Voltage offset

If you are getting 4.45 Vp-p with those resistor values, there must be an additional load on Vout of 2.4 kohm, or there isn't 12 Vp-p going in. The indicated circuit actually gives 0 to 5.5V. Whether that's actually unsafe depends on the ADC maximum input ratings. Another possibility is that there is source impedance in the sine source, but that would show up as a reduced voltage at its terminals. Nothing shown would put noise on VCC, if you mean the power supply voltage. If you are worried about noise on the attenuator output, going into the ADC, the impedance isn't very high, so it wouldn't pick up much noise in a short distance. However, any noise on the 5V offset source will appear on ADC input, attenuated only by 1/2.2.

--
John

55% of any noise on your 5V source will appear on the output you are digitising, while you are scaling down the sine wave you want to digitise to about 46% of its original value.

These ratios are determined by what you want to do, not the method you chose to make it happen.

If you use a good low noise reference source to generate the 5V offset, the additional noise from the off-setting source is unlikely to be a problem. If you use a 5V rail as your offset source you could well be in trouble.

The circuit you have drawn is simple, but it presents a 2k2 to your sine wave source, and 545R to your A/D converter. If you sum the sine wave signal into an op-amp-gnerated virtual earth, along with a negative off-setting voltage, you could generate your desired offset sine wave at the low impedance output of an operational amplifier.

If the A/D converter ground and the sine wave source grounds aren't identical or constant, you can add a few more resistors to the circuit to make a subtractor that can compensate for this varying difference - if only fr frequencies belwo the bandwidth of the op amp involved.

-- Bill Sloman, Nijmegen

Pardon me, but I don't understand the term "rail", do you mean my power supply source? If so, yes my 5V *reference* is directly from my main power supply source.

From the thow reply a see, once said it's ok and you said I can be in trouble!!! What's the real story?

I have design a circuit using a op-amp to obtain the same result, but op-amp is *expensive* compare to the above circuit, but if a running after trouble with this design, I will go for an op-amp.

I would highly suggest doing the attenuation/level shift using a good, high bandwidth opamp to drive the ADC input. The ADC sucks a narrow, high peak current when it samples and a 500ohm source impedance will surely wiggle in this condition. A good high bandwidth opamp (say an AD8031) would make a good buffer for the ADC input. Running the opamp off 0-5V will also protect the ADC from damage if your sine input signal gets zapped or goes higher than the expected voltage.

This is about the same as your circuit but greatly attenuates the 5V noise to be expected in a digital system... View in a fixed-width font such as Courier.

. . . . +5V . | . [560] . | . +-----. . | | . [560] | . | | . >---[1.3k]---+-------->

. | . === . 0.33u| . --- . /// . . . . . .