Ok so I just observed something very interesting
The original scope I used where I got the images from was a LeCroy Wave Ace 213 100MHz, 1GS/S 4 channel oscilloscope. Then when I used a different oscilloscope, the ISO-TECH IDS 8062 60MHz 1G Sa/a 2 Channel scope, I got no spikes??

Finally I used a Tektronix 3012 2 Channel oscilloscope (100MHz, 1.25 GS/s) and the spikes returned...

To summarize:

LeCroy Wave Ace 213 100MHz, 1GS/S 4 channel oscilloscope: Spikes
ISO-TECH IDS 8062 60MHz 1G Sa/a 2 Channel scope: No Spikes
Tektronix 3012 2 Channel oscilloscope (100MHz, 1.25 GS/s): Spikes

 

I knew I saw your problem before. When I use cheap probes and ground right next to a power supply filter cap, I get the overshoot. Moving to a remote ground point reduces (eliminates) the overshoot. Disconnecting the scope probe altogether greatly increases the overshoot (as mentioned earlier).

 

I knew I saw your problem before. When I use cheap probes and ground right next to a power supply filter cap, I get the overshoot. Moving to a remote ground point reduces (eliminates) the overshoot. Disconnecting the scope probe altogether greatly increases the overshoot (as mentioned earlier).

DO you mean "Disconnecting the scope probe altogether greatly DECREASES the overshoot"???

 

No, sorry, disconnecting the GROUND lead greatly increases the overshoot.

 

No, sorry, disconnecting the GROUND lead greatly increases the overshoot.

AH ok! Well then with that statement I can confirm that, the spike I am seeing is not in my circuit, it is due to the inductance inside the scope probe. So I can move forward? Is that a fair statement.

 

Increase the resistance from TLP250 pin 2 to ground from 100 ohms to 470 ohms.

Connect a 0.1uF ceramic capacitor in parallel with a 100uF electrolytic capacitor across the power supply lines of the Arduino.

Connect a 0.1uF ceramic capacitor across pins 8 and 5 of TLP250. Place a 100uF electrolytic capacitor in parallel with a 0.1uF ceramic capacitor from +12V to ground.

Observe the signal at the input of the IR2117.

 

Increase the resistance from TLP250 pin 2 to ground from 100 ohms to 470 ohms.

Connect a 0.1uF ceramic capacitor in parallel with a 100uF electrolytic capacitor across the power supply lines of the Arduino.

Connect a 0.1uF ceramic capacitor across pins 8 and 5 of TLP250. Place a 100uF electrolytic capacitor in parallel with a 0.1uF ceramic capacitor from +12V to ground.

Observe the signal at the input of the IR2117.

Hi

I placed all the caps as you suggested, and I am very pleased to say it HELPED! The signal going from my IR2117 to the gate of my mosfet is clean. The load resistor on the mosfet source to ground however, the spikes are still there. I placed the filter caps at the drain of the mosfet aswell.

Attached is the image that is the measurement taken at the source of the MOSFET.

 

AH ok! Well then with that statement I can confirm that, the spike I am seeing is not in my circuit, it is due to the inductance inside the scope probe. So I can move forward? Is that a fair statement.

I think you can. I do not see this on my Arduino tests.

 

I think you can. I do not see this on my Arduino tests.

what scope did you use?

 

A Tektronics 465 analog and a Uni-T UDT2102CEL DSO.

 

Ah ok, I am just afraid now it is actually in my circuit, Because after adding all those filter caps (Tahmid reccomendations, post 26) the spikes went away, just not on my load resistor at the source of my mosfet.... so I am stuck...

 

If it's coming back from your circuit, you can put a 5.3V zener on the signal from the Uno to the circuit, reversed biased (of course) between the signal and ground. That will cut any spikes that might harm the ATmega328.

 

If it's coming back from your circuit, you can put a 5.3V zener on the signal from the Uno to the circuit, reversed biased (of course) between the signal and ground. That will cut any spikes that might harm the ATmega328.

Ah thanks yes that is a good idea, but with the luck i currently have is that after all the nice filter caps,the PWM signal has absolutely no spikes (i have measured) from the arduino PWM output through the opto coupler, into the IR2117 and at the gate of the MOSFET. The PWM is clean. BUT now the issue is the spikes is at the load..

I was thiking to maybe connect a op-amp, then compare a fixed voltage from a DC power supply source that is slight lower than the amplitude of the spike but still higher than the actual pwm signal.Then maybe place a LED or some kind of indicator (still need to look) at the output. SO if the spike is in my cct the indicator will show that to me... just an idea

 

Might be interesting, but the spike is probably harmless. Maybe you can post your circuit so we can have a look. Show us where the spike is seen and where it is not.

 

Yes thats my plan, i will be at the lab tomorrow morning, in 12 hours time basically.

 

Increase horizontal so half the screen is the spike. See if it is ringing, or a single spike.

 

Hi

Here is my circuit: Please note: Only the 0.1uF caps are ceramic, the rest are electrolytic.

 

Increase horizontal so half the screen is the spike. See if it is ringing, or a single spike.

seems like ringing. Measurement is taken across the load resistor. Vr

 

Is the power part of your circuit on a breadboard? What kind of resistor is rload?

 

Is the power part of your circuit on a breadboard? What kind of resistor is rload?

Its a 50W, 330 ohm resistor. The whole circuit is on a bread board