adc quantization levels

[mubi17]

adc input frequency quantization levels

i m a beginner in analogue design and trying a 0.18u adc. i have following problems.

i am not getting 63 themometer codes for 6 bit adc after quantization while using 1 Vp-p input signal. i m trying an aleady developed circuit with track and hold

 

[A.Anand Srinivasan]

adc quantization

check the input range of the ADC it might be more than 1Vpp...

 

[mubi17]

quantization level adc

the defined range for adc is 1 Vp-p. is there any problem which i could correct

 

[A.Anand Srinivasan]

capacitor in 0.18u site:edaboard.com

then check your quantisation level... i think it is more than what it should be... what value r you getting when input is maximum...

 

[mubi17]

adc+quantization circuit

i m trying an already developed system to generate thermometer codes the inputs used by me are

clock 1ns with 100psec width 250p rise and fall time for sample and hold

ac input is 8.172 Mhz with0.5 Vp-p. the out put of s/h is from 0.4 to 1.35v with 70p capacitor. using 0.18u technology.

the thermometer code at 63rd level is around 75ns width pulse and 1st level with is around 7ns

please help me out

thanks

 

[A.Anand Srinivasan]

i didnt get a clear idea of what you've posted.... can you please post your circuit and the waveforms that you have obtained....

 

[mubi17]

quantization adc

anand thanks for taking interest, was too busy to check it, will definitly post the circuit. can u please tell me while using a 1 nsec clock with 100psec high time for 6 bit adc design should i get the thermometer codes with a difference of 1 nsec(time scale on x axis, voltage on y axis, pulse) every time or there could be no variation in adjecent codes. i m getting some thermometer codes right perhapes then in few adjecent codes there is no difference and 2 nsec width (x axis, time scale) difference in very next code. in short missing a code and then next code with twice the difference at x-axis

 

[A.Anand Srinivasan]

quantization levels

i think checking your input signal is the best idea.... i think it is rising too fast than it is supposed to....

 

[mubi17]

should i reduce the input frequency i.e i m using 9 M Hz or change the value of sample and hold capacitor

 

[A.Anand Srinivasan]

now i got some idea on what could be wrong.... i think your capacitor is non-linear as the voltage across it increases and so you better try linearising it.... because something which works for the lower values must work for higher values too....