Technique to reduce delay in analog circuit

I have designed current cooperator. In that i want to measure delay. i have two basic doubts

1. what parameter should i use to measure delay as my input is current pulse and output i am measuring voltage. ???? (as delay is varying if i m measuring between V(out) and I(in) or V(out) and V(in) )

2. What are the techniques that i should use to reduce this delay ????

Im not sure what is current cooperator but delay in analogue circuit depends not only on the component like transistor or IC but also on its configuration like closed loop feedback or open loop configuration with what external components u have used.
U can find the delay out by finding out the transfer function of entire circuit and calculating delay. Or u can use an MCU or 555 to connect input and output of analogue circuit for starting and stopping the timer. But MCU may has its own software and timer start/stop delay. U have to find them theoretically and make corrections, same holds for 555 based circuit. While doing things with 555 u can use a separate MCU or a CRO.
U can also directly connect high end mixed signal oscilloscope to input and output and capture the inout waveforms and delay. But they are costly.

The delay can be reduced by simplyfying ur transfer function or state space simulation of ur circuit.

Hope that helps.

oh sorry.... typing mistak. The term is current comparator. I made a schematic of current comparator and trying to get delay by (S-edit/T-Spice)Tanner-tool.

I have never used simulation software but using that u can determine the total capacitance and stray inductance from output to input then use the output of comparator as load resistance. Using these RLC values u can calculate time constant and determine how long it takes for output to appear. As comparator is an open loop circuit, this time constant will be moreover constant across GBWP.
The other way is to determine open loop transfer function and determine the total output response time using rise time and steady state time. Spice or other tools as per I know can find these TF's.

Hope that helps.

For non sine waves "delay" is influenced by the amplitude. If you consider a rising edge as an input and an over driven amplifier. The output of the amplifier might start later but reach its maximum value sooner (and then have a flat top) then the input waveform. So what is the delay?
Delay is caused by shunt capacity and series resistance, so keep the circuit impedances low.
What sort of delays are you thinking of? milliseconds, microseconds or nanoseconds, the techniques for dealing with these will be very different.
Frank

chuckey said:

For non sine waves "delay" is influenced by the amplitude. If you consider a rising edge as an input and an over driven amplifier. The output of the amplifier might start later but reach its maximum value sooner (and then have a flat top) then the input waveform. So what is the delay?
Delay is caused by shunt capacity and series resistance, so keep the circuit impedances low.
What sort of delays are you thinking of? milliseconds, microseconds or nanoseconds, the techniques for dealing with these will be very different.
Frank

Click to expand...

The delay i m talking is of nanosecond.

my input signal to circuit is current pulse and output i m getting is a voltage pulse. (so type of input is clear)

if i measure delay between V(out) and I(in) thn will it be valid ??? or i should always go to measure delay between same excitation and response ?