CRT Question - Old Calculator

Hello, all, Here's another update on the Wyle Scientific WS-02 calculator. Yesterday I tested out the power supplies with the logic chassis and CRT connections and deflection board disconnected, and dummy loads made from automotive brake light lamps in parallel on the +12 and -12V supplies. These provided enough load on the supplies to check for ripple under load, and assure that regulation was also working properly under load. The power supply passed with flying colors, still <5mV ripple on all of the supplies, and the regulation was within a tens of millivolts of specified voltage. I also tested the supply by varying the mains voltage by +/- 10%, and regulation stayed stable. I also loaded the -30V supply with a single bulb, not knowing what kind of current it was capable of sourcing, and it also was stable and had minimal ripple. I also did a check of the high voltage supply, using a rubber-insulated screwdriver with a heavy piece of stranded copper wire wrapped around the shank tightly, connected to a NE-2 lamp, the other side of which was connected to chassis ground. I was able to get a solid discharge through the lamp from the supply by moving the end of the screwdriver close to the point where the high-tension wire connected into the HV power supply. So, it appears that the power supply is working. Thankfully, I didn't get zapped. Another discovery is that the tubes in the power supply are not triodes, and I also miscounted the pins. These are 9 (not 8) pin mini tubes, and I was able to get a part number off of the second tube by examining it very closely with a bright light while it was still plugged into its socket. The tubes are 1V2 diodes. I'm not sure what they are for, but on closer inspection of the circuit board that they plug into, there are a number of transistors that I suspect are wired as an oscillator, but what the diode tubes do is a mystery. The datasheet for the 1V2 tubes says they are primarily used in the flyback circuitry of television sets, so perhaps they are used in a similar way in the HV power supply in the calculator. After running the power supply under load for a few hours, and seeing the voltages and ripple remain very stable, I was confident that I could put the logic chassis back in, and reconnect everything, and try powering it up. I discharged the high voltage section to make sure I wouldn't zap anything when I was installing the logic chassis. I put everything back together, carefully checking everything to make sure it was connected correctly. The designers were thoughtful in that they put keys on all but one of the connectors (the one without the key is the one that plugs into the deflection board, but it was very obvious how it plugs in). I hooked up the DVM's to each of the power supply lines, and slowly ramped up the Variac. After about 70% of mains, the supplies had stabilized at their proper voltages. I could also see the heater in the CRT starting to glow. At around 85%, the display started showing signs of life, but it was very unfocused. As I slowly ramped up to 100%, the display sharpened, and I was greeted by six lines with decimal points that I could move around by changing the decimal point selector. Also, the line that displays the ENTRY register had an intensified 0 at the left end of the display, and there was an intensified 0 positioned at the left-most digit of the ENTRY register. Per the manual, I pressed CLEAR ACCUMULATOR, CLEAR MQ, and CLEAR ENTRY, and the display flickers all zeroes when these keys are pressed, then returned to the "two zeros" display, which is the expected behavior. I then pressed a digit key. The intensified zero moved one digit position to the right, but no digit was entered. Repeated digit entry would move the cursor one digit position to the right for each digit press, but no digit would enter. The BACK SPACE and FORWARD SPACE keys would move the cursor digit as expected. The TO and FROM register keys would move the register selection "0" around as expected. Math operations, which couldn't do much since the only thing in the registers is zero, caused the display to flicker, and then return to the normal display. Pressing square root causes the display to show all zeroes, with an intensified 0 in the MQ register moving from left to right, then returning to the 'idle' display. Pressing the DIVIDE key resulted in the display going to all zeroes, and the machine locking up. I'm assuming that this is because it is trying to divide by zero. So, clearly there is something up with digit entry. Either that, or the delay line is not properly serving as a memory for all of the registers. There is going to have to be some deeper diagnostics done to try to figure out what is wrong. The difficulty is that the logic chassis in the Wyle Scientific was designed to only be replaced as an entire module. The 8 circuit boards are all hard-wired together with countless individual wire jumpers that interconnect the boards, so getting to any of the boards except the front-most and rear-most boards is going to be a challenge. Despite the issue, I'm very pleased it is as alive as it is after all of these years. Thanks again to all for the input on the HV power supply, and the interest in this old calculator. -Rick Bensene