Author Topic: Power in transient circuit  (Read 3421 times)

0 Members and 1 Guest are viewing this topic.

Offline raff5184Topic starter

  • Regular Contributor
  • *
  • Posts: 213
  • Country: us
Power in transient circuit
« on: April 06, 2017, 12:20:01 am »
Hi,
how should one think in terms of power if the circuit is not in a stationary state?
I have a circuit to charge a large capacitor, it takes entire minutes to charge it, but I want to be able to calculate the power of my supply chain for different voltages, or give an indication of what is the average power that is being delivered to the load. How is this done? Should I use a standard load for example a 50 ohm resistor without the capacitor?

Thank you
« Last Edit: April 06, 2017, 12:22:38 am by raff5184 »
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Power in transient circuit
« Reply #1 on: April 06, 2017, 12:37:40 am »
One should avoid thinking in terms of "RMS", which is only meaningful with some time to average over (root mean square).  For periodic signals, it's a cycle.  For non-periodic signals, it's not meaningful.

Instead, you'll most likely be working with instantaneous power.  Which is just fine, nothing wrong with that.  Perfectly real and easy to calculate (P = V*I no matter what the time scale).

For a capacitor charging circuit, you're likely limited more by (input or output) current, than by power, but you can do a constant-power charge too.

What are you doing?  What does 50 ohms have to do with this?

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 13243
Re: Power in transient circuit
« Reply #2 on: April 06, 2017, 12:49:49 am »
Power is the instantaneous product of voltage across and current through a circuit element.  Inductors and capacitors can store power applied for a time as energy, so when considering transient conditions, you need to distinguish between power that became stored energy, (or stored energy that became power), and power converted to heat (so-called ohmic losses).  If you are calculating or simulating an AC circuit or a DC switching circuit, its important to calculate or measure the power over either N whole cycles (so it has returned to the same state) or very large number of cycles, so that the power used to change the stored energy is a negligible part of the total.

If you connect your charging circuit to a 50R load it will almost certainly NOT deliver or use the same power as it would while charging the capacitor.  Depending on the level of detail and accuracy you need (and your budget), you could instrument your circuit with current and voltage probes and use a DSO that can do waveform arithmetic, or a data logger to measure the actual power while charging the capacitor or if you have enough component data you could model it in a SPICE simulator.


RMS is not meaningful for power!  P=V*I or P=V2/R or P=I2*R all contain a square or equivalent product.  Just integrate over time and divide by the time interval for Average power.  If you have a RMS voltage and a RMS current and you *KNOW* they are always in phase, you can multiply them to get the Average Power.  If they are not in phase the true power will be less and may even be zero.
 

Offline raff5184Topic starter

  • Regular Contributor
  • *
  • Posts: 213
  • Country: us
Re: Power in transient circuit
« Reply #3 on: April 06, 2017, 12:56:26 am »
I am building a wireless charging system, so I have a rectified DC signal applied to the capacitor, and I want to be able to evaluate the performance of my system, or in other words what are voltage and current "coming out" of the rectifier.

The 50 ohm have nothing to do with this, I was trying to figure out a method to evaluate the performance of my system somehow "independently" of the value of the capacitor I want to charge
 

Offline raff5184Topic starter

  • Regular Contributor
  • *
  • Posts: 213
  • Country: us
Re: Power in transient circuit
« Reply #4 on: April 06, 2017, 01:05:28 am »
Depending on the level of detail and accuracy you need (and your budget), you could instrument your circuit with current and voltage probes and use a DSO that can do waveform arithmetic
Yes, I have lab instrumentation, professional DSOs and multimeters. I'll use an oscilloscope
 

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 13243
Re: Power in transient circuit
« Reply #5 on: April 06, 2017, 01:10:04 am »
Do you have a DC current probe, or can you stick a 1R (or lower) resistor between the -ve side of your charging circuit and the capacitor so you can scope the voltage across it to get the current and simultaneously scope the voltage across the cap, do a slow one-shot acquisition and get your scope to show you the product of the two channels?  That should be repeatable if you always start with the cap discharged.

Otherwise you'll need an active constant voltage load to simulate an infinite capacitor charged to a particular initial voltage.  You may be able to build something suitable with a TL431 adjustable shunt regulator chip, as long as the minimum current is always over 1mA.   Use TLV431 for lower voltages/currents.  For high power levels, you may need an external Sziklai pair acting as a PNP emitter follower to boost its current capability. Put a *BIG* heatsink on the NPN power transistor!
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Power in transient circuit
« Reply #6 on: April 06, 2017, 01:17:04 am »
Oh, well in that case you simply want the V(I) curve of the supply.  For well filtered DC, Vdc * Idc = P.

Ideally, you'd do it with a variable resistance or load, so you can find the maximum power point, which will probably be a resistance (ratio V/I), that depends on distance, and what the LC parameters are (matching).

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline Conrad Hoffman

  • Super Contributor
  • ***
  • Posts: 2092
  • Country: us
    • The Messy Basement
Re: Power in transient circuit
« Reply #7 on: April 06, 2017, 03:40:54 am »
Think about this problem- http://www.mogami.com/e/puzzle/pzl-03.html

Correct power measurement has confounded a lot of people, particularly the perpetual motion and free energy folks. You'll know your doing it right when the answer doesn't violate conservation of energy.
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Power in transient circuit
« Reply #8 on: April 06, 2017, 04:15:34 am »
Energy conservation problems, analyzed without E and M and loss (internal resistance or external radiation), are sure to cause confusion. ;)

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 13243
Re: Power in transient circuit
« Reply #9 on: April 06, 2017, 08:17:26 am »
Think about this problem- http://www.mogami.com/e/puzzle/pzl-03.html

Correct power measurement has confounded a lot of people, particularly the perpetual motion and free energy folks. You'll know your doing it right when the answer doesn't violate conservation of energy.
I dislike puzzle pages that lie to you:
Quote from: pzl03.html
Surprisingly as soon as switch S2 is closed, the total sum of the energy in this space becames half!

That is untrue - it ignores the propagation velocity of the cable, which in real life will be some fraction of C, and will be C if the only dielectric is vacuum. 

If you assume that SW2 and its connections is fully screened, it closes instantaneously without sparking, the cables have screened non-shorting end caps so it cannot radiate and that the cables are lossless (i.e. superconducting with a vacuum dielectric), it will oscillate forever!
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Power in transient circuit
« Reply #10 on: April 06, 2017, 04:26:08 pm »
If you assume that SW2 and its connections is fully screened, it closes instantaneously without sparking, the cables have screened non-shorting end caps so it cannot radiate and that the cables are lossless (i.e. superconducting with a vacuum dielectric), it will oscillate forever!

Even then, the energy will eventually tunnel out (which is to say, no shield is perfect, even a superconducting one).  Or if we're using real superconductors (not ideal ones), there is real AC loss, small but present nonetheless.  Axial resonators, for LINAC duty, typically have a Q in the 3e8 range -- better than quartz crystal, but not infinite!

Again, analyzing a system without full consideration of its properties, leads to confusing ones' self over edge cases that don't actually exist. :)

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf