Three Phase Thyristor Control with microcontroller

[zab]

I have seen some circuits where thyristor are used to switch on and off ac supply. Like wise thyristors are used as regulated rectifier using phase angle control.I have used micro controller to sense zero crossing and then using timer interrupt to control the ac volt with adjusting firing angle of The thyristor in single phase . I want to use this method(phase angle) for three phase for power regulation. There are two possibilities for that.
1 to use three thyristor and three diodes just positive side control
2 use 6 thyristors for positive as well as negative control.

My questions are how to do that. any flow chart , code .algorithm.
More questions are:-
1.Use 1 zero crossing and adjust phase difference  for each?
2. Use 3 separate zero crossing for each phase?
3. What type zero crossing will be the best?
4. For star and delta will there be a difference?
5. Optocoupler is used for smaller thyristors how about transformer for switching thyristors
5. Is there any way to improve power factor?

 
Thanks in advance.

[thunderer]

You could start here: http://www.nxp.com/documents/application_note/AN467.pdf. I'd think about using opto triac drivers with built in zero crossing detection. It is an interesting project, I wanted to have the time to do it.

[zab]

You could start here: http://www.nxp.com/documents/application_note/AN467.pdf. I'd think about using opto triac drivers with built in zero crossing detection. It is an interesting project, I wanted to have the time to do it.

The quoted example again related with single phase.
Any three phase example ?
I gone through google and got some doc. one example ie "thyristor driver  circuit" in eda board used 4 pic control for this. why ?
http://www.edaboard.com/thread194314.html
http://www.edaboard.com/attachments/55530d1303382533-thyristor-driver-circuit.pdf
Just for complexity or some thing else?

[Sideshow Bob]

I have this book, General Electric corporation scr manual 5th edition. It may be old but the basics are still the same
Ps do not post this link outside this forum

[zab]

The book is really special.Will it provide guidance for what i am looking for I do not know, but it is good one for understanding of scr especially for new ones to power field.
 

[sam_des]

Hello zab,

I have done that for Stud Welding Machine. Following points might be helpful for you,

1) You will need three different ZCDs. Either Star or Delta will do. Step-Down Voltage Transformer followed by OpAmp will give you best results with good safety. Try to minimize the phase difference between actual ZCD and detected ZCD.
2) Having 3 ZCDs will also help you determine Phase-Sequence of R-Y-B. This is required for Correct thyristor firing sequence. You can also detect incorrect phase wiring and prevent undesired operation.
3) For Thyristor firing, some comments from my Source-Code header,

Code:

/*
  Description:-
  ==============
    We need to drive the 3-Phase Fully Controlled SCR bridge for Voltage/Current
  control. Since AC input to this bridge is 50Hz, base frequency is 50Hz, but
  actual drive signal must high frequency, i.e. beyond audio, to minimize humming
  noise during operation. Also to reduce power requirements for driver, drive
  signal is applied only for a pre-defined period instead of entire cycle.

    1) Base Frequency       - 50/60 Hz
    2) Carrier Frequency    - 20,000 Hz
    3) ZCD Input            - Actual +ve & -ve Zero Crossing of each phase.
                              +ve Zero Crossing - Falling Edge.
                              -ve Zero Crossing - Rising Edge.
    4) Bridge Config        - 1   3   5
                              4   6   2
    5) SCR Conduction angle - With firing angle = a,
                              6+1 -  (a+30) to  (a+90)
                              1+2 -  (a+90) to (a+150)
                              2+3 - (a+150) to (a+210)
                              3+4 - (a+210) to (a+270)
                              4+5 - (a+270) to (a+330)
                              5+6 - (a+330) to  (a+30)
    6) Max. Period for Drive- 1mSec
    7) O/p Voltage with Free-wheeling diode at the output,
            V(dc) = (3*sqrt(3)*V(ll)*cos(alpha))/pi,                                for 0 <= alpha <= 60
                  = (3*sqrt(3)*V(ll)(1+(cos(alpha)/2)-(sqrt(3)*sin(alpha)/2)))/pi,  for 60 < alpha < 120
        NOTE: alpha=0 is 30 deg from actual zero crossing of line voltage waveform.
              v(ll)=Line-To-Line voltage of secondary (not w.r.t. to neutral if secondary is star)
*/
/*======*/
  Base frequency is 50Hz.
   A ZCD interrupt will occur for each +ve zero crossing of each phase i.e.R,Y,B.
   ZCD sequence will be,

        INTx Fired      INTx Polarity       ZCD Detected        SCRs to Fire
      =========================================================================-
            INT1        +R - Falling            1                   6+1
            INT2        -B - Rising             2                   1+2
            INT3        +Y - Falling            3                   2+3

            INT1        -R - Rising             4                   3+4
            INT2        +B - Falling            5                   4+5
            INT3        -Y - Rising             6                   5+6
    ** No 2 ZCDs(Edges) occur simultaneously.

