I have taken a L6565 smps design and altered it to be driven by a PIC.
The PIC fires the mosfet driving the transformer and also switches off the mosfet if an over current event occurrs.
I am finding the overcurrent event happens at the very first mosfet pulse at power up. So the PIC turns off the mosfet and flashes the error LED.
The circuit should work fine.
I have tried ignoring the over current event but this just blows the fuse.
The only part that might not be right is the transformer.
I have wound on 50 turns on the primary and 13 on the secondary.
There wasnt enough room on the transformer to wind on 50 turns horizontally so I put on 25 turns then went back along the previous 25 turns with the final 25 turns. I thought this would be OK.
The secondary is simply 13 turns wound horizontally.
Any clues as to why the over current event occurrs would be helpful.
The PIC fires the mosfet driving the transformer and also switches off the mosfet if an over current event occurrs.
I am finding the overcurrent event happens at the very first mosfet pulse at power up. So the PIC turns off the mosfet and flashes the error LED.
The circuit should work fine.
I have tried ignoring the over current event but this just blows the fuse.
The only part that might not be right is the transformer.
I have wound on 50 turns on the primary and 13 on the secondary.
There wasnt enough room on the transformer to wind on 50 turns horizontally so I put on 25 turns then went back along the previous 25 turns with the final 25 turns. I thought this would be OK.
The secondary is simply 13 turns wound horizontally.
Any clues as to why the over current event occurrs would be helpful.
Did you gap the transformer? If not, it will saturate right away.
No I didnt.
Do you mean the ferrite cores need a gap ?
I tried it with a gap tonight and stupidly just parted the cores slightly. Of course Iswitched it on and the two halves were attracted and blew the fuse and the mosfet. I have no mosfets spare so will have to order some more.
How do you normally cut down the cores ? Do you put something of 1mm between them or physically cut them down with a hack saw ?
How do you normally cut down the cores ? Do you put something of 1mm between them or physically cut them down with a hack saw ?
You need to do some background reading on how flyback SMPS work. I would stick with the basic L6565 circuit and get it working first before trying to do anything fancy with a PIC.Search the ST web site - they probably have a design spreadsheet somewhere that will allow you to design a proper transformer.
And it still needs a gap, too -not just any gap. Do you have any spec goals for this supply? Post them, (AC input voltage range, Vout, Iout, proposed operating frequency),along with the core you're using, and maybe I can help some.
You should also realize that the L6565 is not your usual fixed-frequency PWM controller. Since it's quasi-resonant, it operates in variable frequency borderline continuous mode, kinda like a self-oscillating RCC flyback with some extra smarts. Your choice of primary inductance will affect the operating frequency.
You should also realize that the L6565 is not your usual fixed-frequency PWM controller. Since it's quasi-resonant, it operates in variable frequency borderline continuous mode, kinda like a self-oscillating RCC flyback with some extra smarts. Your choice of primary inductance will affect the operating frequency.
And it still needs a gap, too -not just any gap. Do you have any spec goals for this supply? Post them, (AC input voltage range, Vout, Iout, proposed operating frequency),along with the core you're using, and maybe I can help some.
.
I am trying to get around 5 amps at 40 volts out.
Input is 240VAC mains. I am running the PIC at 50KHz with 50% duty cycle.
I am using an ETD54 with n97 core, 50 turns primary and 13 turns secondary.
I am getting a pulse out of the transformer with a lot of ringing.
Is there anyway to stop this ?
I think the ringing is interfering with pulsees close together.
http://www.ckp-railways.talktalk.net/00.jpg
Is there anyway to stop this ?
I think the ringing is interfering with pulsees close together.
http://www.ckp-railways.talktalk.net/00.jpg
Is the current limit circuit peak/pulse detect or level detect?
Is the limit pin cap bypassed?
What is your topology?
I have a resistor in series with the triac and measure the voltage across it.
The PIC looks at the over current pin.
Post your circuit.
I have used the L6565 example circuit (the first one in the datasheet) and replaced the L6565 with a PIC.
My screenshots dont come out well.
Here is the latest waveform.
The low part is the transformer primary charging up.
The high part is the secondary passing current through a diode into a smoothing capacitor.
Then there is a large gap which if I try to put out another pulse I get an over current event. Still needs a bit of thought as to why.
http://www.ckp-railways.talktalk.net/000.jpg
The low part is the transformer primary charging up.
The high part is the secondary passing current through a diode into a smoothing capacitor.
Then there is a large gap which if I try to put out another pulse I get an over current event. Still needs a bit of thought as to why.
http://www.ckp-railways.talktalk.net/000.jpg
That waveform looks good to me.
you should easily be able to double the frequency of pulses with that setup.
Do you have any load connected.?
use bits of paper between the outside legs of the cores, and watch the differences with different gaps. then do the same with different loads.
you should easily be able to double the frequency of pulses with that setup.
Do you have any load connected.?
use bits of paper between the outside legs of the cores, and watch the differences with different gaps. then do the same with different loads.
That waveform looks good to me.
you should easily be able to double the frequency of pulses with that setup.
Do you have any load connected.?
use bits of paper between the outside legs of the cores, and watch the differences with different gaps. then do the same with different loads.
I have a 25R resistor connected.
I am using an etd34 N97 core with 2mm gap.
As soon as I decrease the spacing between pulses I get an over current detect to the PIC.
I read a bit on the L6565 and it waits for transformer current to go to zero before it outputs another pulse. I am just using a timing loop in the software after looking on the scope.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Amplifiers
- Power Supplies
- SMPS power up problems.