Group Actions
Engagement
  • Author Author: peteroakes
  • Date Created: Date Created
  • Views 1875 views
  • Likes 2 likes
  • Comments 7 comments
Related
Recommended

The Modular Bench Power Supply ++, Integration theory and practical

peteroakes

Back to the Main post The Modular Bench Power Supply ++, The Essential DIY Build for Every EE Student and Old Timer alike...

 

So now we will look at how to use integration to control the power stage of a High Power MOSFET and how the response time can be tuned to suite the characturists of your power supply enabling you to adjust the transient response times etc

 

To be up front about this, I did get some inspiration from studying 30 year old agilent power supply schematics and those of other manufactures  So im not claiming the basic ideas are all mine. I already knew what I wanted to do like floating the control on the power rails, N Channel Power Mosfets etc. the studying helped me to understand how this could be acheived and to learn from good design. I think we can all afford to learn from the hard work of others. It can take exaustive amounts of time and effort to fully inderstand an electronic design and from what I have learned, it was worth every minute

 

None of my schematics are from those that I studied, they have all been re-worked for my needs and the components I have available. I guess it is like learning to paint, you study the work of old masters in the area of interest and learn from their endevours.

 

here is the schematic I ended up with for this part of the power supply

image

and Power Control

image

 

Again all of the descriptions and a practical demo are within the video

  • Thanks you Peter for all the tutorial. It have been really interesting and helpfull, and it have engaged me to start my own PSU. I don't quite very well undestand the operation of the V control and I control points againts Vgate that you show in the second picture. As far as I undestand, the integrator goes from -12 to 12 depending on the output of the previous state of sensing. Take it for example that I have a setpoint of 3V and I am sensing 3V on the output of my PSU. The output of the integrator will this case do nothing and remain at 0V, turning off the mosfet because not voltage is being put into the gate (all is going to the integrator because of the diode's direction).

    Now, let's imagine that I have a setpoint of 3V and I am sensign 5V (-5V in differential amp), so the summing junction will be -2V--> It will generate a decreasing ramp on the integrator output until reaching the -12V. As this voltage is lower than the 6V that we have in the Vgate path, this will turn off the Mosfet.

    On the other hand, if I would have a  setpoint of 5V and I am sensign 3V (-3V in differential amp), the summing junction will be 2V--> It will generate an increasing ramp on the integrator output until reaching the 12V. When it exceeds the 6V, because the direction of the diode, it will allow the mosfet to turn it on, allowing the 6 V on the Vgate path to go into its gate. 

    I don't know if this is correct, I was not pretty sure before starting to write, but while writing it has been usefull to me to clear my ideas (sorry ;) ). In this case, we will all the time be working with the mosfet in a saturation mode, where it is either fully ON or fully OFF right? The question is--> ¿Is this a switching mode PSU or a linear mode PSU? I think it is linear because based on what i have understood and taking the latest example, where I needed to increment the output voltage of my PSU from 3V to 5V, once it has reached those 5V, is going to be 0V into the integral amp's input, generating no ramp in its output, so keeping its output in 12V until another setpoint is set?

    What is the difference then between a setpoint of 4V or 5V, because the integral output is always 12 V or -12 V, I don't very well understand how the Vds of the Mosfet changes in order to put the PSU's output higher or lower. Because if we work in saturation mode, is always going to be the same voltage in the source right?

    I don't know if i am right, and if it is a switching mode PSU, what is its frequency?

    I am sorry for my explanations and my english. I hope I have explain myself clearly enough to understand my questions. Let me know if it is not like that

    Another question, what is the purpose of the analogue 0V that you have in common on all your amps in your final schematic of this tutorial?, and where do you take it from?

    Thank you so much

    Nagore

  • in reply to clem57

    Clem Martins wrote:

     

    portmanteau

     

    A large leather suitcase that opens into two hinged compartments.

    portmanteau word: a word formed by combining two other words.  This usage was coined by Lewis Carroll in Through the Looking-Glass:

    Humpty-Dumpty said:

     

    'Well, “slithy” means “lithe and slimy.” “Lithe” is the same as “active”. You see it’s like a portmanteau – there are two meanings packed up into one word.'

  • in reply to johnbeetem

    Nope, I was referring to the company that produced the 30 year old schematics as Agilent rather than HP, of course by 2020 they will change from Keysight to just Sight image

  • in reply to johnbeetem

    portmanteau

     

    A large leather suitcase that opens into two hinged compartments.

  • in reply to peteroakes

    Peter Oakes wrote:

     

    OK... You got me

     

    HP 

     

    They're often now referred to as Agilent / HP and is the same company still (Well Almost)

    Peter, you missed a name change.  The test and measurement part of Agilent is now Keysight Technologies.  The rest of Agilent is now primarily biotech or something like that.

     

    I think Keysight is a silly name.  According to Wikipedia, "the name is a portmanteau of key and insight, and was chosen to convey that the company "unlocks" critical or key insights".  OK, that explains it but I still think it's a silly name and a way-too-cute explanation for it.