (237 ohm resistor - 1%, channel fault)
R52 (on Q4 base) => 1.759 V (237 ohm resistor - 1%, channel fault)
The 1% resistances are all within tolerance.
I also tried to disconnect one pair of final transistors and then the other (reconnecting the one I had disconnected before) but nothing changes.
Basically by powering the circuit with 26 V positive and 26 V negative (with the output of the AM2 module disconnected) at the output there are 23.5 V negative. The negative output voltage disappears if I disconnect one terminal of the 2K7 - 1W resistor R50 (or R48).
Last edited:
I grounded the base of Q2 and Q3 and the output voltage dropped to - 115mV (it was -23.5V).
On the working channel, if I ground the base of Q6 and Q7 the output voltage is + 38 mV, removing the ground it is about + 8 V.
On the working channel, if I ground the base of Q6 and Q7 the output voltage is + 38 mV, removing the ground it is about + 8 V.
OK, I connected everything, including the AM2 module output and the faulty channel started working again. The 8 ohm output power is approximately 54 Wrms with 1.1 Vrms input at 1 KHz without distortion.
However, I noticed that the 1% resistances are not so critical: I varied one of the resistances from 237 ohm up to 330 ohm without appreciable output variations, I think it is due to the servo loop that compensates for the imbalance.
The amplifier does not have an power on switch and connects to the mains directly when turned on the DC voltage on the output is about 3 V and gradually drops to a few millivolts after a few seconds, I don't know if this is normal, but it does on both channels.
I noticed that even with no input signal the heatsink gets slightly warm on both channels.
However, I noticed that the 1% resistances are not so critical: I varied one of the resistances from 237 ohm up to 330 ohm without appreciable output variations, I think it is due to the servo loop that compensates for the imbalance.
The amplifier does not have an power on switch and connects to the mains directly when turned on the DC voltage on the output is about 3 V and gradually drops to a few millivolts after a few seconds, I don't know if this is normal, but it does on both channels.
I noticed that even with no input signal the heatsink gets slightly warm on both channels.
Last edited:
Yes, i think so too. The initial fault was the two 2SB555 transistors and the 10 ohm and 237 ohm resistors. After changing these components I had probable problems with imperfect connections. On the board of the AM1 and AM2 modules there are rivets soldered to the PCB pads where the various cables are inserted, sometimes the contact between the rivets and the PCB pads is unstable or interrupted.
Those are funlets, and prevent delamination of the pcb traces from stress and heat when attaching wires.
They should be soldered carefully. Without those, the traces easily become loose.
They should be soldered carefully. Without those, the traces easily become loose.
Last edited: