CAPACITOR POLARITY

Thread Starter

VernonLS

Joined Oct 9, 2019
42
If I have a simple RC network with a zero to plus 5 square wave as input, the voltage across the resistor in the network shown will step instantaneously (almost) to +5v and then fall slowly to ground at a rate dependent on the rc time constant of the network. For my network I am estimating that the resistor is about 10K and the capacitor is about 1 microfarad so the time constant is about 10 ms. My question is what happens if I use an electrolytic capacitor. Should the plus side connect to the signal source or the resistor? And, if it is connected in reverse what will happen assuming that the capacitor rating is 50v. I don't think it will go pop, but what happens over time? Does it develop an internal short, does it work but at a lower capacitance than expected? I dunno. Can anyone suggest what happens. RC NETWORK.jpg
 

AlbertHall

Joined Jun 4, 2014
9,022
The positive side should connect to the signal source.
If you connect it backwards there isn't enough current available to make it go bang. It will gradually degrade the insulation on the electrode and become more and more leaky until it is a short circuit.
 

TeeKay6

Joined Apr 20, 2019
560
If I have a simple RC network with a zero to plus 5 square wave as input, the voltage across the resistor in the network shown will step instantaneously (almost) to +5v and then fall slowly to ground at a rate dependent on the rc time constant of the network. For my network I am estimating that the resistor is about 10K and the capacitor is about 1 microfarad so the time constant is about 10 ms. My question is what happens if I use an electrolytic capacitor. Should the plus side connect to the signal source or the resistor? And, if it is connected in reverse what will happen assuming that the capacitor rating is 50v. I don't think it will go pop, but what happens over time? Does it develop an internal short, does it work but at a lower capacitance than expected? I dunno. Can anyone suggest what happens. View attachment 193075
Do not deliberately perform the experiment of connecting electrolytic caps with reversed polarity. Depending on the voltage and current that the attached circuit can apply, the electrolytic cap can indeed explode. You don't want you eyes to be exposed to that.

Many years ago in a lab where I worked, I turned on a newly-built instrument for checkout. As I waited for it to "warm up" I walked to the other end of the lab. I had walked about 30 feet when there was an explosion like a shotgun being fired and a binding post from the front of the closed instrument was propelled violently, landing at my feet. In that case the voltage was about 150VDC and the cap was about 100uF, in a metal can mounted to a chassis...and essentially connected directly to the AC mains via a diode bridge. The inside of the instrument afterward did indeed look as if a shotgun shell had detonated inside.
 

AlbertHall

Joined Jun 4, 2014
9,022
Do not deliberately perform the experiment of connecting electrolytic caps with reversed polarity. Depending on the voltage and current that the attached circuit can apply, the electrolytic cap can indeed explode. You don't want you eyes to be exposed to that.

Many years ago in a lab where I worked, I turned on a newly-built instrument for checkout. As I waited for it to "warm up" I walked to the other end of the lab. I had walked about 30 feet when there was an explosion like a shotgun being fired and a binding post from the front of the closed instrument was propelled violently, landing at my feet. In that case the voltage was about 150VDC and the cap was about 100uF, in a metal can mounted to a chassis...and essentially connected directly to the AC mains via a diode bridge. The inside of the instrument afterward did indeed look as if a shotgun shell had detonated inside.
One place where I worked a guy (aka an idiot) used to connect an electrolytic by long wires to a mains plug. throw the capacitor up in the air, then switch on the power. He did this at Christmas and he said it was his way of decorating the place with some streamers.
 

crutschow

Joined Mar 14, 2008
24,097
Here's the LTspice simulation of your circuit.

As you can see, the voltage across the capacitor (yellow trace) always stays positive between the top of the capacitor (node 1) and the bottom (node 2).

Thus a polarized capacitor should have the (+) connected to node 1.

1574786980061.png
 

Thread Starter

VernonLS

Joined Oct 9, 2019
42
The positive side should connect to the signal source.
If you connect it backwards there isn't enough current available to make it go bang. It will gradually degrade the insulation on the electrode and become more and more leaky until it is a short circuit.
Thanks, of all the replies, yours was the most helpful in understanding how a working circuit might degrade. Let me pose another question. What if the input was a+5V to -5V square wave? Would an electrolytic capacitor simply be a bad choice?
 

AlbertHall

Joined Jun 4, 2014
9,022
What if the input was a+5V to -5V square wave? Would an electrolytic capacitor simply be a bad choice?
You could get away with it if the square wave is not 50:50 mark space ratio and the time constant does not allow a reverse polarity.
Another way is to connect two electrolytics in series back to back. This effectively gives you a capacitor of half the value but which can be used in AC circuits.
 

Thread Starter

VernonLS

Joined Oct 9, 2019
42
You could get away with it if the square wave is not 50:50 mark space ratio and the time constant does not allow a reverse polarity.
Another way is to connect two electrolytics in series back to back. This effectively gives you a capacitor of half the value but which can be used in AC circuits.
Wow, thanks, I learned a lot from our posts.
 

MisterBill2

Joined Jan 23, 2018
4,859
You could get away with it if the square wave is not 50:50 mark space ratio and the time constant does not allow a reverse polarity.
Another way is to connect two electrolytics in series back to back. This effectively gives you a capacitor of half the value but which can be used in AC circuits.
One microfarad non-polarized capacitors are available and fairly common and so they would be the best choice
 
Top