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.

Current sense with a .01 ohm

Status
Not open for further replies.

Mosaic

Well-Known Member
I'm trying to upgrade an existing 0.1 ohm current sense as at 9A the 0.1ohm dissipates 8 watts. An LM358 at 10x handles the signal.

Current range is about 0.5 to 9A. Thus with the 0.01 ohm, 5mV to 90mV difference. Desired output either 10x (to feed into existing LM358) or 100x.

I was considering a current shunt monitor:
**broken link removed**

But I notice it has been spec'd from 20mV to 100mV and has a 2mV input offset. 2mV isn't much better than the LM358.

So I'm asking for suggestions on the approach.
I have access to the following opamps.

LM358, TL072/82, MCP6L72T, MCP6001,TLE2022, TLE2082,TLC2272, MC33072, OPA2134, OP07, NE5532, LT1013
 
Without knowing any more, I'd use a rail-to-rail opamp, LM358 is pretty shoddy in that regard.

What is your circuit like though, high-side current measurement? Are you using a difference amplifier or something similar with one side at the rail and the other after the resistor? Because that can be tricky even with rail-to-rail op amps. You may still have to drop the voltage, perhaps buffer as well, then feed into a difference amplifier. But I'm just speculating now, without knowing more.
 
Given the additional costs of a decent diff/Inst amp and the fact that the LM358 that's in use now will still have to remain (second opamp used elsewhere) I am leaning toward using a linear hall sensor , ACS713 - 20A.
**broken link removed**

That way i can reconfigure the LM358 current sense amp to operate as a difference amp and cancel the 0.5V , 0A min. signal from the ACS713. The new overall amplification looks to be around only 5.4X to achieve the original LM358 0.1 ohm sense output.

Benefits:
1) Very little common ground offset with the AC713's 1.2mohms lifting the ground by 10.8mV max.
2) No Instrument amp required or other opamp.
3) ACS 713 linear sensor is cheaper and smaller than the 0.1 ohm MP930 30W, 1% sense resistor in use.
4) Less circuit modifications.
5) Virtually no heating (0.1W peak) compared to 8.1W now.

I think by adding a diode (inside the feedback loop) as the output of the LM358 amp I can achieve a close to ground rail output. Output load is 2.5K resistive.
The supply is 12V, subtracting the LM358's 1.5V limitation from the top rail = 10.5, another 0.7 diode drop leaves 9.8V, which equates to a max of 9.8A sensed which is fine for a 0.5 to 9A range.
 
The diode may cause some problems with temperature and non-linearity.
The LT1013 has much better specs than the 358, so you might think about that one.
 
My favorite current sense amp is the INA195 from TI.

Even with a 5V supply, it has a common mode range from -16V to +80V. This allows you to measure current in a path other than ground. That may, or may not be of value to you.

The 195 has a gain of 100, but family members are available with gains of 20 and 50.
 
LTC6102. Chopper stabilised current sense amp. it is the input offset voltage that will kill your measurement accuracy, especially at low currents. Fit this part and never look back
 
Hello there,

The LM358 will have some significant input offset when using for signals as low as 5mv. You'd have to find a way to adjust that, possibly with a resistor network and voltage reference.
Yes, the output doesnt go exactly to ground, so the diode in the feedback is used, and this forces an extra 0.5v drop from output pin to actual output signal terminal, which means the output of the op amp does not have to go to zero. The diode does not introduce any temperature problems because it will be in the feedback loop.

The LM358 is fun to work with because it's so cheap, but if you decide to go to a higher priced device then by all means be sure to explore the chopper stabilized op amps. I had one from Microchip that had only 10uv offset (gain of 10), and with no adjustment necessary!
 
Last edited:
The LTC6102 seems capable but costs 15% more than the ACS 713 and still requires a sense resistor. I ended up using the ACS713 ratiometric linear sensor at full bandwidth, feeding into a LM358 subtracting amplifier to produce a linear 1V/A output.
As the current to be measured is pulsed, the voltage output is pulsed as well so that goes thru the second LM358 opamp which acts as an integrator & comparator (control seems more stable than using a low pass filter) allowing avg current set point control. All seems well in operation and there is a side benefit of isolation from the current bearing circuit and also insignificant voltage bounce in that circuit, 1.2 m ohm x 10A = 12mV. I used to get up to 1.2V bounce before.

Thanks for the alternatives!
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top