Author Topic: Ripple current calculation / frequency for psu design  (Read 7983 times)

0 Members and 1 Guest are viewing this topic.

Offline amateur_25Topic starter

  • Regular Contributor
  • *
  • Posts: 106
Ripple current calculation / frequency for psu design
« on: January 15, 2013, 09:08:46 pm »
Hi
I looking to understand how would an enginner know how much ripple current is likely to flow within a circuit without building it so
that when he comes to purchase capacitors he knows the capacitors ripple current rating is sufficiently high so it doesn't cause overheating of the
capacitor.

 Also say I had transformer feeding linear regulator, how does one know the frequency of the circuit in order to determine the regulators ripple
rejection rating? Surely you could work this out without having to build a prototype?
 

Online AndyC_772

  • Super Contributor
  • ***
  • Posts: 4223
  • Country: gb
  • Professional design engineer
    • Cawte Engineering | Reliable Electronics
Re: Ripple current calculation / frequency for psu design
« Reply #1 on: January 15, 2013, 09:33:49 pm »
Simulation works for me. Have you discovered LTSpice yet? If you're interested in power supply design, it's worth downloading it and spending about a week of your life learning, experimenting, reading the manual and then experimenting some more.

If you draw up an accurate model of your power supply, the predicted rms ripple current can be computed and shown with a single click.

Offline Bored@Work

  • Super Contributor
  • ***
  • Posts: 3932
  • Country: 00
Re: Ripple current calculation / frequency for psu design
« Reply #2 on: January 15, 2013, 10:41:40 pm »
You can do it old school, the way they taught you in your engineering courses. One classic homework is to proof/derive the transformer dimensioning equations, so you don't forget them.

The ripple voltage for a bridge rectifier circuit is



With Iout the output current through the output load RL, smoothing capacitor C, AC frequency f and internal transformer resistance Ri.

If you then start to simplify by choosing/assuming a transformer where you have



you get



You specify a maximum ripple you can accept, and solve that equation for C.

If you are into simulation you can then use the values as a start for a simulation.

In reality you use a few more equations to calculate required transformer properties (which I am too lazy to look up), which in turn requires to solve a non-linear equation system, which you typically do via a relatively simple iteration in the particular case. And again you can plug the result of the iteration or just an initial ballpark estimate into a simulator if you like.
I delete PMs unread. If you have something to say, say it in public.
For all else: Profile->[Modify Profile]Buddies/Ignore List->Edit Ignore List
 

Offline RossK

  • Contributor
  • Posts: 19
  • Country: us
Re: Ripple current calculation / frequency for psu design
« Reply #3 on: January 16, 2013, 01:57:56 am »
when looking at LT Spice here http://www.linear.com/designtools/software/ i found LTpowerCAD as well which I haven't looked into at all but hey, it reads like a silver bullet

Quote
The LTpowerCAD design tool is a Microsoft Excel based power supply design tool program that can significantly ease the tasks of power supply design with Linear Technology µModule and some monolithic products. Unlike conventional simulation tools, this tool guides users throughout the whole supply design process: it searches suitable LTC parts according to user’s supply specifications; then it guides the user to select and optimize circuit component values with suggestions and warnings. The tool shows real-time results of feedback loop bode plot, as well as the power stage performance. The design can be exported to LTspice; simulation circuit for users to further verify their designs. The tool also provides PCB layout example and a design summary. Currently, this design tool support Microsoft Excel 2000, 2002, 2003, 2007 and 2010 on a Windows based computer. Contact LTpowerCAD@linear.com for questions or support.
 

Offline amateur_25Topic starter

  • Regular Contributor
  • *
  • Posts: 106
Re: Ripple current calculation / frequency for psu design
« Reply #4 on: January 17, 2013, 09:52:54 pm »
For example in the circuit below is a crude dc-dc converter I intend to use to drive the cooling fan @12v.
I should take the peak to peak current indicated on the probe as the ripple. My capacitor should be greater
than the highest value which is about 314mA. yes?



 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf