What determines the duration of off for a gated scr?

Status
Not open for further replies.

macmikeman

Senior Member
I understand the workings of scr's ok. I understand how the gate controls the circuit. What I don't quite understand is how do they set the duration of the off or adjust the phase angle off times per cycle? Is this done via a program inputed into an integrated circuit? Or is that what all them little capacitors I see on printed circuit boards are doing ? I tried to find out from Google, but the search only returned high end gobbledegook way above what I can deciper. Any help would be appreciated. Thanks.
 
I understand the workings of scr's ok. I understand how the gate controls the circuit. What I don't quite understand is how do they set the duration of the off or adjust the phase angle off times per cycle? Is this done via a program inputed into an integrated circuit? Or is that what all them little capacitors I see on printed circuit boards are doing ? I tried to find out from Google, but the search only returned high end gobbledegook way above what I can deciper. Any help would be appreciated. Thanks.

By "off" I assume you mean the delay angle before the SCR turns on since it will turn off naturally at the end of the half cycle?
There are various ways of generating the delay angle before the gate pulse is applied. Obviously, the firing pulses have to be synchronised to the supply voltage.
One way of generating the delay is to (digitally) count the time from the zero crossover. A longer delay means a longer off period and a shorter conduction angle.

This can also be done with analogue circuits.
One technique is to compare a DC level with a portion of the sine wave, typically using op amps.

Firingangle.jpg


Hope that helps rather confuses......
SCR firing is initiated when the control voltage exceeds the synchronising voltage. By adjusting the control voltage between -10V and +10V, the firing angle can be varied over a 180deg range.
 
A little more simplely. An SCR when gated on will continue to conduct until one of two things happens. Either the current through it quits(goes to zero) or the voltage is removed. SCRs are not controlled by regulating the Off time but rather by setting where in the waveform the On time will occur.
 
SCRs are not controlled by regulating the Off time but rather by setting where in the waveform the On time will occur.
Not sure I agree with you on that. The off time or delay angle is the means of control and you can adjust that. Once the SCR is on, it's on until the current goes to zero. The on period is a function of the circuit parameners.
 
090426-0853 EST

Besoeker:

I believe that iaov's usage of "on time" means the "time of turn on" and for "off time" it means "time of turn off". I use these terms in the same way. The time during which something is on I would usually describe as "on duration".

.
 
Capacitor timing

Capacitor timing

One of the most common (or at least most basic) ways to control the on time of an SCR is with a capacitor. (I a litte rusty on these, but I think the basic description below is correct.)

Basically, a capactior is placed in parallel with the gate (parallel with the gate to cathode, I think, but I'm not sure I remember that right??). The capacitor is charged through a series resistor.

When the SCR is on, the capactior discharges through the SCR, so it starts every half cycle already discharged. So synch. with the incoming wave is automatic. It doesn't take any special monitoring of the input wave.

When the SCR turns off, the capactior starts charging through the resistor. When the capactior voltage reaches the voltage necessary to forward bias the gate junction, SCR fires.

Making R a variable resistor changes the RC time constant, which changes how long the capacitor to charge up the the firing voltage, and in turn the time it takes to fire the SCR. So that controls the off time (and the on time.)
 
090426-0853 EST

Besoeker:

I believe that iaov's usage of "on time" means the "time of turn on" and for "off time" it means "time of turn off". I use these terms in the same way. The time during which something is on I would usually describe as "on duration".

.

Fair enough.
For the avoidance of doubt, we generally refer to the "on" period as the "conduction angle". There is some logic for this.
In this context of this application, it is often measured or described in degrees
 
Firing angle generation is the heart of SCR based control systems. Before the advent of digital control systems, the phase angle firing decisions were determined by a complex series of analog sensing and follower circuits with descrete components. Over time, those got boiled down into a condensed set of components on a "thick film" firing board, kind of like an IC only cheaper to make in smaller quantities. The firing angle command was then sent to a "pulse generator" to provide the actual firing sequence of the gate circuit by waiting long enough to affect the RMS output. Then finally, depending on the firing method chosen for the SCRs, the signal goes through a "pulse transformer" that amplifies it to make sure there is enough energy to ignite the SCR gate. More modern controllers are now using digital firing techniques that do what the thick film and pulse generator did, but still go through the pulse transformers.

How you refer to the timing of the firing is, as you can see from the previous posts, a bit of a senmantics issue. But suffice to say that the turn-on (gate) timing is determined by first sensing where the zero-cross point is (or is supposed to be). There are several methods of accomplising that, each with their own advantages and drawbacks. Once you know where the the zero-cross point is, it's compared to the desired phase angle and then sent to the firing control system. If it's a soft starter, the determination of the desired phase angle is dynamic (changing) based on the ramp profile and/or current limit set up by the user. If the SCR control is in a rectifier, it's usually fairly static and based on the desired DC voltage level needed after rectiication.
 
How you refer to the timing of the firing is, as you can see from the previous posts, a bit of a senmantics issue. But suffice to say that the turn-on (gate) timing is determined by first sensing where the zero-cross point is (or is supposed to be). There are several methods of accomplising that, each with their own advantages and drawbacks. Once you know where the the zero-cross point is, it's compared to the desired phase angle and then sent to the firing control system.
We have been making SCR based controllers for about four decades. The early firing circuits were all analogue and and the firing delay angle was based roughly on the sort of comparison technique I posted earlier. Supply voltage "notching" is a known characteristic of SCR equipment so using the supply voltage for the comparison isn't such a good idea. We opted for a cosine wave comparison. Generating the cosine effectively removes the rubbish. That, plus a bit of bias confers some quite nice benefits.
The input to output transfer function is linear and changes in supply voltage are corrected open loop. Some of the ideas and calculations are maybe a tad more complex but the practical implementation is surprisingly simple. We still use this circuit where the supply is know to be or expected to be "dirty". The stuff practically bullet-proof.

Zero crossover detection as a synchronising point is OK with clean supplies.
I have seen it go horribly wrong where the supply was not so good. Contributed to my grey hair, I'm sure.:wink:

If the SCR control is in a rectifier, it's usually fairly static and based on the desired DC voltage level needed after rectiication.
We do quite a lot of projects that use SCRs for rectification. DC drives, anodising and alternator static excitation are examples. All are variable voltage.
We have done some for fixed voltage. If you ever visit Bermuda, our phsophremitically weighted chargers power the telecoms.:)
 
This has been very informative. Thanks. What I am actually trying to do is replace the digital controls (shot and too expensive to replace in 2009) of an electric conduit bender with a dc motor, to relay and contactor controlled, which is well and easy enough for me (the reverse relays and brake circuit part) . The harder part is dealing with inrush thru the motor armature which if I am right is probably best done with scr control of the dc power. This is a permanent magnet motor 115 dc, 1 hp. I might be wrong about it but from what I think I know about dc motors, across the line starting at 1 hp is definately a no-no. Is the timing of the scr for something like this a function of the scr type, or is it done thru a seperate control over the gate voltage?
 
Status
Not open for further replies.
Top