How to design electronic circuits

WBahn

Joined Mar 31, 2012
29,979
I noticed there's a huge difference between electrolytic and non-electrolytic capacitors; I used a 47μF 100V non-electrolytic capacitor instead of the 10μF 63V electrolytic and the frequency went up instead of going down. How can that be? :confused:
Leakage current. Using an electrolytic as a timing capacitor is seldom a wise thing to do. You can end up with such severe leakage (and it doesn't take much) that the circuit actually stalls and sits there looking stupid (sorta like my dog).
 

Thread Starter

adam555

Joined Aug 17, 2013
858
But there are no electrolytic capacitors for the nF range. How do you do in those cases?

I always wondered: is a non-electrolytic capacitor in series with a diode the same as an electrolytic capacitor (disregarding the diode's voltage drop)?

Another problem I found: I tried to bring the square wave down -so that it's mid position would be at 0v and the peaks positive and negative- and I imagined that it would be similar to, well, I think it's called decoupling (getting rid of the DC current in the wave). But if I add the capacitor, then it's not a square wave anymore, it becomes a mixture between a square wave with a triangular tip. How do I achieve this without messing with the form of the wave?
 

WBahn

Joined Mar 31, 2012
29,979
The reason that electrolytic capacitors are used is because it is often the only practical and economical way to get large capacitances in a reasonably small package. I'm not aware of any (legitimate) applicaiton in which you would WANT to use an electrolytic if a viable non-electrolytic option were available. Now, if you are WANTING to use the capacitor as a detonator or incendiary, then perhaps....

No, putting a diode in series with a non-electrolytic cap does not make the end result an electrolytic cap. The reason that electrolytic caps shouldn't be reverse biased is that doing so drives electrochemistry that breaks down the dielectric barrier and shorts out the capacitor (to keep things simple -- the actual mechanisms and effects are more complicated, of course).

Are you talking about removing the DC component from the square wave at the output? You need to add a decoupling capacitor that is properly sized so that it removed only the lowest frequency components. Provide a screen shot of your output and the circuit as it looks with your capacitor in it and perhaps we can tell you more.
 

Thread Starter

adam555

Joined Aug 17, 2013
858
Thanks WBahn,

I misunderstood your previous answer; I thought you were saying the electrolytic were preferable.

Also got the decoupler working by changing the capacitor. It was a bit tricky because it had to decouple a wide range of frequencies.
 

Thread Starter

adam555

Joined Aug 17, 2013
858
Now I'm having trouble converting a square wave to a triangular wave. I can do it by using a simple capacitor, but it only works for a fixed frequency.

I've been checking out other circuits, and found one with an OpAmp. I got it to work, but it also does the same as the capacitor: the higher the frequency, the smaller the peak to peak voltage.

How can I do it without squeezing the amplitude?

This is the basic design...



The resistors and the capacitor on the top are adjusted for a certain frequency, and I need to change these if I increase the frequency; otherwise the Vpp drops.
 

Attachments

Top