On this site:

PM5126 sine/square-wave oscillator

PM5126-oblique.jpg
The PM5126 is a transistorised LF generator covering a range of 10 Hz to 1 MHz. A really useful instrument for the electronics workbench.

The PM5126 is a transistorised LF generator covering a range of 10 Hz to 1 MHz. It delivers an output voltage of 0-12 V on two outputs: a 600 Ω and a low-Z output. The output signal is a low-distortion sine was or a square wave. An output meter indicates the output voltage. The oscillator also has a synchronisation input. Though I didn't find any production codes on the components, I suspect it dates from the early seventies. The documentation I found is from 1972. My specimen has a test & maintenance sticker that refers to a date in 1978.

This AF generator consists of a number of circuit boards. They plug into edge connectors that are interconnected by wires. The power transformer and both power transistors of the power supply are mounted on the back panel


    The PM5126 seen on the back, power transistor on back panel.

The PM5126 seen on the back, power transistor on back panel.

I assumed that it would be a Wien bridge oscillator like many other AF generators. When I set out to fix this generator and read the schematics, I was surprised, by the circuitry. Philips had used a different type of RC oscillator. No doubt, at the time they have patented this circuit to gain an advantage over manufacturers of “ordinary” Wien bridge oscillators. This is not an oddball circuit, though. It works well. The oscillator consists of three amplifier sections each consisting of three transistors in a Darlington arrangement. Two stages are used for tuning. A phase shifting RC member is connected between the emitter and collector of the output transistor of these stages. The output signal of the third stage is taken from the collector, so this stage is inverting and has an amplification of more than 1. At the oscillating frequency, the two phase shifting amplifier stages give a phase shift of 90 degrees each. With the 180 degrees of the inverting stage, this adds up to 360 degrees. The loop gain is regulated by an NTC resistor in the collector load of the output stage of the third amplifier. This is yet another difference from the classical Wien bridge oscillator, where usually a small incandescent lamp is used as a PTC resistor in the feedback loop of the oscillator.

    Simplified diagram of the PM5126 oscillator.

Simplified diagram of the PM5126 oscillator.

All amplifier stages are DC coupled. A zener diode (not in the diagram) is used to bias the stages such that the output transistor has enough headroom.

That concludes the oscillator board, The other circuit boards contain:

The frequency range switch and attenuator switch are mounted to the front and under the circuit boards.


    Measurements of the signals on the oscillator board.

Measurements of the signals on the oscillator board.

I bought this generator through an Internet auction site, together with a PM5100 valved LF generator and a PM2454 LF voltmeter. At first sight my PM5126 seemed to work fine, but after some time I noticed it was a bit unstable. This became slightly worse over time. Every time I changed the frequency setting, the output meter swung wildly and the oscillator often stopped working for tens of seconds.

Fortunately, somewhere in a corner of the Net there was a manual. First I checked the bias of the transistors in the oscillator but could find nothing. I exchanged the electrolytic (not in the diagram) that serves as a coupling capacitor and keeps the DC component of the output amplifier from the NTC resistor. That didn't make any difference. I started to fear that maybe the NTC was bad, which would be quite an awkward situation as the chances of finding this very special component were close to zero. Finally it dawned to me that the frequency range switch might be corroded. I sprayed some contact cleaner on a Q-tip and cleaned the switch. Black goo. I dried the switch using some more Q-tips and carefully applied some contact lubricant. After this, the oscillator worked fine! The meter still went to the right hand corner just after power-on and changing frequency, but that was just the overshoot of the control circuit around the NTC resistor. The oscillator didn't halt any more.

Now my PM5126 was ok and I had a decent, reliable LF generator.

Copyright © 2012 by Onno's E-page         published 2012-09-29