Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

ADC problems

Status
Not open for further replies.

ashfsk

New Member
Hello to all.

Im using a ADXL203 accelerometer to measure tilt and accleration and a 10-bit ADC to ditize the o/p.
The o/p of sensor varies from 0.05V to 4.8V. I need to measure a change of 0.099mV.
The smallest measurable voltage using ADC comes to be (4.8-0.05)/1024=4.6mV (Plz correct me if im wrong on this).

Can anyone plz suggest on how to proceed??
I dont know if there r sensors with a better sensitivity than this one.
Some sort of diferential amplifier maybe required, but Im not sure on how to exactly go abt this one.

Thnx.
 
ashfsk said:
Hello to all.

Im using a ADXL203 accelerometer to measure tilt and accleration and a 10-bit ADC to ditize the o/p.
The o/p of sensor varies from 0.05V to 4.8V. I need to measure a change of 0.099mV.
The smallest measurable voltage using ADC comes to be (4.8-0.05)/1024=4.6mV (Plz correct me if im wrong on this).

Can anyone plz suggest on how to proceed??
I dont know if there r sensors with a better sensitivity than this one.
Some sort of diferential amplifier maybe required, but Im not sure on how to exactly go abt this one.

Thnx.
If you need to digitize the entire range of the sensor, you'll need a 16 bit A/D. If you only need part of the range, what part do you need?
 
You need a 16 bit A2D to meet your resolution target, do you REALLY need that degree of resolution? - and are the rest of your components within 0.002% tolerance?.

I think you probably already know how noise is going to affect it?.
 
ashfsk said:
Hello to all.

Im using a ADXL203 accelerometer to measure tilt and accleration and a 10-bit ADC to ditize the o/p.
The o/p of sensor varies from 0.05V to 4.8V. I need to measure a change of 0.099mV.
The smallest measurable voltage using ADC comes to be (4.8-0.05)/1024=4.6mV (Plz correct me if im wrong on this).

Can anyone plz suggest on how to proceed??
I dont know if there r sensors with a better sensitivity than this one.
Some sort of diferential amplifier maybe required, but Im not sure on how to exactly go abt this one.

Thnx.

How about starting with good English.

Next, look into a higher resolution a/d.

Finally look at your noise margin to see if this is feasible.
 
ashfsk said:
Hello to all.

Im using a ADXL203 accelerometer to measure tilt and accleration and a 10-bit ADC to ditize the o/p.
The o/p of sensor varies from 0.05V to 4.8V. I need to measure a change of 0.099mV.
The smallest measurable voltage using ADC comes to be (4.8-0.05)/1024=4.6mV (Plz correct me if im wrong on this).

WOW... You must have some really stable components in your circuit to need 0.099mv resolution... Why do you need microvolt resolution? Are you looking to detect the acceleration of tectonic plates?

The ADXL203 has a range of -1.7G to +1.7G.. Now assuming 0G is approx 2.4V and +1.7G is 4.8v this equates to approx 1.412V per G, or 1412mV per G. Why the need to try and resolve 0.099mv?
I built a 3 axis accelerometer, just for fun, using the MMA3202D, and even using decent filtering, 10 bit A/D and lots of averaging math, the displayed acceleration wasn't stable, even when stationary! It varied due to noise, both incircuit (even resistors make noise), and external. Admittedly, I was displaying to 3 decimal places, but it kinda proves a point.....
Microchip have a 22bit A/D chip on their website. They even give freebies away.. Maybe you need to order a few! Link below.....

**broken link removed**
 
datasheet required

hi all

i m doing my final year project.i need the datasheet for 5460 adc.if any one knows plz help me to do my project.
 
I worked for an outfit that did industrial control. The A to D they used was 25 bits. All .01% resistors and hand-selected switching diodes. Plus lots of overdriving the bridge circuit and signal averaging.

Getting accuracy like that is not simple - you'll even need active zeroing on the amplifier feeding the A to D. Talk a bit more with your advisor and see if a decimal place got slipped a bit.
 
I am doing project with ADC 5460 : I want to use for two channel ADC measurement and sending data on USB bus : can any one guide with proper Hardware and software with related links
 
Status
Not open for further replies.

New Articles From Microcontroller Tips

Back
Top