Oscilloscope repair
The 2465B power supply consists of two boards mounted together as a sandwich: The A2A1 Regulator and the A3 Inverter. The picture above highlights the two capacitors on board A3 that are swapped in Figure 10-13 in the service manual.
Components, A2A1 Regulator board:
Components, A3 Inverter board:
|
Tektronix 2465B repair - Part 1
One day my otherwise reliable 2465B would not power up, but would just flash its front panel LEDs. There was also a clicking sound from the power supply at the same rate as the flashing of the LEDs. My best guess was that this was a symptom of the well-known "worn-out capacitor" illness. Rather than looking for the perhaps single culprit I decided to replace all electrolytic capacitors in the switch mode supply, in view of the age of the oscilloscope, with the exception of the two input capacitors C1022 and C1021 (the two large blue cans in the picture.) All capacitors were replaced with 105 C types, and low ESR types were used on locations handling switching currents. A couple of bipolar capacitors were replaced with polyester film capacitors. You may find a complete list of the replaced components in the table on the left. WARNING: I had a fine help from the 2465B service manual to help disassembling correctly, but the two capacitors C1132 and C1115 are swapped by mistake on "Figure 10-13. A2A1 - Regulator and A3 - Inverter boards". If you populate the board with new capacitors according to Figure 10-13 you will fry the capacitor on the location printed C1115 due to a much too high rail voltage, and the inverter will not work properly as the capacitor inserted on the location printed C1132 will be too small. The two capacitors in question are marked with red arrows on the figure. Another slip by Tektronix relates to the two current limiting resistors A2A1R10 and A2A1R1019, both 15 Ohm / 0.5 W, in parallel with the NTC resistors A2A1RT1010 and A2A1RT1016. During power on, they will have to sustain a surge, and sure enough: Both turned out to be blown up into an open circuit, despite no visual damage. Both resistors were replaced with surge resistant types with 3 W rating. After the replacement of components (and after applying some contact cleaning spray inside the beam finder switch that tends to be stuck) the 2465B is now back to normal. |
The arrows indicate where the tracks were broken due to leakage of the three electrolytic capacitors above.
|
Tektronix 2465B repair - Part 2
After the repair of the power supply, the 2465B behaved well for about a year or so, but one day, nothing would appear on the screen, and the power-on cycle just ended with the ADD indicator being lit. It turned out that the 2465B now suffered from two another well-know 2465B failure mechanisms: 1) The calibration memory had lost its contents, and 2) Some electrolytic capacitors on the A5 board had leaked and had made life miserable for the DAC circuitry that produces a number of voltages which define vertical position, brightness, among other. This summarizes the flow of the repair: By slightly turning up the GRID BIAS trimmer the error message TEST 04 ERROR 11 appeared. This is a symptom of calibration data errors, or loss of calibration data. I soldered out the Dallas_DS1225Y, and a readback using a GQ-4x4 USB Universal Programmer revealed that all data was indeed lost. After having read the note by Dr. H. Holden on the 2465B calibration issues, which I recommend you look up on the Internet, I replaced the Dallas DS1225Y with a FM16W08, a 64k FRAM in 28-pin SOIC (Farnell 2077743) mounted on a 28-SOIC to DIP adaptor (Farnell 2476045), and using two 76341-314LF 1-row 14-way sockets (Farnell 1098045). An image of the calibration data from another 2465B was programmed into the FM16W08 by the GQ-4x4. Once the new calibration memory was in place, the error message TEST 04 ERROR 11 disappeared. Instead, I now got the message TEST 05 ERROR 44. For some 2465B owners this may also ring a bell: The error message is a symptom of voltages from the DAC on board A5 being wrong. The DAC produces a number of voltages that define the vertical position, the brightness, etc. To cut a long story short, and sparing you the details of the tedious measurements around the DAC circuitry, I found that leaking electrolytic capacitors close the DAC circuitry had etched the copper tracks at two locations: Very close to the pad of DAC U2101 pin 19, and very close to one of the pads of R2014. I had to solder a wire from DAC U2101 pin 19 to R2521 to ensure the connection to multiplexers U2530 and U2531. Also, I replaced R2014 and added a piece of wire to bridge the missing connection. Having done this, neither message TEST 04 ERROR 11 nor message TEST 05 ERROR 44 appeared. Next task: To go through the calibration steps.... |