   For first 30 deg of phase it useless to try & drive the SCR, since it won't
   be forward biased, so we add the 30 deg = 1.66mSec phase shift to our calculations.

   60-ZCDs per 360 deg, Angle between successive ZCDs is 60 deg.
   We can have, Min Prefire delay = 1.66mSec + 0.00mSec - Fired on ZCD + Phase Shift.
                Max Prefire delay = 1.66mSec + 3.33mSec - Never fired at all.
                Max. Fire Period  = 1.00mSec
                Min Fire period   = 0.00mSec - Never fired at all.

   Since we generate pre-fire/fire delays using interrupts & with RTOS, we need
   certain minimum & maximum values for these delays.

   For maximum possible resolution we use Timer3 16-bit timer in Fast PWM interrupt mode.
   @ 16MHz clock, we have, Timer3 tick      = 62.5nS
                           Prescalar        = 8
                           Eff. Timer3 Tick = 500nS
                           6.66mS           = 6.66m/500nS

Note that code is for ATmega128 @ 16MHz, running FreeRTOS 7.0.0.

For actual driving, again 3 transformers were used. Since frequency was 20kHz, we used Ferrite Transformers, resulting in small size. Note that design of the transformers greatly depend on SCR and snubbers must be on actual SCR terminals or very near to it.

I have some docs regarding this which I will try to get from my backup.

Hope this helps,
sam_des

[max]

Hello sam_des

I am not expert in this field, I have a question

Is it possible to use the three uCs independently with common set point?
In this case phase sequence RYB is necessary?

Regards

[sam_des]

@max,

Yes, it is possible to use 3 uCs. But then you will need to sync them with each other somehow. All 3 must have same firing angle at any cycle. And R-Y-B phase sequence detection is required. What if instead of R-Y-B, it is connected as R-B-Y or anything else, while our uC, happily assume correct connections ? We can flag user of incorrect connections. There are instruments which do this and are regularly used during 3-phase wiring, even when wires usually are color-coded.

Also where to connect V/I feedback to regulate DC output if you have multiple uCs ? Do you have any special reason for doing it ?

In my opinion, it will much easier to use single uC. A 8-BIT uC can do it. Even good-old 8051 will do the job with lower carrier frequency.

It might look complex at first glance for 3-Phase, but once you understand the basic principle of SCR firing, it is actually quite straight-forward. Just read few Power-Electronics books. Most entry-level texts explain in detail, 3-phase operation for uncontrolled, half-controlled and fully-controlled bridges with various loads -R, L, C.

The book posted by Sideshow Bob is indeed great theory and practical oriented book. Another good one is Power Electronics by Cyril W.Lander. This also quite old-1981.

Just a word of caution, 3-phase is dangerous and not to be played with candidly. Be sure of every bit of safety before you turn the switch on.

regards,
sam_des

[zab]

sam_des Thanks for guidance and good tips.
The books you have mentioned is good one I just found it  here http://www.mediafire.com/download/8a2yflcjaa74s3u/Power_Electronics_3rd_ed-_Cyril_W._Lander.pdf

I intend to use one Controller for this I want to keep its cost low so only one transformer and one zcd.For Other I want to use 2 or 3 optocoupler for phase rotation detection. Will it be ok?
If the input transformer is attached to 2 phase instead of 1 phase and neutral. Will it make any difference to zero crossing detector?If yes then how to adjust that?.
What type circuitry one should used for feedback for the stability of out put? ADC will be enough or comparator type of circuitry will be required?

Zero crossing get disturbed by voltage variation and noise.How to keep it trouble free in noisy environment?
Drive transformer or optocoupler is best  for SCRs drive with respect to performance?



 

[max]

Hi,

I found an article from my archive, please see the attached pdf.

Regards

[zab]

Thanks, your archives contain valuables what more store there.This way is really simple but if possible share high resolution pages it is not fully understandable especially drawing .
I intend to make some thing unique having little more functions in the power controller.

[max]

hi zab

I am sorry I do not have hi res copy, see the attached image file, I enlarge the
pdf image and combine the both pages.

[sam_des]

@zab,
I have attached the image of SCR driver from my Stud Welding machine. It drives, 1200Amp/1600V SKT1200/16E hockey puck SCR.
Main point of the design is driver transformer. Also attached image of actual implementation of 3-phase driver card and Snubber card.

Hope this helps,
sam_des

[zab]

Thanks for the informations . I was little busy in other problems So still I can not make it.Information collection by yet.

  What was the size of the core and turn ratio you use for drive transformer? I saw that you used three winding for it while you are using half wave rectifier at out put.Is that mean you are not using in fly-back mode?

 What type of zero crossing you select for your project?
and are the problem you faced during prototyping that?

One more thing what should be the width of trigger pulse?
I previously used the continuous pulse upto next zero crossing to ensure the triggering of SCR. Because right after the zero crossing voltage are low to sustain holding current .