Hi. Thank you for your useful tutorials. Now I want to use uc3845 as controller but I don't know how to enter the gain of the input circuitry in the simulation program and where. I should consider it in the voltage controlled current source or somewhere else?
Many of the PWM is based on the UC1843 series of PWM. The comparator is ussually limited to around 1V. So to keep the same gain, I try to keep the error amp output at 1V
How would you approach this if you were to have a variable output voltage flyback converter? That is, effectively squeezing in a potentiometer between the output voltage sampling resistors and feeding the inverting input of the op amp with the wiper.
Hi Jacob. Good question. If you were to replace the two feedback resistors with a potentiameter, it would make is a variable power supply. But you would have a limit on how low of a voltage you can go or adjust down. For example, if the Vreference voltage that is connected to the noninverting side of the error is 2.5V, then the lowest voltage the supply could be set it would be 2.5V. With this configuration, you could not adjust lower than the Vreference voltage. Hope this makes sense. Rbolanos
@@RobertBolanos Thank you for your response. The design i have in mind is a DCM converter using a UC3844 in current mode. Input = 24VDC. Output = ca 30-400V, 50W. I've seen another design doing this kind of thing with the same chip, but with no feedback components at all at the error amp, which made me a little hesitant.
Hi! Congratulations! It is impressive the way you conduct the tutorial. You match the theory with the practice very well. I learned and understood mainly the closed loop gain that you showed on part 8 - “How to Measure and Set the Control Loop”. I supposed that the PWM frequency is considered very high to influence the closed loop response. Is that right? I have only one question, if you can explain it to me, I will be grateful. What happens with the stability of closed loop if the load resistance (R12 24 Ohm) increases, say, then times? Can the flyback turn instable? Sorry! English is not my mother language. Thank you.
Hi Mr Bolanos, thanks for this tutorial. I am designing a power supply now. I have a few questions for you: (1) Please, what is the prototype board you are using called and what are the metallic connectors you are connecting components to the board with called, I'll like to have such. (2) Is it okay to have an offline SMPS without earth connection? (3) Assuming that (2) is yes, then I have a project that will require very low power, probably less than 5W. I am thinking of using an offline flyback with no earth connection but I can't seem to wrap my head around dealing with common-mode noise filtering in this case. Please what are the things I'm supposed to consider as regards common-mode noise and absence of earth connection.
at 42min26sec : is it a breadboard? what is the name of the adapter(made of copper like) you use to connect easily your components(like that cap you show) ? (if you have a link to a online shop, it would be perfect too ^^)
Dear Robert l would like to measure frequency response of the output stage in case of isolated flyback. Where I should connect channel A of cleverscope in this case? So far: I shorted primary and secondary ground. On my primary side I have a pull-up resistor connected to collector of optocoupler and this collector is connected to err. amp. output. My channel A is exactly connected to output of err amp. I decided to ask because power stage response measured in this way has strange behaviour. I see higher DC gain when I decrease input voltage. I expected that flat part of output stage response will move down. Please advise.
Hi Kamil, To measure the output stage or the output filter response, you connect one of the channels to the output voltage that you are regulating. Let say that you have a 5V output voltage. When then, you connect one of the change to this output. The other channel you will connect to the output of the error amp. The output of the error amp is connected to the input of the PWM. Send me a email to rbola35618@aol.com and I will send you a schematic of where to hook up. Robert
The historical reason for having a voltage regulated power supply, as compared to a unregulated power supply, was the output impedance effect at lower frequencies, where you had to have a very large capacitor and/or a very large (low % regulation) transformer and you would still not get the same low frequency benefits as you would using a voltage regulated supply. The focus for this realization was, in part, due to audio power amplifiers in driving a loudspeaker where we want operation down to 20Hz or slightly less. So, the power supply to the power amplifier output stage must exhibit a low output impedance, preferably, all the way to DC (0 Hz), so the power output stage will not modulate its supply voltage (low AC impedance source at the supply rails). Feedback provides this effect while avoiding the size and cost of very large capacitors and transformers. If you look at traditional audio power amp design one small signal analysis assumption is the power rails act as an AC short (low impedance). So, in other words, power supply regulation results in very low power supply AC output impedance.
Marvellous Sir!...I just love the way you connect simulation in the lab to a prototype, to the bode plot on clever scope!
Appreciate your efforts!
Thank you for the compliment. I am glad you found it usefull. Robert
Thank you, I saved this video some time ago and today I sat down and watched it. its very nicely explained.
+Yaghiyah Brenner Thank you Yaghiyah for the feedback. I am glad you found the video useful. Robert
Thank you Aaron for your kind comments!
Robert
Thanks for the great video.@22.36 how you are making sure that US 741 is giving you 1v? What is the math involved in that?
Is there a reason you start off with the same value resistor for both the bottom Vout sense resistor(R5) and Rint(R6)?
Thank you for this helpful and detailed video!
Hi. Thank you for your useful tutorials. Now I want to use uc3845 as controller but I don't know how to enter the gain of the input circuitry in the simulation program and where. I should consider it in the voltage controlled current source or somewhere else?
