| Vibrator - replacement with "solid-state circuit"
Following on from Astor CR car radio restoration 12volt car radio with vibrator type HT supply. The method of obtaining high voltage (200-250V DC) needed for the valve-anode circuits, was (in earlier times) acheived by using an electro-mechanical vibrator and step up transformer. The vibrator would switch the incoming 12v DC supply back and forth to each side of a centre tapped transformer primary. This constituted a (somewhat crude) alternating potential which would then by transformer action, be stepped up to 250v or so. We all know transformers only work on alternating voltages, and that they must obey the law of electromagnetic induction. Rectifier diodes and smoothing filter capacitor then provide the required DC HT voltage. As replacement Vibrators are un-obtainable; best idea is to replace/upgrade a (faulty) mechanical vibrator with a solid state equivalent circuit. Advantage: No moving parts, no contactor arc-ing, noiseless and as some will say, greater reliability. The pre-existing laminated steel core step-up transformer is retained. The circuit shown below will essentially replace a vibrator.
A CMOS integrated circuit (74C04 Hex inverter) oscillator provides a 100Hz clock pulse. (equal mark/space ratio 1:1) I also tried other IC's like 4069, 74HC7404, which are pin compatible and also work well. However the HC high speed cmos is restricted to a 5 volt max supply rail. This frequency is the same as the mechanical vibrating reed inside the vibrator. The vibrator contact behaviour is a break before make action. The electronic circuit should emulate that of the vibrator which it is replacing. This is acheived in the electronic circuit by the R-C-diode network and inverter. 33K 0.01uF and 1N5817 schottky diode See diagram for waveforms. With the values shown a 300uS delay is introduced at the onset of each "gating" pulse that alternately turn on each MOSFET This is the needed "commutation" and prevents both Mosfets being gated on simultaneously, at switch-over point. This would be like a brief short circuit across transformer primary. After prolonged use power dissipated in the Mosfets could exceed their specification. Because of power MOSFETs very low "on" resistance; almost all of the available voltage is impressed across each half winding of the primary, cycle by cycle. There are well designed solid-state vibrator replacement assemblies now available (see references - below) Most seem to use a PIC microcomputer to achieve the needed pulse rate & widths etc which then gate on the Power Mosfets. The above being a hardware only design, any variation in clock frequency, and delay is achieved by changing of the appropriate "timing" resistance.
In the test construction shown, I have used a mains transformer 240V : 12v centre tapped (connected "in-reverse"). From a 12V input supply, the output is 300V DC when driving a 22K load (4 Watt @13mA) The Power Mosfets run cool! References: RVB 1 Solid State radio vibrator inverter by PeKo (USA) Silicon Chip Magazine: Solid State vibrator replacement (Pg 97/98) Solid State vibrator replacement in old radios Engineering Data on Vibrators CMOS driven MOSFET DC to DC converter Solid-State-Vibrator-Replacement-for-Old-Radios (Ebay Aug 2019)
� 3Q 2019 |