Author Options:

Will this circuit work? Answered

Ok, so I'm building a giant led array and I'm running it off of a PWM chip and a few shift registers. Given the number of addresses and variables I need for each one, I'm guessing that the micro running the show won't like that too much and will be a bit slower than I'd like, and i'm not sure if it will be slow enough to cause irritating visual inconsistencies. This got me thinking, how did tube TVs do it? They had phosphors that slowly released their charge and therefor looked alright. Thats how I arrived at this circuit.

Will it pretend to be a phosphor well enough or will it need something different?



5 years ago

Good logical thinking.

Impossible circuitry, mixing analog to digital does not do it at all :-p


So that wouldn't light th led up and simultaneously charge the cap when theres power and then discharge through the led when there isn't?

What might do that? first thing i thought of was hooking an inductor up in parallel, but since I'm not terribly familiar with them I dismissed the idea in favor of the one in the pic

If you have a true Totem pole PWM output,
The high will charge the cap as you said.
But the low will discharge your capacitor and
the LED will not light.during that time

You know about POV persistence of vision.
All you need to do is make sure the PWM frequency is above that.
and it will appear to your eyes as though the LED is steady ON.

Why would the cap not light the led? Not enough current? What If I used something like a polarized cap?

When the PWM line is high the capacitor is being charged.

When the PWM line is low ( essentially ground )
which holds the LED at low ground essentially zero voltage
preventing the LED from getting any current flow and
discharging the any capacitor to ground..

Go ahead and build it to see,
 don't stop on my account....

Would any combination of those 3 parts work?

Now I'm curious about just what could give a typical LED some sort of "residual" glow with the fewest and smallest parts possible.
Any ideas?

Ok, so I googled the original inductor idea and found something that might work kind of like I want
If I put the LED in reverse biased in parallel with an inductor(no resistor), and invert the PWM signal, I could get something of an analog pulse from the LED. Would the discharge be "instantaneous" like with a cap, or would it decay slower? Or am I on the wrong track an it would blow it up?
(Pic relevant, I can't recall which way is which)


Try this variation.

The high PWM causes current to flow through the inductor and LED.

The low PWM has no current flow But the inductor wants to keep the
current of its collapsing magnetic energy going and
it does so through the left pointing Sckottky diode which incidentally
includes the LED in it's path.


Does the left pointing one need to be a schotttky, or can it be of the more typica/basic variety? What about the other one?

YES....  "left pointing Sckottky diode"   you can use
the 1N5817 IS A 20PIV 1AMP   14¢ diode at Jameco.

The low forward voltage helps prolong your LED post light.

This pic works too but triples your current ( don't need Shocttky ).