Yes that is correct. If you know the output current the just divide by 1 volt which is what want .
Many of the PWM is based on the UC1843 series of PWM. The comparator is ussually limited to around 1V. So to keep the same gain, I try to keep the error amp output at 1V
How would you approach this if you were to have a variable output voltage flyback converter? That is, effectively squeezing in a potentiometer between the output voltage sampling resistors and feeding the inverting input of the op amp with the wiper.
Hi Jacob. Good question. If you were to replace the two feedback resistors with a potentiameter, it would make is a variable power supply. But you would have a limit on how low of a voltage you can go or adjust down. For example, if the Vreference voltage that is connected to the noninverting side of the error is 2.5V, then the lowest voltage the supply could be set it would be 2.5V. With this configuration, you could not adjust lower than the Vreference voltage. Hope this makes sense. Rbolanos
@@RobertBolanos Thank you for your response. The design i have in mind is a DCM converter using a UC3844 in current mode. Input = 24VDC. Output = ca 30-400V, 50W.
I've seen another design doing this kind of thing with the same chip, but with no feedback components at all at the error amp, which made me a little hesitant.
Hi!
Congratulations! It is impressive the way you conduct the tutorial.
You match the theory with the practice very well. I learned and understood mainly the closed loop gain that you showed on part 8 - “How to Measure and Set the Control Loop”. I supposed that the PWM frequency is considered very high to influence the closed loop response. Is that right? I have only one question, if you can explain it to me, I will be grateful. What happens with the stability of closed loop if the load resistance (R12 24 Ohm) increases, say, then times? Can the flyback turn instable?
Sorry! English is not my mother language.
Thank you.
Hi Mr Bolanos, thanks for this tutorial. I am designing a power supply now. I have a few questions for you:
(1) Please, what is the prototype board you are using called and what are the metallic connectors you are connecting components to the board with called, I'll like to have such.
(2) Is it okay to have an offline SMPS without earth connection?
(3) Assuming that (2) is yes, then I have a project that will require very low power, probably less than 5W. I am thinking of using an offline flyback with no earth connection but I can't seem to wrap my head around dealing with common-mode noise filtering in this case. Please what are the things I'm supposed to consider as regards common-mode noise and absence of earth connection.
at 42min26sec : is it a breadboard? what is the name of the adapter(made of copper like) you use to connect easily your components(like that cap you show) ?
(if you have a link to a online shop, it would be perfect too ^^)
Dear Robert
l would like to measure frequency response of the output stage in case of isolated flyback. Where I should connect channel A of cleverscope in this case?
So far:
I shorted primary and secondary ground. On my primary side I have a pull-up resistor connected to collector of optocoupler and this collector is connected to err. amp. output. My channel A is exactly connected to output of err amp.
I decided to ask because power stage response measured in this way has strange behaviour. I see higher DC gain when I decrease input voltage. I expected that flat part of output stage response will move down. Please advise.
Hi Kamil,
To measure the output stage or the output filter response, you connect one of the channels to the output voltage that you are regulating. Let say that you have a 5V output voltage. When then, you connect one of the change to this output.
The other channel you will connect to the output of the error amp. The output of the error amp is connected to the input of the PWM. Send me a email to rbola35618@aol.com and I will send you a schematic of where to hook up.
Robert
The historical reason for having a voltage regulated power supply, as compared to a unregulated power supply, was the output impedance effect at lower frequencies, where you had to have a very large capacitor and/or a very large (low % regulation) transformer and you would still not get the same low frequency benefits as you would using a voltage regulated supply.
The focus for this realization was, in part, due to audio power amplifiers in driving a loudspeaker where we want operation down to 20Hz or slightly less. So, the power supply to the power amplifier output stage must exhibit a low output impedance, preferably, all the way to DC (0 Hz), so the power output stage will not modulate its supply voltage (low AC impedance source at the supply rails). Feedback provides this effect while avoiding the size and cost of very large capacitors and transformers. If you look at traditional audio power amp design one small signal analysis assumption is the power rails act as an AC short (low impedance).
So, in other words, power supply regulation results in very low power supply AC output impedance.
Well said Wind Ward! Thank you for your insight! All true facts!
Who are you Wind Ward? You are obviouly an older engineer like me. Are you still practicing engineer or retired? Email me at rbola35618@aol.com
How do you calculate the ESR ?
Check out our webinar on loop gain measurements: th-cam.com/video/lMMB3FU5HrY/w-d-xo.html
So the Crolf, Rin+ and Cin+ will cancel the pole,zero on the out put filter and that on error amplifier to have the open loop total
Yes that is correct. Check this earlier video on the Zero-Pole cancelation.th-cam.com/video/2doS4y9KsIM/w-d-xo.html
I do not see how RS and R1 are in parallel. THey look in series .
You do the Thevenin transformation
Can you please explain why inductor is drawn as constant current source? Your explanation is not clear..
Because as you have a constant current source it will force a constant current through the inductor making zero voltage across it, which is a short