Why is a polarized capacitor needed for this circuit?

Thread Starter


Joined Dec 7, 2012
I built a circuit like this to make a blinking LED (actually my circuit is from page 59 of Make Electronics, but very similar)

I wanted to see what would happen if I used a larger capacitor but I only have a 3000uf non-polarized one. I thought it would make the LED blink at a slower rate but instead the circuit acted as if there was no capacitor at all. Why is that?


Joined Mar 31, 2012
Are you sure it's 3000 microfarads? What kind of capacitor is it? How do you know it's non-polarized? Have you tried putting it in both ways?

Can you upload a picture of it?

Thread Starter


Joined Dec 7, 2012
Here's a photo of the 2 capacitors I've tried. The black one works as expected (slows the rate of flashing). The green one is what I thought was a 3000uf unpolarized cap. I've tried it in both directions and there's no change (although it doesn't seem to matter which way I put the black one either).



Joined Mar 14, 2008
The black one is polarized and should be connected as the polarity is marked.
It may work in the short term connected opposite, but likely will fail with time.

The green cap may be 39k pF or 39nF, which is why it didn't appear to work.


Joined Apr 11, 2010
That’s what I came up with, too. I found another coding scheme that suggests that it could be 3*10^9 pF. It’s late and I couldn’t get my head around the powers of 10.

Thread Starter


Joined Dec 7, 2012
Yes, 0.039 uF, the 'K' is the tolerance (10%) rated 100 volts.
Thanks. I clearly read it very wrong and that explains the result. But could someone explain why it's read as .039uF and not 3,000,000,000pF? I realize this is a pretty basic question, but an important one too.


Joined Dec 31, 2017
But could someone explain why it's read as .039uF and not 3,000,000,000pF?
That calculator you were using doesn't recognize a leading zero, so the output is bogus. That cap must be pretty old, a newer cap of that value would be marked 2A393K. The 2A stands for 100 volt DC rating.


Joined Nov 13, 2015
Beyond the resistor color code, there is only loose adherence to marking 'standards'. Experience tell me that this is a 0.039 uF (39 nF) cap and not a 3,000,000,000 or a 39,000 uf cap. The later large values would be physically huge.

Sometimes, even us 'old timers' need to bring out the meter to confirm the value of a part.