PWM converter: Difference between revisions

From apm
Jump to navigation Jump to search
basic page
 
added linebreaks
Line 4: Line 4:
The main trick here is to cross the intermediate states between fully open and fully closed as fast as possible. <br>
The main trick here is to cross the intermediate states between fully open and fully closed as fast as possible. <br>


To elaborate a bit: Both  
To elaborate a bit: Both <br>
– in [[FAPP]] zero resistance state and  
– in [[FAPP]] zero resistance state and <br>
– in [[FAPP]] infinite resistance state  
– in [[FAPP]] infinite resistance state <br>
the power lost on the switching transistor is [[FAPP]] zero.
the power lost on the switching transistor is [[FAPP]] zero. <br>
But in-between these two states there is a level of resistance across the switching transistor that leads to maximal power lost on it.
But in-between these two states there is a level of resistance across the switching transistor that leads to maximal power lost on it. <br>
That is called impedance matching. We want to do the exact opposite here. Maximal "impedance mismatching" so to say.
That is called impedance matching. We want to do the exact opposite here. Maximal "impedance mismatching" so to say. <br>
   
   
It's not avoidable to cross the matched impedance resistance across the switching transistor but  
It's not avoidable to cross the matched impedance resistance across the switching transistor but <br>
one can do so as fast as possible and as rarely as possible (both withing limits),  
one can do so as fast as possible and as rarely as possible (both withing limits), <br>
such that on average most of the time one is in the zero losses state.
such that on average most of the time one is in the zero losses state. <br>


== Application cases ==
== Application cases ==
Line 23: Line 23:
== Novel direct mechanical analogy to well established existing electrical systems ==
== Novel direct mechanical analogy to well established existing electrical systems ==


The idea here is to convert PWM converter circuits based on discrete elements  
The idea here is to convert PWM converter circuits based on discrete elements <br>
more or less 1:1 over to (macro/meso/micro/nano)mechanical circuits using the [[mechanical-electrical analogies]].
more or less 1:1 over to (macro/meso/micro/nano)mechanical circuits using the [[mechanical-electrical analogies]]. <br>
It might become sensible to do so at the nanoscale once we have [[advanced APM]] capabilities.
It might become sensible to do so at the nanoscale once we have [[advanced APM]] capabilities. <br>


From above there are in direct analogy the tasks of:
From above there are in direct analogy the tasks of:
Line 32: Line 32:


The first task (constant force sources) can be done by just gears. Or linearly for small displacements by levers. But the second one not. <br>
The first task (constant force sources) can be done by just gears. Or linearly for small displacements by levers. But the second one not. <br>
For the second task (constant speed sources) there are escapements in clocks  
For the second task (constant speed sources) there are escapements in clocks <br>
but these are only optimized for slow precisely timed release not for maximal energy efficiency (the ticking sounds are part of the losses).
but these are only optimized for slow precisely timed release not for maximal energy efficiency (the ticking sounds are part of the losses). <br>


While sort of straightforward to do as of time of writing (2025-01)  
While sort of straightforward to do as of time of writing (2025-01) <br>
it seem this mechanical analogy has not yet been experimentally demonstrated at any scale.
it seem this mechanical analogy has not yet been experimentally demonstrated at any scale. <br>


== Related ==
== Related ==

Revision as of 12:29, 28 February 2025


The main point here is energy efficiency.
The main trick here is to cross the intermediate states between fully open and fully closed as fast as possible.

To elaborate a bit: Both
– in FAPP zero resistance state and
– in FAPP infinite resistance state
the power lost on the switching transistor is FAPP zero.
But in-between these two states there is a level of resistance across the switching transistor that leads to maximal power lost on it.
That is called impedance matching. We want to do the exact opposite here. Maximal "impedance mismatching" so to say.

It's not avoidable to cross the matched impedance resistance across the switching transistor but
one can do so as fast as possible and as rarely as possible (both withing limits),
such that on average most of the time one is in the zero losses state.

Application cases

There are several application cases. Mainly:

  • Decreasing or increasing the voltage but retaining ideal voltage source behavior.
  • Approximating an ideal current source given an ideal voltage source.

Novel direct mechanical analogy to well established existing electrical systems

The idea here is to convert PWM converter circuits based on discrete elements
more or less 1:1 over to (macro/meso/micro/nano)mechanical circuits using the mechanical-electrical analogies.
It might become sensible to do so at the nanoscale once we have advanced APM capabilities.

From above there are in direct analogy the tasks of:

  • Decreasing or increasing the force/torque but retaining ideal force/torque source behavior.
  • Approximating an (linear/angular) speed source given an ideal force source.

The first task (constant force sources) can be done by just gears. Or linearly for small displacements by levers. But the second one not.
For the second task (constant speed sources) there are escapements in clocks
but these are only optimized for slow precisely timed release not for maximal energy efficiency (the ticking sounds are part of the losses).

While sort of straightforward to do as of time of writing (2025-01)
it seem this mechanical analogy has not yet been experimentally demonstrated at any scale.

Related

External links