of course i could adjust the gain of the opa and thus make the -3 to +3V reference, but why? i thought the CM106 has this value as offset...
The reason I asked is because you need a reasonable range of adjustment to get proper DAC offset correction
and that depends on the circuit you use. My circuit is a summing amp, so the offset reference (or correction
voltage) must be negative to add and cancel out the positive offset of the DAC. I suspect the Weartronics
circuit at the top of this thread might actually require a positive correction voltage, because the DAC is
applied to the positive input of the op-amp rather than being summed together with the correction voltage on
the negative input (I have never wired a circuit that way, so I'm just guessing).
Take a closer look at the summing amp circuit. When R1 and R2 are the same value, you get zero output when
both inputs have a voltage that is equal and opposite polarity, but summing amps are quite flexible- for example,
let's say the DAC offset is 2.5v and I make the offset reference -5v, all I have to do is make R1 twice the
value of R2 and I'll still get 0v at the output! Here's the formula for zero balance (E=applied voltage):
E(R1)/R1 + E(R2)/R2 = 0 and R1 = -R2*(E(R1)/E(R2))
The point is, I could make the offset reference any fixed negative value and adjust R1 to get a perfect
balance for each channel. The DAC offset does vary a small amount from channel to channel, but it's probably
not enough to worry about.
The DAC gain of the first stage of the circuit is -R3/R2. The second stage of the circuit is just a unity gain
inverter where gain = -R5/R4. With R4=R5, the output is equal and opposite of the first stage, and the total
voltage across -X & +X will be twice that of the first stage measured to GND. If R2=20K, set R3 (gain) to about
to about 32K to get a full +/- 5v for your Chinese projector...
So this is strange. the "real" voltages on the divider are coming out unaltered with a impendance amplifier / follower; but are changing extreme when i use something with gain.
Now, you're getting deeper into electronics... The ideal voltage source has zero output resistance, so no
matter what load you put on it, you get the same voltage. A real volt source such as a battery, can be thought
of as a perfect volt source with a small amount of built-in resistance in series. As the load gets greater, some
of the battery voltage starts to drop across that internal resistance, causing the final output at the load to
drop. Devices such as op-amps, regulators, and DACs use feedback which allows them to maintain a constant
voltage as the load changes (within reasonable limits). My point is this: in a summing amp, the load seen by
the voltage source is equal to the series resistor (R1 or R2). If the volt source is coming from a resistor
divider, than that source resistance will affect the final result, making it less predictable. In an op-amp
voltage follower, the input presents almost no load at all to the volt source, yet the output is regulated,
and that (or a regulator) is what you need to use for your offset reference source.
Look on the bright side, if you bought a correction amp that was ready-to-go, you would have missed out on
the fun and frustration of learning some electronics...
