Maker Pro
Maker Pro

How to measure millivolt drop?

Hello,

I am doing a project and one of the requirement is to measure
millivolt voltage signal, need 16-bit ADC. The actual input range is
below 100milivolt. If I choose a ADC with input range to 4V,
1LSB=0.06mV. Shall I just measure the input without any voltage
amplifier added in front of the ADC input pin?

Another requirement is another analog measurement, whose input
impedance is over 20MOhm. Can I just use a normal ADC chip connect to
that input without any other circuit? Or I need a very low bias
current OPamp to amplify the input signal, and have the voltage
amplifier's output connected to the ADC input.

Many thanks,

Regards
Jerry
 
J

Jim Thompson

Jan 1, 1970
0
Hello,

I am doing a project and one of the requirement is to measure
millivolt voltage signal, need 16-bit ADC. The actual input range is
below 100milivolt. If I choose a ADC with input range to 4V,
1LSB=0.06mV. Shall I just measure the input without any voltage
amplifier added in front of the ADC input pin?

Another requirement is another analog measurement, whose input
impedance is over 20MOhm. Can I just use a normal ADC chip connect to
that input without any other circuit? Or I need a very low bias
current OPamp to amplify the input signal, and have the voltage
amplifier's output connected to the ADC input.

Many thanks,

Regards
Jerry

Them thar homework problems are getting difficult ;-)

...Jim Thompson
 
C

ChairmanOfTheBored

Jan 1, 1970
0
Them thar homework problems are getting difficult ;-)


Controlling 15kV to within 20 Volts meant the capacity to read microvolt
changes in the feedback loop. We had no problem achieving this, despite
the huge fields present in HV supply proximities.

This should be a cakewalk.
 
P

Phil Allison

Jan 1, 1970
0
I am doing a project and one of the requirement is to measure
millivolt voltage signal, need 16-bit ADC. The actual input range is
below 100milivolt. If I choose a ADC with input range to 4V,
1LSB=0.06mV. Shall I just measure the input without any voltage
amplifier added in front of the ADC input pin?


** Wasting the available resolution of the ADC if you don't.

Another requirement is another analog measurement, whose input
impedance is over 20MOhm. Can I just use a normal ADC chip connect to
that input without any other circuit? Or I need a very low bias
current OPamp to amplify the input signal, and have the voltage
amplifier's output connected to the ADC input.


** Never heard of FET or BiFet op-amps ?

Input resistance in the Gohms.

Look up the TL071.


......... Phil
 
R

Rene Tschaggelar

Jan 1, 1970
0
Hello,

I am doing a project and one of the requirement is to measure
millivolt voltage signal, need 16-bit ADC. The actual input range is
below 100milivolt. If I choose a ADC with input range to 4V,
1LSB=0.06mV. Shall I just measure the input without any voltage
amplifier added in front of the ADC input pin?

What do you gain by not having an amplifier
doing a gain of 40 ?
Another requirement is another analog measurement, whose input
impedance is over 20MOhm. Can I just use a normal ADC chip connect to
that input without any other circuit? Or I need a very low bias
current OPamp to amplify the input signal, and have the voltage
amplifier's output connected to the ADC input.

What happens if the ADC is having a lower
impedance than expected ?

Rene
 
W

Winfield

Jan 1, 1970
0
ChairmanOfTheBored said:
Controlling 15kV to within 20 Volts meant the capacity to read
microvolt changes in the feedback loop. We had no problem
achieving this, despite the huge fields present in HV supply
proximities.

Right, that's only 0.1% and electric fields are easy to tame.
 
C

ChairmanOfTheBored

Jan 1, 1970
0
Right, that's only 0.1% and electric fields are easy to tame.


Except that our success with it was typically to within 5 volts, and the
HV feedback resistor is in the same area as the HV multiplier stage, and
you don't know what the **** you are talking about in this HV realm.
 
W

Winfield

Jan 1, 1970
0
Except that our success with it was typically to within 5 volts, and the
HV feedback resistor is in the same area as the HV multiplier stage, and
you don't know what the **** you are talking about in this HV realm.

I regularly make unusual push-the-envelope designs to
20kV and beyond, what HV realm are you talking about?
It's generally the magnetic fields I struggle with.
 
W

Winfield

Jan 1, 1970
0
Except that our success with it was typically to within 5 volts, and the
HV feedback resistor is in the same area as the HV multiplier stage, and
you don't know what the **** you are talking about in this HV realm.

Well, I regularly make unusual push-the-envelope designs
to 20kV and beyond; what HV realm are you talking about?

For example, a recently-finished instrument, a Rydberg-
atom expeller, creates two 10 to 15kV voltages, which
are applied to two electrodes in a vacuum chamber.

The voltages are independently controllable, but in a
precision-dump mode one electrode must be remotely set
to a small voltage away from the other, ranging from 0
to 200 volts, within 0.2 volts (the experiment requires
a sweep where the effect is proportional to V^2, so the
values below 5 volts are especially significant). This
is all measured and controlled from ground potential, so
that's 0.2/10kV = 0.002% or 20ppm precision. I was able
to easily create the appropriate electric-field shields.

It's generally the magnetic fields I struggle with. This
case was no exception, given my (perhaps poor) choice to
design a square-wave PWM with custom HV transformer to
create the precision 0 to 200 volts (in retrospect, a
sine wave transformer drive would have been better).

The rejected, but perhaps better choice, would have been
to precisely control the 0 to 200V right at the HV end,
using an optical digital link. I balked at that choice
when I worried that reporting on the resulting voltage
from the HV side with another optical link didn't seem
fully honest, so at least one analog link should be used.
 

neon

Oct 21, 2006
1,325
Joined
Oct 21, 2006
Messages
1,325
let mw put it to you this way. if you got 10v input one bit is is what? 3.276-4
if you expect to get this kind of resolution from a 100mv input from a ADC you are dreaming. And finaly 4v 16 bits adc 1 LSB is not .06 mv. you are realy not playing witha full deck are you?
 
Top