Need help in SCR power control circuit

HI, I am building a single phase full wave controlled rectification circuit. The concept is to control the output voltage by controlling the firing angle of thyristor. I am building a circuit from this webiste: **broken link removed**

There are 2 main circuit which is zero crossing detector circuit and firing angle control circuit.
Since this project output is a AC. So I had modify the circuit output to become a DC.
But I am not sure whether the circuit is correct or not. So hope guru and senior here can help me check it out. If wrong please guide me. thank you.
The attachment picture is mine modified circuit.

Still got 1 more question, This project he used an Atmel microcontroller, but if I change to PIC microcontroller should be no problem right???
Thank you

Hi,


1. You might need a resistor in series with the output of the opto to limit current through the opto and the gate.
2. Is that a DC meter there? You'd need a resistor across the meter to act as a decent load for the SCR or you may end up measuring the voltage due to the leakage of the SCR as well as the voltage when it is turned on.
3. Might also need a resistor across the output of the bridge rectifier to make sure the output goes close to zero volts so the SCR can turn off when required.
4. For better control you might be better off connecting the SCR in the ground lead of the bridge rectifier instead of the positive lead. This allows for better on/off control of the SCR, with the SCR cathode connected to ground.

Thanks for suggestion. I Had made some changes based on your suggestion, please have a look.
And because I want a variable DC output voltage, so I has use a DC motor as a load for demonstration, the motor spin faster when higher voltage.
Please help me check out this circuit, If no problem I am gonna to start the coding ^^.View attachment 60940

Hi,

Actually you should connect the SCR between the bottom lead of the motor and ground.
Also, since you are using a controller to drive it you can make the drive pulse short. You'll have to check the specs on the SCR you are using though to find out what the pulse width should be and also the minimum drive current to the gate. The shorter pulse width keeps the gate dissipation down.

It's also wise to connect a resistor from the gate to ground. Usually 1k should be good, but remember that the other resistor has to be able to drive the gate up to the required voltage as well as supply the required gate current too.

For the minimum current and voltage to trigger the gate I had found on some datasheet which is here..
This is the picture that highlighted.
View attachment 60987

But for the pulse width isn't it I have determine the time to trigger the gate and off the pulse?? something like delay???
Sorry if i am wrong ^^

Hi,

Oh ok, well you seem to be working with a very hefty SCR there, which would have a decent max average gate power rating too. That means you can probably drive it with 200ma continuous and not worry about the pulse duration. That should ensure turn on every time.

Usually the gate power is more important because the SCR packages are smaller. This one is quite a big one

When will you be ready to test this?

haha.. oic.. I am still not quite sure yet what kind of SCR I am going to use yet.
I am going to test it out after I had develop the code for the microcontroller.
I am planning to test this circuit using proteus software first before building the entire circuit out.
I am planning to use pic18f4520 as the microcontroller, but this circuit use an Atmel chip and don't really understand the code.
**broken link removed**
Can you help me explain some of it. I don't really understand those code really well.
U don't need to explain line by line, but just the concept on how the entire code is running.
Basically this is what I understand from the code but not really sure right or wrong.
Steps:
1. Enter the time delays
2. Make a interrupt when there is zero crossing detected.
3. In the interrupt program, read the time delay inserted,
4. delay the specific time and on the give a pulse to the SCR.
5. On the next interrupt, the pulse is off and repeat step 3 and 4.

Thank you.

Hi,


It sounds like you have the basics there already.

I would think you would trigger each half cycle though, so for any zero crossing (0 degrees or 180 degrees) you would do the very same thing, unless you only want half wave output. Full wave is better though.

The delay the uC waits is the half cycle time minus the required 'on' time.

It's also a good idea to build in some sort of 'over delay' protection, if you intend to fire at small conduction angles. This helps protect against accidentally triggering a bit too late, which instead of a small conduction angle (and small output voltage) you'd get a large conduction angle (and very high output voltage) which would defeat the required time delay setting.

Also, if you are using a positive trigger pulse you have to watch out what triac you pick. The triac you pick has to support a positive trigger with both polarities of MT1 and MT2. Some dont support both polarities with a positive trigger pulse, but most do with a negative trigger pulse.

Yea. I am going to do a full wave output.
But I do not have any idea on how to do the "over delay" protection. Can you provide some ideas??

Beside that, I think I am going to use a thyristor as the SCR. Because what mine lecturer taught me is that a triac normally used to control AC output.
Please correct me if I am wrong. (and may be mine lecturer is wrong too ^^)

One more things is that I am not really understand MT1 annd MT2. Can you please explain a bit about it??
Thank you ^^

mt1 mt2

I would be interested in info about MT1 and MT2 as well.

Hi again,


Im sorry, MT1 and MT2 refer to a 'triac', not an SCR. The SCR has anode and cathode.
If you do decide to use a triac however those notes will be of use to you.

The over delay protection means having some way to make sure that the delay can not be too long no matter what. For example, a half cycle of 60Hz is 8.333333ms, so you want to make sure that the delay can never be more than 8.333333ms minus the zero cross detect delay minus some safety margin, like 0.33ms or something like that. So you have to figure out how much delay you have in your zero cross technique and then subtract that from 8.333333ms and then subtract say 0.3ms for safety.
Lets say the zero cross delay time is 0.2ms, then that leaves you with about 8.1ms, and subtract 0.3ms from that and you get 7.8ms as your maximum delay. In the code you would simply do a check to make sure that the delay is not greater than 7.8ms, and if it is you set it explicitly to 7.8ms (whatever that comes out to in machine instruction cycles).

Not too difficult.

Oic.... I will try this out..
I will test the circuit out as soon as possible.. Hope everything go well.
I will keep u update all the time on the progress. ^^
Thank you...

Hi,

Thanks, i'd like to hear more about your results once you get up and running.
It's always nice to hear about how the project being discussed worked out in the final design. It's not only enjoyable just to hear about it, it's also sometimes informative.

Oh yeah, another thing nice to see is some photo snap shots of the board, circuit layout, etc., or anything else that seems interesting about it. Pictures help a lot. Scope pictures too when possible.

Hi, there are some question I want to ask... From the picture is some of the others people project. But I am wondering why the pulse that send from the microcontroller must go through an optocouple before going to the SCR???
View attachment 61921
View attachment 61922

Thank you ^^

Hi,

I dont see any peak gate current limiting resistor in that circuit.

The opto is used to isolate the low voltage DC supply from the mains AC power. Feeding the SCR gate directly would mean loss of isolation. With an opto or pulse transformer the low voltage part of the circuit (the control part) is isolated electrically from the mains AC power.

Not shown is the low voltage power supply line transformer, which would isolate the mains line from the rectifier diodes for the low voltage power supply.

I do believe that you are taking your chances with that circuit. -the SCR may not turn off under certain circumstances.
since you are going to use optical isolation anyway, the fix is to put a triac on the ac side of the bridge rectifier.