Hi,
I have 0.005 Ohm current sense resistor for a motor control application (PMSM). What's the best differential amplifier to measure the current (max 10 A) using kelvin connection.
i.e., V=IR
Drop across current resistor = 0.005*1 (@ 1 Amp) = 0.005 V. This is very low voltage drop.
Any referense schematics would be of great help.
thanks and regards
Hello,
Please find a schematic attached for a low-side current sense application using the OPA320. I modified the schematic you submitted to the "SPICE" forum using the TLC2274. I maintained most of your component values and have included the math I used to calculate the component values that were changed.
To minimize errors when performing current shunt measurements during PCB layout, be sure to take connect the inputs to the amplifier directly at the shunt resistor terminals. Connecting the inputs close by on a plane or through a wire will induce errors due to the PCB resistances.
In the case of the schematic below, be sure to connect R83 and R4 directly at the R2 shunt resistor terminals on the PCB. Connect the GND connection of R4 directly at the GND of the sense resistor to minimize any errors due to GND plane resistances (small as they may be).
The circuit provides a linear output from roughly 0V to 3V with a 0 - 10A input current through a 5mOhm sense resistor.
The negative output stage saturates at around 10mV and measurements below 50mA are not possible with this design.
Please let me know if you have any further questions.
Regards,
Collin Wells
Linear Applications
Thanks for the info and the schematics. Really nice of you sending them.
I have a query.
1) Could I use OPA320 amplifier if the motor works as a generator i.e., the direction of current would be reverse. Probably if you look at the TLC2274 schematics it has a mid point setting to take care of this requirement.
Please suggest.
2) Could you suggest how to implement a kelvin connection for motor controller.
The 4-Wire 'Kelvin' method (circuit 4) is used in difficult cases when lead
resistances vary, RX is very small, or when very high accuracy is required. The
method is immune to the influence of lead resistance and is limited by the
quality of the constant current source and voltage measurement. Thermo-electric
voltages can be eliminated by averaging two measurements with the polarity of
the excitation current reversed.
http://www.arab-training.com/vb/t13647.html
please suggest
thx
Hello,
1.) The circuit I previously provided would not work if the current source went negative. Below please find a schematic showing a modified version of the midpoint circuit that was in your original design. I've included the math required to get the circuit to operate. I chose to center the output voltage at 1.65V and designed for a span of 3V for the 20A current range. This prevents the both the positive and negative output stages from saturating during operation by limiting the output swing from 0.15V to 3.15V off a +3.3V supply.
2.) I may be misinterpreting what you're asking here, so please correct me if needed. The "Kelvin" or 4-wire measurement technique is typically referred to when measuring small resistances by forcing a current through the resistor and then measuring the voltage drop across it. Because the lead resistances of the probes have voltage drops of their own when current passes through them, the measurement accuracy is affected. 4-wire measurements solve this issue by using separate force and sense probes. The sense probes are very high-impedance so the measurement is not affected by them.
In this case you are not trying to measure resistance, you're trying to measure the current with a known resistance. The force and sense lines are already separate because the forced current is from the motor load and the sense lines are the high-impedance inputs of the OPA320. Please follow the layout advice in the previous post to minimize any PCB resistance errors.
2046.Low-Side Current Sense - BiDirectional.TSC
Best regards,
Collin Wells
Precision Analog
Linear Applications
