I believe the circuit should work. It will not like to start a motor. The stall current (startup current) of a motor is much higher than the running current. The supply will current limit and maybe never get a motor started. In current limit mode the MOSFET will run hot.Either this video is fake or there is a problem with my work and I can't find my problem
Thanks, I will try it with something other than the DC motorI believe the circuit should work. It will not like to start a motor. The stall current (startup current) of a motor is much higher than the running current. The supply will current limit and maybe never get a motor started. In current limit mode the MOSFET will run hot.
Hello friendsThanks, I will try it with something other than the DC motor
YesLamps don't make good test loads, because the inrush current is 10x the running current, and the PSU will have a hard time starting up.
Remind us again what your target output voltage/current was, let's see some waveforms.
I have used MBR 30100 diode.So, your input supply is 27V with a maximum output current of 15A?
Don't forget that the peak voltages on the MOSFET and diodes are Vin+Vout in a SEPIC converter.
The problem is not the coilSo, your input supply is 27V with a maximum output current of 15A?
Don't forget that the peak voltages on the MOSFET and diodes are Vin+Vout in a SEPIC converter.
I could not understand youLet's see some waveforms.
Maybe I did not explain correctlyWhat?
It is very important that the transformer is 1:1.
10:15 is not good.
You need to wind the transformer pulling two wires at the same time. Do not wind two separate windings.
View attachment 273410
View attachment 273409
Not like this:
View attachment 273411
YesSeveral times "we" have asked about the phasing on the transformer. Without an oscilloscope it is hard to trouble shoot.
The wire with 12V or 30V on it, and the wire on ground should be on the same end of the transformer.
Thank youC8, C9; Under light loads, you can operate without these two caps.
I was able to solve the problem to some extent, of course I think.Let's see some waveforms.
Can you highlight the part of the datasheet that you do not understand? MOSFETs can get hot for a couple of reasons which include but are not limited to:I was able to solve the problem to some extent, of course I think.
I set the resistance of R10 below the value of 0.05.
Now the circuit works, there are only a few problems
For example, we have 6 volts without load, and when the load is connected to it, the voltage first becomes zero and within 2-3 seconds, the voltage reaches about 5.5 volts.
Another problem is that if you remember, I asked about the input of the circuit, how can I give about 30 volts to the input
And the answer was that with an 18V zener and a 680 ohm resistor on pin 7 of the IC, the result can be reached
But now, when the circuit input is more than 18 volts, the MOSFET gets extremely hot
I would be grateful if you could explain to me in a simple way what the permissible voltage between the MOSFET pins should be during use.
I read the MOSFET datasheet and did not understand it
Nothing wrong with separate windings on the same core for SEPICs and Cuks (but the green core is ferrite, and it won't work without a gap). SEPICs and Cuks don't need coupled inductors to work.What?
It is very important that the transformer is 1:1.
10:15 is not good.
You need to wind the transformer pulling two wires at the same time. Do not wind two separate windings.
View attachment 273410
View attachment 273409
Not like this:
View attachment 273411
UC3843 has a good enough gate-driver circuit for these not to be a problem, provided that the supply voltage is high enough (and UC3843 has a UVLO so won't even produce an output at low voltages)Can you highlight the part of the datasheet that you do not understand? MOSFETs can get hot for a couple of reasons which include but are not limited to:
There may be other reasons for your problem, but those are two of the big ones. I don't see the schematic in post #54 addressing either of those two concerns.
- Inadequate Gate to Source voltage. The gate threshold voltage is the gate to source voltage where the device BEGINS to conduct. In this application you want to turn the gate on very quickly to a voltage that could be much higher than your input voltage. The datasheet should tell what fully on means. It is where rds(on) is a minimum.
- SLOW turn on and turn off. During this phase of operation there is Miller effect which arises because of how the MOSFET channel interacts with the node capacitances on the device. Normally you would use a gate driver that can move charge on to and off of the gate just as fast as electronically possible. Any delay in this process because of LONG RC time constants will allow for substantial heating.
I certainly agree that waveforms will be helpful. So if the gate driver is adequate for charging and discharging the gate in an appropriate amount of time, what is the purpose of R8(6.8KΩ)?UC3843 has a good enough gate-driver circuit for these not to be a problem, provided that the supply voltage is high enough (and UC3843 has a UVLO so won't even produce an output at low voltages)
We're not going to resolve this until the TS posts some waveforms - anything else is just speculation