Hi all
I have a question
I use atmega 8 for making pulses on OC1A pin
I used oscope to zoom in one pulse details
The problem is that when an overshoot is created on pulse level change, the overshoot maximum is higer than the input ic voltage ?
How does it can occur ?
I attached the oscope result

the blue is input voltage and the yellow is overshoot

 

How long is your ground lead on the oscilloscope probe? Try attaching the spring ground contact and try again.

 

Try attaching the spring ground contact and try again

What do you mean about spring ground ?

 

http://i.stack.imgur.com/PSo3N.jpg

 

Why this occur ?

 

Because the 10cm of the ground lead has lots of inductance, which will change the shape of the edges of the signal. So it is ok to use the long lead for slow signal, but if you want accurate measurement of fast edges you need to use the shortest ground possible.
Also the long lead will often pickup stray magnetic fields from other parts of the measured circuit which again might fool you into thinking that something is wrong when it really isnt.

 

It occurs because the resistance, inductance, and capacitance of the circuit force it to.

 

You mean the inductance of ground wire produces voltage because of frequency and the voltage shows higher amount right ?

 

The inductance of the ground lead of your scope probe and the capacitance of the probe form a series resonant circuit, and it is the "ringing" of this resonant circuit you are seeing on the scope; there's nothing wrong with your circuit. To reduce this overshoot, use the shortest ground connection possible.

 

Yes. The probe has some capacitance between the tip and ground, and the ground lead has some inductance. With high enough dV/dT (rise time) the resonance of the LC becomes significant and creates that ringing.

 

OBW0549
You mean there must be no overshoot at all ?

 

Yes. The probe has some capacitance between the tip and ground, and the ground lead has some inductance. With high enough dV/dT (rise time) the resonance of the LC becomes significant and creates that ringing.

But why it is higher than IC voltage ?

 

Look at this, red is the actual voltage at V1, blue is what your scope probe would measure.

 

Ok I tested on scope I get the result

 

OBW0549
You mean there must be no overshoot at all ?

Ummm... no. What I'm saying is that you should stop worrying about this "overshoot" because it is occurring in your scope probe, NOT in the circuit you are testing. If this apparent overshoot still bothers you, you can minimize it by using a shorter probe ground lead, which will have less inductance.

(If you want to demonstrate the effect of excessive ground lead inductance, try extending the scope probe ground lead with a long clip lead, like several feet long; then observe on the scope, you will see a LOT of ringing!)