Paul, and all,
The gates of SCR's need to be pulse fired for reliable
phase control. Commercial units have pulse train burst applied to the gates
through a pulse transformer. The phase angle of the burst is derived any
number of ways which can be very precise and repeatable.
It just takes some op amps and in some cases a few logic ICs, or a
special function IC, to obtain to control signals.
Back in the age when SCRs were high in fashion, unijunction transistors
were also fashionable, and many SCR circuits of the day used UJTs to
generate the gate trigger pulses.
In my own SCR and TRIAC circuits I never used UJTs, though. Instead I
either generate a nice trigger pulse using a DIAC behind a capacitor, or
I trigger the devices with a long pulse, often as long as 1ms, generated
by logic ICs, op amps, or more commonly a microcontroller. Reasonably
modern SCRs have trigger currents below 50mA, often far below, which is
low enough that 1ms pulse length isn't a big waste of energy. When one
has to trigger a device at the zero crossing, such a long pulse is very
useful, because one can start the pulse right at the zero crossing, it
will turn on the device, AND hold it on until the current has risen
enough to keep the device on for the remainder of the semicycle. With
old SCRs that required several amperes to trigger, a similar effect was
achieved with a train of narrow pulses.
When the load is inductive, current will ramp up at a certain (possibly
low) rate even when the device is triggered in the middle of a
semicycle, so the long trigger pulse, or train of short pulses, is often
required anyway, with the single long pulse being safer.
Manfred
========================
Visit my hobby homepage!
http://ludens.cl
========================
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|