Bipolar to Unipolar Design +-10 to 0-3.3V

Greenja

Other Parts Discussed in Thread: CC2540, LM258

Hello,

I have simulated a +-10V to 0-3.3V bipolar to unipolar circuit. Although it works in the simulator, I would like to know if there are any design issures, particularly with R1 having such a low value. It is being fed into the Analog input of a MCU (CC2540, MSP or Stellaris LM3S/LM4F). Exclude the choice of Op Amp from the analysis unless there is a crucial flaw in the choice. I have lots of them and haven't had any issues with other unipolar designs.

VCC is a precision reference 2.5V

Thanks,

over 12 years ago

0 Ron Michallick over 12 years ago

TI__Guru**** 164656 points

Greenja,

The output sinking current max is about 15V/(R8+R1) = 3.6mA; the LM258 can do that.

Converting an input range of -10V to +10V to a range of 0V to 3.3V can be implemented passively.
The attached circuit does not need a op-amp, but a unity gain buffer can be added to the output.

The value of the "A" resistors is 2.03 times the 'B' resistors.
A = B * (10-3.3) / 3.3

0 Ron Michallick over 12 years ago in reply to Ron Michallick

TI__Guru**** 164656 points

Greenja,

I see that the 2.5V supply is the most accurate supply. It could be used as the supply for my circuit but the resistor values would need to be changed otherwise the output would be centered around 1.25V instead of 1.65V.

0 Greenja over 12 years ago in reply to Ron Michallick

Guru 22270 points

Hello Ronald,

Your input was greatly appreciated. Just goes to show, you can learn something new everyday.

From you circuit I was able to not only eliminate an Op Amp, but a negative power supply. This now means I have to go and redesign all my analog input boards.

0 Ron Michallick over 12 years ago in reply to Greenja

TI__Guru**** 164656 points

Greenja,

I made an excel spreradsheet that calculates the resistors needed. The cells in yellow must be set by the user. They are gain (less than 1), output for input=0, power suppy voltage, and a third formula (currently set to A=10k, a=1/10k)

3 Resistor divider with DC offset.xls

0 Greenja over 12 years ago in reply to Ron Michallick

Guru 22270 points

Excellent work once again Ron!

This could really be developed into a useful tool. In my calculations I tried to stick with the standard 5% resistor values. If your time permits, an option would be to calculate based on the standard resistor values.

I took a look at the RESISTOR-CALC from TI and the EXCELRESISTORCALC. Combining these with your spreadsheet would be very useful. Your values seem to fall into the E192 tolerances which is more precision than I need.

But for now, your spreadsheet does an effective job and I found it very useful.

Thanks,.

0 Bharat Gupta7 over 9 years ago in reply to Ron Michallick

Prodigy 160 points

Thanks for all the help but, i'm not sure why this isn't working for my sinusoidal input signal.

0 Ron Michallick over 9 years ago in reply to Bharat Gupta7

TI__Guru**** 164656 points

Bharat,

The circuit just as well for AC signals.
Do you have more details about input range, output range, and any DC supplies available?

0 Bharat Gupta7 over 9 years ago in reply to Ron Michallick

Prodigy 160 points

well I'll describe what I'm trying to achieve,

I'm stepping down the 230VAC to 10VAC using a transformer.

now after when I connect this resistor network, there is no shift in voltage above the zero level.

just reduction in the voltage swing.

Supplies available are +15V, -15V and 5V. I also have LM1117-3.3 to get a 3.3V from the 5V

hope this information is helpful

Amplifiers

Amplifiers forum

Bipolar to Unipolar Design +-10 to 0-3.3V