The PM2420 is a digital bench multimeter, made in France. Its display has 3 ZM1000 nixies and a neon bulb as a leading "1" digit. It's documentation does not mention a date of manufacture, but I found the year 1969 on an addendum to the manual. Like the PM2433, it uses discrete logic. The A/D conversion principle is different though: the PM2420 uses the "ramp" principle. It contains only 93 transistors (plus 8 in the power supply) and an opamp. The design fits into a series of measuring instruments Philips sold at the time.
The meter has 23 ranges, selected by a single central knob. The range switch doubles as a power switch, position 24 is "OFF". The range of the A/D converter is 0-1000 on the ohms and DC ranges, so the "1" neon bulb is seldomly used. The display will show values of up to 1450, but an overrange input of more than 5% will result in larger measuring errors.
The circuitry is divided over a number of circuit boards. Flip-flops and gates in the logical circuitry are built up like a sandwich, components placed between two small PCBs.
I bought this meter in March 2002, and was lucky to find a manual with it. When I switched it on, it appeared to work. But not to specs. The higher DC and AC voltage ranges functioned. But the lower voltage, current ranges and ohms ranges gave an overrange reading (around 1450) and did not respond to the input.
Looking at the schematics, I noticed there were two fuses that protect the input amplifier for low voltages and the current source for the ohms ranges, respectively. These were blown. I replaced the fuses and hey! The low voltage and ohms ranges worked again. But on the current ranges, I got a high reading again, even when the input was not connected. I checked the shunt resistors. It turned out that the shunt for the 1 A range was interrupted. And indeed, it looked a bit burned. That explained a lot. Probably the previous owner had tried to measure mains voltage while his probe cord was in the 1 A input. This fried the 0.1 Ω shunt and blew up the fuses. I removed the burned wire from the shunt resistor and put some new nichrome wire in place.
After all this, this is now my first really usable nixie multimeter.