Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.
Too small, but one thing jumps out at me, you are returning the emitter of the feedback opto to the isolated side of the transformer, it should probably go to somewhere on the primary side?
It is small, but I can see all the essential elements of a full bridge power supply.
But I do have a couple of observations:
1) For the inrush limiter you are using a NTC thermistor. At full load, the losses of the thermistor may be too large. You may want to use a relay to bypass the thermistor after a couple of seconds.
2) The supply has positive and negative outputs. But you are taking the feedback from only the positive output. If the loads are reasonably balanced this may be no problem. But what happens if, for instance, the negative load is completely removed while the positive is at full load? An over voltage condition will occur. You may want to add some monitoring circuit on the negative load, to either shutdown everything or add a "bleed" load.
3) How about temperature? You may want to add a thermistor to sense the heatsink temperature and start up a fan or even shut everything down.
4) Fuses? I don't see any fuses.
5) Lastly and this is optional: I would put status LEDs. At the very least one LED on each output voltage. If the temperature sensing is added, another LED to alert that condition. Of course additional monitoring conditions (overcurrent, AC OK, etc) may be added with additional circuitry.
Yes i know but when i upload it to somewhere, it becomes impossible to read. You can save it in your pc and read the values. Actually, PFC is necessary but i have designed it for my friend who didnt want to have PFC before SMPS. Yes you re right, i fixed it.
It is small, but I can see all the essential elements of a full bridge power supply.
But I do have a couple of observations:
1) For the inrush limiter you are using a NTC thermistor. At full load, the losses of the thermistor may be too large. You may want to use a relay to bypass the thermistor after a couple of seconds.
2) The supply has positive and negative outputs. But you are taking the feedback from only the positive output. If the loads are reasonably balanced this may be no problem. But what happens if, for instance, the negative load is completely removed while the positive is at full load? An over voltage condition will occur. You may want to add some monitoring circuit on the negative load, to either shutdown everything or add a "bleed" load.
3) How about temperature? You may want to add a thermistor to sense the heatsink temperature and start up a fan or even shut everything down.
4) Fuses? I don't see any fuses.
5) Lastly and this is optional: I would put status LEDs. At the very least one LED on each output voltage. If the temperature sensing is added, another LED to alert that condition. Of course additional monitoring conditions (overcurrent, AC OK, etc) may be added with additional circuitry.
I thought to add bleed load but it may cause additional power loss. Therefore, i didnt add it. I really forgot to add fuses =) i will also add monitoring circuits and leds. Thnk u so much schmitt trigger.
You can use rar file for better quality schematic.
Thnx. I have also changed thermistor part as finally in schematic below. Now i will implement low voltage cutoff for UC3825 and mosfet drivers. I mean for LM350 voltage regulator. When AC line becomes about 180VAC, output of mosfet drivers and UC3825 should be shut down.
In proteus, i couldnt work this cut off circuit properly.
23.14VDC is rectified voltage of 16.36VAC. Primary of transformer is 180VAC. Turns ration is 220VAC:20VAC. So 180VAC:16.36VAC.
I want to add cut off circuit as i said before but this circuit does not work. - pin of LM358 is stable 5.1V, when + pin of LM358 reaches 5.1V or over, led turns on but relay also must be turn on. However, i still see about 12V at output of voltage regulator. By the way, assume that LM350 gives about 12V at output. There is no LM350 in proteus. So i just changed the name of LM7812 by LM350.
I have completed cut off circuit which shuts circuit down when AC line is 180VAC or 23.11VDC and 242VAC or 31.11VDC. I have simulated in proteus and it worked nice. I have also added hysteresis so that circuit activates itself above 24VDC and below 30.5VDC.
I don't understand what you are doing with M9 and M10.
I see that you would like to create a discrete AND circuit, but if the gates of the two mosfets are tied together..this wouldn't work, correct?
I don't understand what you are doing with M9 and M10.
I see that you would like to create a discrete AND circuit, but if the gates of the two mosfets are tied together..this wouldn't work, correct?
Actually i tried to make overvoltage and undervoltage protection for SMPS. Each mosfet is for that protection. Upper one works for undervoltage and lower one works for overvoltage. I simulated it in proteus it worked good but it does not mean that it will work in real life. I couldnt see the fault. Thnx again. I will fix it.
I am really confused that which topologies have a half frequency of their oscillator frequency.
My oscillator frequency is 100kHz and i thought that it has 50kHz switching frequency. Isnt that correct ?
In application note of UC3825, switching frequency is half of the oscillator frequency in push pull and half bridge but i dont know it for full bridge.
Schmitt trigger, do you know how to regulate both + and - rails with TL431 ? I am stuck in how to use TL431 and optocupler for both positive and negative rails. I saw a circuit it had two TL431 and optocouplers and they both share the feedback current just before PWM chip. Is that true ? I saw feedback circuits in both power supply cookbook and ON semiconductor document but when there is a negative rail, they both dont apply feedback for negative rail. They just apply multi output positive rails.
Do u think this feedback network is correct ? The aim as i know is to share feedback currents. I did it via calculation but i am not sure whether it is correct or not.
Cross regulation will be a tremendous problem whenever the loads are not fully matched and one of the inductors goes into discontinuous mode while the other remains in continuous mode. The transfer function varies as the inductors cross boundaries.
So no, if there is an opportunity for that scenario to occur, your proposed scheme will not work.
-Are the currents in both outputs equal? Always?
-Which of the two outputs is most important?
I believe there is a technique which utilizes a coupled inductor...but it has been a long while when I read that.
Cross regulation will be a tremendous problem whenever the loads are not fully matched and one of the inductors goes into discontinuous mode while the other remains in continuous mode. The transfer function varies as the inductors cross boundaries.
So no, if there is an opportunity for that scenario to occur, your proposed scheme will not work.
-Are the currents in both outputs equal? Always?
-Which of the two outputs is most important?
I believe there is a technique which utilizes a coupled inductor...but it has been a long while when I read that.
Yea, i was concerning about that issue. I just did some research and found that coupled inductor is just to wind the wires on the same core. I will use toroid and i read that bifilar winding is best one but there will be large currents like 14A full load current. Now i concern about how to isolate the two wires from each other. By the way, i guess bifilar winding is to wind the wires at the same time as parallel on the core. Not twisting them together before winding.
Now as i understood from u, i just need to use +70V for regulation. So -70V will be cross regulated via coupled inductor.
These two optocouplers are for overvoltage protection for outputs. When one of the output is 72V, ILIM becomes about 1.2V and shuts the IC down. I hope transients will not be problem for this protection. Otherwise, i need to use another one SCR and clamped overvoltage protection that is less efficient.
Make absolutely sure that, the coupled inductor is properly phased...in other words, that both the positive and negative currents enter the inductor at the dot (start of windings) for each winding.
Second, I would make the I_lim O'ring diodes from a dual package diode like the BAT54C to make sure the forward voltages match and track.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
