Here¡¯s an update on the circuit I mentioned earlier for avoiding the occasional ¡°thunggg¡± turn-on sound which is due to remanence-induced current surge.? This circuit might reduce stress on the transformer, but is unlikely to remove the risk associated with improper de-powering of the transmitter.? It avoids the remanence problem by always turning the transformer on at the beginning of one polarity of the applied AC power and always turning it off after the other.? Thus, the initial magnetization of the transformer cannot build upon remanent magnetization already present.?
?
A small transformer scavenged from a ¡°wall wart¡± power supply connects to the switched AC line.? Two diodes in series with one side of the secondary feed a 220 uF (100 uF should be plenty) capacitor and then a 5.4V LDO regulator.? The other side of the secondary goes to this circuit¡¯s ground (not connected to the HT-37¡¯s ground).? A voltage divider connects after the first diode to provide a pulse once per cycle to an NPN transistor, which in turn supplies the Trigger input to a 555 timer IC.? The 5.4 VDC supplies the 555¡¯s Vcc, enabling the 555¡¯s output to provide 4.0 V to activate a pair of OMRON G3MB-202P solid state relays connected in parallel.?? The 555¡¯s timing resistor and capacitor provide an output pulse approximately 15.3 ms long.? The 555¡¯s Reset line connects to Vcc through a resistor.?
The 555¡¯s output pulse begins during the AC half-cycle (A) that triggers it.? It¡¯s still high during the subsequent zero-voltage crossing of the AC waveform, at which time the SSR connects the AC power to the transformer primary at the beginning of the next half cycle (B).? The SSR control input is still high during the next zero-crossing, enabling the next half-cycle (A again) to continue powering the primary.? During this half cycle, the SSR control pulse ends.? The AC power pulse retriggers the 555 1-2 ms later, maintaining the power connection when the next half-cycle (B) begins.? This continues until someone turns off the transmitter¡¯s power.? When that happens, the 555 does not get retriggered, so there¡¯s no voltage on the SSR¡¯s control pins at the next zero crossing.? Thus, half-cycle A is the last one powering the primary.
The images below illustrate the turn-on and turn-off using this circuit.? One of the turn-offs occurs only a few seconds after turn-on while the filaments are still warming up.? The other occurs after warm-up.? I think the bump on the current waveform represents the additional current drawn to top off the filter capacitors.
While this circuit works, I¡¯m not certain it will work well with every turn-on of the HT-37. ?This is because turn-on can occur at various times during the ¡°A¡± half cycle, possibly sometimes providing a 555 output pulse that starts late.? If it starts late enough, the 555¡¯s output pulse to the SSR might still be high when the next trigger pulse arrives, causing it to be missed.? In such a case, the transmitter would turn off momentarily and then on again.? One could shorten the 555¡¯s output pulse so that it¡¯s guaranteed to end before the last possible time that the next pulse could arrive.? But that would increase the risk of sometimes turning off the SSR just before the current crosses zero (the current zero-cross slightly lags the voltage zero-cross).? It¡¯s not practical to test this circuit for all possible timings of turn-on relative to the AC power cycle.? So, I¡¯m considering narrowing the window of possible trigger times by substituting a chip that activates its output pulse on a falling transition instead of a voltage level (CD74HC4538).? I¡¯m also considering setting up an additional SSR control pulse that¡¯s a few AC power cycles long to cover the initial several cycles while the pulse timing settles down.?
Please post or send comments and suggestions!
Cheers
Halden VE7UTS
Waveform images:? Yellowish orange is voltage applied to the primary.? Turquoise is the voltage across a 0.2-ohm resistor in series with the primary.
?
?
