# 3 Phase Controlled Rectifier

Discussion in 'General Electronics Chat' started by roro36, Jan 25, 2012.

1. ### roro36 Thread Starter Member

Sep 7, 2010
52
0
Hi..

I'm having trouble undersdaning how 3 phased controlled rectification works. 6 Thyristors are used to complete the action. It's the firing sequence and theory that I cant grasp. Most places say to fire them in pairs, which means we can fire 1A and 2B(3A should be on as well), then when it is 2A's turn to fire, 2B should turn off reverse biased and 2A can be pulsed on along with 3B... Not sure if that is correct, but that means that it's not perfect rectification as 3B should have been on 30degrees earlier? That is my first issue...

Then how would actually go about deciding when to pulse each thyristor? Would you wait for zero crossing and then just hold it on for a few degrees before the thyristor latches?

2. ### jimkeith Active Member

Oct 26, 2011
539
99
Three phase non-controlled rectification operates in similar fashion, except at the peaks of the applied voltage--may be easier to visualize.

Initially, to complete the circuit and get current started, two thyristors must be fired simultaneously; one in the top half and another in the bottom half of the bridge--after that, a 'new' device must be fired or rotated in every 60° of phase rotation, alternating between the top half and bottom half of the bridge--each device conducts for a total of 120°. While they conduct in 'pairs' of adjacent 60° segments, they are fired individually.

While everything is timed relative to zero crossing, they are generally not fired at this point--depending upon the desired output voltage, they are fired after zero crossing--the gate pulse must extend long enough for the thyristor current to integrate past its holding current level that is a function of load inductance--perhaps 1us for resistive load or 5us for inductive load. At low voltages where the load current is discontinuous, the gate pulse must have a tail that extends for an additional 60° or more to guarantee that the subsequent thyristor that fires completes the circuit--this may be done as a long DC pulse or a series synchronized shorter pulses that may be easily isolated via a gate drive pulse transformer.

To simplify synchronization, I suggest that your circuit sample phase to neutral rather than line to line voltage--this makes synchronization indifferent to phase rotation that can be a complex issue. Create your own artificial neutral via (3) resistors and sample the voltage via attenuating differential amplifiers with high resistance values. Then phase shift the analog AC voltage 30° (if I remember correctly) via a simple RC filter--also helps to remove line voltage hash that can mess with your synchronization.

Last edited: Jan 25, 2012
roro36 likes this.
3. ### roro36 Thread Starter Member

Sep 7, 2010
52
0
Thanks... That is exactly what I was looking for...How, in your example would the system know when the zero crossing is for positive cycle or negative, as this would determine the correct thyristor to switch on, i.e from zero crossing to positve half cycle or to negative half cycle.

4. ### jimkeith Active Member

Oct 26, 2011
539
99
Simply take a top device and a lower device that are connected to a given main. When the voltage (phase to neutral) is positive, fire the top device and when negative fire the lower device. Zero crossing may synchronize the phase delay function for these two devices, but the subsequent polarity of the voltage further steers the gate logic to fire the upper or lower device.

Last edited: Jan 25, 2012