Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronic Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

What's the most accurate method for measuring current?

Status
Not open for further replies.

Stewie911

Newbie level 6
Joined
Aug 12, 2008
Messages
12
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,373
What is the best way to measure current. It must be very accurate. Current can go as high as 1A.
It is basically current going into ground, so at the moment I am thinking of a precision 50 milli-ohm resistor just before ground with a 16bit ADC just before the resistor. The ADC has programmable gain of about 8 and I will use an OPAMP before the ADC that will also give gain.

As the current goes higher, the resolution doesn't have to be that good anymore.
For example:
0 - 10mA : possibly 50uA, but at least 100uA accuracy.
10-100mA : maybe 500uA accuracy
100mA - 1A : 1mA - 5mA accuracy.

Is there maybe a better way, like hall effect current sensors?
 

FvM

Super Moderator
Staff member
Joined
Jan 22, 2008
Messages
47,928
Helped
14,143
Reputation
28,545
Reaction score
12,827
Trophy points
1,393
Location
Bochum, Germany
Activity points
278,099
Re: current measurement

Direct (shunt) measurement is simple and accurate, although a chopper stabilized amplifier would be needed for the 2.5 uV voltage accuray required by your specification with a 50 mΩ shunt. A higher shunt resistance would be usual, I think.
A hall sensor isn't accurate enough without a true current compensating magnetical circuit.
 

rikie_rizza

Full Member level 6
Joined
Aug 5, 2006
Messages
339
Helped
29
Reputation
58
Reaction score
9
Trophy points
1,298
Location
Bikini Bottom, between a rock and a pineapple
Activity points
3,173
current measurement

I agree with higher resistance because the resistance value is more stable in case of temperature fluctuation. Don't forget to calculate the trace length, total resistance is included with trace resistance.
 

FvM

Super Moderator
Staff member
Joined
Jan 22, 2008
Messages
47,928
Helped
14,143
Reputation
28,545
Reaction score
12,827
Trophy points
1,393
Location
Bochum, Germany
Activity points
278,099
current measurement

I didn't mention explicitely, but a low-ohmic shunt measurement has to use a 4-wire connection of the shunt in any case, otherwise the said accuracy specifications would be just an illusion.
 

kender

Advanced Member level 4
Joined
Jun 19, 2005
Messages
1,422
Helped
138
Reputation
276
Reaction score
38
Trophy points
1,328
Location
Stanford, SF Bay Peninsula, California, Earth, Sol
Activity points
9,970
Re: current measurement

Stewie911 said:
Is there maybe a better way, like hall effect current sensors?
Resistive current sensor is a good approach for 1A. Here's an app note on current sensing, which compares resistive and hall-effect sensors: http://www.zetex.com/3.0/appnotes/apps/an39.pdf

FvM said:
I didn't mention explicitely, but a low-ohmic shunt measurement has to use a 4-wire connection of the shunt in any case, otherwise the said accuracy specifications would be just an illusion.
Yes - 4-terminal resistor is ideal for this method. In the US they are relatively easy to obtain. Caddock makes them and sells them through Mouser and DigiKey. An alternative would be to use a simpler resistor and make a calibration curve. Also, SMT chip resistors have lower parasitic lead resistance.
 

Stewie911

Newbie level 6
Joined
Aug 12, 2008
Messages
12
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,373
current measurement

Ok, I've got a 0.1ohm 4-terminal sense resistor.
The minimum current I want to measure is 100uA.
I now want to amplify the voltage with a non-inverting amplifier with gain of about 20.
That gives me minimum 200uA, and with the 16bit ADC and 2.5V reference, I can measure voltages as low as 38uV with maximum 3LSB error( which is kinda ok for me).

But when I run the circuit through LTSpice, then the circuit only 'works' for currents of about 20mA and higher. Is this because of the spice OPAMP module(generic) that I use?
If I use lower currents in spice, then the output is to high.

Any suggestions?
What about instrumentation amplifier in place of non-inverting amp?
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top