A sample text widget

Etiam pulvinar consectetur dolor sed malesuada. Ut convallis euismod dolor nec pretium. Nunc ut tristique massa.

Nam sodales mi vitae dolor ullamcorper et vulputate enim accumsan. Morbi orci magna, tincidunt vitae molestie nec, molestie at mi. Nulla nulla lorem, suscipit in posuere in, interdum non magna.

Bigger is not always Better – Filter Capacitors for Power Supplies

A filter capacitor to smooth a half-wave rectifier.

A filter capacitor to smooth a half-wave rectifier.

This is a rewrite of the previous post. I had two problems with the previous post. First, I miscalculated the Diode resistance by a factor of 10 and second, I did not put the full load on the power-supply. Ripple is going to look a lot worse this time around because of the increased load. The actual power supply is a kit I bought over the holidays and this is an exercise in understanding why it has the component sizes it does.

The power supply consists of a transformer with a 12 Vrms secondary with both a positive and negative half-wave rectifier.  The filter capacitor on both sides of the power supply is 2200μF.  The filter is followed by an adjustable regulator.  We are only going to talk about the positive side and through the filter circuit at this time.  The diode is a 1N4001 with a peak reverse rating, PRV, of 50 V.  The maximum value it will have to block is 12.6  x 2 x √2 -or  35.6 V so the 50 V PRV is fine.

We will be asking the question, what would happen if a 500μF, a 2200μF, and a 22000μF cap was used.  This brackets the value by 1/4 smaller and 10 X larger.   The first question is how much ripple will see at the rated current of the power supply. All of this was simulated using the standard 60Hz power in the US.

Ripple voltage with 750 ma load for the 500 uF Cap.

Ripple voltage with 750 ma load for the 500 uF Cap.

Ripple with a 750 ma load and a 2200 uF cap.

Ripple with a 750 ma load and a 2200 uF cap.

Ripple with 750 ma load and a 22000 uF Cap

Ripple with 750 ma load and a 22000 uF Cap

Bigger is better in this case. Now will will look at the current though the diode in each of these cases.

The current pulses through the diode with a 750 ma load and  500 uF cap.

The current pulses through the diode with a 750 ma load and 500 uF cap.

Current pulses through the diode with a 750 ma load and a 2200 uF Cap

Current pulses through the diode with a 750 ma load and a 2200 uF Cap

Current pulses through the diode with 750 mA load and 22000 uF Cap

Current pulses through the diode with 750 mA load and 22000 uF Cap

The most notable thing of these three graphs is the 500 uF cap draws less current and the pulses are wider. Because the cap has discharged more during the “off” half cycle the diode starts conducting earlier in the forward part of the cycle and charges the capacitor while the voltage is still increasing, so less current is drawn. Because the conduction time is less, the large capacitor does not have time to charge to as high of a voltage as the medium sized capacitor. If you look at the voltage graphs above, the large capacitor does not reach the voltage of the others.

The next three plots show how fast the capacitors charge when power is first turned on.  These were created with a very small load on the power supply.

Initial Charging of a 500 uF Filter Cap.

Initial Charging of a 500 uF Filter Cap.

Initial charging of the 2200uF cap.

Initial charging of the 2200uF cap.

Initial charging of the 22000 uF cap.

Initial charging of the 22000 uF cap.

Notice the very large cap did not reach a very high voltage during these first 10 cycles. The next plots will show the real problem with using very large capacitors.

Diode current for the first 10 cycles to a 500 uF cap.

Diode current for the first 10 cycles to a 500 uF cap.

Diode current for the first 10 cycles with a 2200 uF cap.

Diode current for the first 10 cycles with a 2200 uF cap.

Diode current for the first 10 cycles with a 22000uF cap.

Diode current for the first 10 cycles with a 22000uF cap.

This shows the main problem with using a very large capacitor. We are drawing too much current too many cycles and we will exceed the repetitive current surge rating for the diode. Remember also in this simulation a very small load was connected to the power supply during this charging simulation.

This simulation did not consider any impedance effects of the transformer. To do so would require some equipment I currently do not possess. However, I hope the graphs dramatically show the effects of various sizes of capacitors.

Gary


 

Print Friendly

Leave a Reply

You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>