A.C vs D.C coupling

[Neil Groves]

I am close to finishing a function generator using the XR2206, when i scope the outputs i see the signal has a DC bias, i thought that an audio signal swung around 0v? i intend to use the unit for testing audio filters, what is an audio filter expecting to see at it's input? do i need to AC couple the signal or is a signal with a DC bias ok to use?

thanks guys

Neil.

[Mooly]

Generally an audio signal swings around the zero point.

If your XR2206 design runs on a single supply rail then it's bound to have a DC offset on it's output. Some function generator designs use split supplies and an "output stage" with a user adjustable DC offset control.

So yes... AC couple the generator output with a suitably large cap and with something like a 100k to ground on the output side to ground reference the output.

[DF96]

An audio signal will only swing around zero it that is how you arrange it. That is what coupling capacitors are for. As Mooly says, you need a highish resistance too for DC grounding.

[Neil Groves]

if i am going to AC couple a signal, why to i need to DC ground it? i'm confused.

Neil.

[stratus46]

Originally Posted by Neil Groves

if i am going to AC couple a signal, why to i need to DC ground it? i'm confused.

Neil.

A circuit requires to and from - two conductors. 'Ground' is your reference. Look at any patch cord in your system. Digital or analog has a minimum of 2 conductors.

G²

[Mooly]

Originally Posted by Neil Groves

if i am going to AC couple a signal, why to i need to DC ground it? i'm confused.

Neil.

Because the capacitor if left "open" at the output end will (measure it and see ) have the same DCvoltage on it as the generator output because the cap can not charge.

If you connect that to a circuit you are testing that has a very high input impedance it will take a long time for the circuit to settle. Your meter will do that... watch the voltage fall slowly as the cap charges via the meter.

So you are AC coupling to block the DC voltage at the generator end. You then tie the output end of the cap via a resistor to ground to reference this point to ground potential which is zero.

[Neil Groves]

There are many experiments i wish to perform once i have built my test equipment, i am enthrauled with electronics again, fascinating stuff.

thankyou for the explanation and your time.

Neil.

[DF96]

Just to clarify, you need DC grounding for the output from an item because you should assume that the input to the next item requires a signal based around ground potential. This means that most items will have coupling capacitors and DC grounding at both input and output. Half of these could be omitted, but then you would lose flexibility because only certain outputs and certain inputs would be compatible. Some 'high-end' items miss them out; this is short-sighted IMHO.

Within a piece of equipment you don't need DC grounding as you have control of what comes next. The exception is when you are switching signal inputs, when careful control of grounding can reduce switching noises.