The HP 64000 Logic Development System first introduced in 1980 consists of a mainframe, either the original 64100A or the later 64110A, both of which have a built-in CRT display and a full size keyboard and can be fitted with emulation and analyser cards for many microprocessors in use at that time. I have wanted to own one for some time and recently found a 64110A in the USA sold as faulty with no working display which was shipped at great expense but I was hopeful that I could get it working again. The seller told me that it performed the power on self test correctly so it seemed that the problem was in the CRT or display driver. He also told me that at some time in the past an attempt at repairing it had been made and some connectors had not been replaced. The HP 64110A shares much of its design with the HP 4955A Protocol Analyzer so the information here also applies to the HP 4955A.
When it arrived I found that it was in very good condition apart from a few pieces of the case broken off during transit. Before doing anything else I checked the service manual to find out how to change it from 110V to 220V AC. Unlike most other equipment the selection switch is on the power supply unit inside the case and the manual states that the setting can be changed without removing the side panel but I needed to remove the side panels anyway. The fuse needed to be changed from 8A to 4A but, as I didn't have any, I used a 3A fuse in the mains lead plug until I could obtain some.
This is the view of the four emulation cards which I removed before powering up the mainframe.
Now it is time to switch it on and see what happens.... It performed the self test and beeped a lot which is a good sign but no display and no sign of any high voltage either.
Here is the view of the underside of the mainframe with the display driver PCB A6 swung out.
A closer view of the A6 PCB
You will notice that the white connector socket on the left should be pushed onto the end of the CRT also there is another connector on the right which should be plugged into the A7 board opposite to the A6 board. That would explain why the display did not work but my first thought was that perhaps the CRT was faulty and before reconnecting it I should check it out with a multimeter. I checked between the pins at the neck of the CRT for low resistance and only found low readings between pins 3 and 4, ok because this is the filament, and between 5 and another pin which I didn't make a note of but also ok because there is no wire connected to pin 5. So I replaced the connectors and switched on from a distance..... There was a small flash from the power supply area and the fuse blew.
Could it be a PSU fault ? So I replaced the fuse, disconnected the A6 board from the backplane by pulling out the IDC connector marked 8120-4048 in the above picture and switched it on again. The power supply fan ran but, of course, there was no beeping this time because the loudspeaker is on the A6 board and is now disconnected. There are test points TP1 to TP7 on the backplane to allow the supply voltages to be measured. TP6 marked +12DD for the display driver boards measured 12V and +5V, -5.2V, -3.5V, -12V and +12V were all correct and I switched it off.
I removed the connectors from the CRT and A7 board and replaced the IDC connector for the A6 board and switched on again. As before the selftest beeped correctly and I measured the voltages at the test points again. TP6 measured only 7.5V but all the other supplies were correct and I switched it off.
There is a IDC DIL plug which connects the A7 board to the A6 board and I removed it and switched on again to isolate the problem to one of the two display driver boards. This time before I could start measuring voltages there was a burning smell and I immediately switched it off.
Now that I knew which board had the fault I looked carefully at the A6 board for anything suspicious. One of the transistors was leaning at an odd angle, not that important in itself but when I looked closer there was a lump in the package.
I have previously seen devices with internal faults that blow a hole in the package but never one quite like this. The failed component is Q5 HP part 1854-0798 (MPS U45) NPN darlington transistor, once I had removed it from the PCB it measured 53 ohms between the emitter and collector. The transistor beside it is Q6 HP part 1853-0449 (MPS U95) PNP darlington transistor and its emitter is connected to the emitter of Q5, measuring Q6 showed that it was a dead short between the emitter and collector. This explained the low voltage on the 12DD supply as the Q5 collector is connected to the 12DD supply through a diode and the Q6 collector is connected to 0V so that the 12DD supply was effectively shorted to 0V through a diode and 53 ohm resistor. I checked all other transistors on the A6 board in addition to the transistors on the A7 board but they were all fine and replacements for Q5 & Q6 were ordered.
When the MPS U45 and MPS U95 transistors arrived I fitted them to the A6 board and switched on. Measuring at the 12DD TP6 test point showed 12V. I connected an oscilloscope to the TP7 test point on the A6 board to verify that the vertical sweep, provided by Q5 & Q6, was now operating correctly which it was. I then switched it off, reconnected the connectors for the CRT and the A7 board and plugged the DIL plug into the A6 board. I switched on and watched the rear of the CRT expecting it to glow but saw nothing, I glanced at the screen and the self test was displaying a test pattern.
to be continued
home contact