Continue to Site

# how to convert a bipolar signal (goes below 0V) to a unipolar signal and vice versa?

Status
Not open for further replies.

#### matrixofdynamism

##### Advanced Member level 2
A question asked in 2010 on this website gives only part of the answer.

Suppose a signal varies between +10V and -10V and we want to convert it into a signal that varies only between 0V and 5V, how can this be achieved? Also, how can the reverse be achieved?
For the circuit that does this, what are the implications of its frequency response?

Suppose now that the signal that varies between +10V and -10V is instead to be converted to another signal that that varies between 0V and (-)5V instead of (+)5V.
Suppose now that the signal that varies between +10V and -10V is instead to be converted to another signal that that varies between 5V and 10V instead or between -5V and -10V instead.
How would the approach change in this case?

Suppose a signal varies between +10V and -10V and we want to convert it into a signal that varies only between 0V and 5V, how can this be achieved? Also, how can the reverse be achieved?

This can be done by converting the level of -10V to +10V to -2.5V to +2.5V using instrumentation with gain of 0.25 and then feed this -2.5V to +2.5V to summing amplifier with constant voltage of +2.5V. So your final output will be 0 to +5V.

for reverese condition,
This can be done by converting the level of -10V to +10V to -2.5V to +2.5V using instrumentation with gain of 0.25 and then feed this -2.5V to +2.5V to summing amplifier with constant voltage of -2.5V. So your final output will be -5V to 0V.

So similarly to change a signal of 0 to 5V to a signal that varies between +10V and -10V first summing amplifier will be used to -2.5V and then the resultant shall be put into another amplifier with gain of 2? Therefore this shall need 2 opamp stages. Is it also possible to do all this using mere transistors in a discrete circuit?

Therefore this shall need 2 opamp stages. Is it also possible to do all this using mere transistors in a discrete circuit?

Everything can be done with transistors.
However the amount of transistors, and the trimming required to cancel Vbe offsets, Hfe mismatch and temperature drifts, would be costlier and more complex than using a dual opamp like the LM358.

EDIT: Thomas Fredericksen in its book "understanding Opamps" has a method in which this can be done with a single opamp stage.
Let me see if I can find it.

Transistors? why you need to complicate the method? Opamp is the simple and best solution

Status
Not open for further replies.