Detaining max load capacitance before pulse transitions are distorted

[Macboost]

Hi!

Just starting out in a pulse and digital class and I'm already completely lost. I guess this should be pretty entry level stuff, but is just don't understand what formula to use.

Question:

A signal source with Rs=500ohms produces pulses with rise and fall times of 100ns. Determine the maximum load capacitance the signal source can drive before the pulse transitions are distorted.

 

[Gryd3]

Have you taken any other electronics courses?

 

[KrisBlueNZ]

A signal source with Rs=500ohms produces pulses with rise and fall times of 100ns. Determine the maximum load capacitance the signal source can drive before the pulse transitions are distorted.

I must be missing something. ANY amount of load capacitance will distort the signal somewhat. Also they probably need to specify the shape of the edges. Or do you assume they're straight? How do they define "distorted"?

 

[Macboost]

Have you taken any other electronics courses?

Yes, but it's been over 5 years since I've seen the inside of a classroom - don't remember much.

 

[Macboost]

I must be missing something. ANY amount of load capacitance will distort the signal somewhat. Also they probably need to specify the shape of the edges. Or do you assume they're straight? How do they define "distorted"?

This is literally the exact question with no other info given:

 

[KrisBlueNZ]

"significantly" distorted... Again this needs to be defined. Have a look through your notes and see how they define it.

 

[Macboost]

Looks like f(high cutoff) needs to be greater than or equal to 2 times the highest significant harmonic.

 

[KrisBlueNZ]

OK. Can you determine the highest "significant" frequency component of a signal that rises from 10% to 90% of its total swing in 100 ns and also falls from 90% to 10% of its total swing in 100 ns?

 

[Macboost]

I would say 5Mhz;

FH=1/(2 x rise time)

 

[Macboost]

Thanks for pointing me in the right direction. Now, correct me if I'm wrong, but Tau would have to be less than or equal to rise time divided by 2pie?
And since I know Rs is 500ohms...

 

[KrisBlueNZ]

According to http://www.edn.com/electronics-blog...-The-bandwidth-of-a-signal-from-its-rise-time the highest "significant" frequency component in a signal of rise time t is 0.35 / t. So in your case that would be 3.5 MHz.

So that means the cutoff (-3 dB) frequency of the RC filter (pole) formed by the output resistance and the load capacitance would have to be at least ...?

So as you said, you can work back to calculate C from f_-3dB and R

 

[Macboost]

Looks like I understand.

Thank you!

 

[Arouse1973]

I must be missing something. ANY amount of load capacitance will distort the signal somewhat. Also they probably need to specify the shape of the edges. Or do you assume they're straight? How do they define "distorted"?

Absolutely Kris. There needs to be a level of acceptance for the circuit with regards the rolling off of the signal.
Adam