I am experimenting with some ADCs used like audio front-ends. I do not have any problem with digital circuits, but I would like to know if I am doing the correct way with the analog portion.
If I use for example an 8-bit ADC that accepts inputs levels from 0 to +5 only, the silence level would be at +2.5 volts (128 binary). However, my input signal level could be more that 5 volts (destroying the ADC) or could swing negative. So I am using a capacitor at the input to block dc and then I am routing the signal to a unity-gain opamp that it is supplied with a positive only power supply (5 volts and gnd for the +vcc and -vcc power pins of the opamp, and generating a virtual ground of 2.5 volts with a voltage divisor made from resistors). May I suppose that the blocking capacitor is enough to prevent my opamp and the ADC to see any negative swing of the input signal? Do I need another capacitor between the opamp and the ADC? A sine signal has the mid-point as required by the ADC, but what about if the input signal is distorted some way that appear like a square wave with duty cycle? What would be the mid-point then?
I do not have an oscilloscope right now so I am wondering how all those questions will be affecting my signal path. And the tricky point is that when I would be using an 3.3 volt ADC (audio input only from 0 to 3.3 V max), how I could apply to it a signal that comes from an +/-15 volts opamp?
Thank you very much for any help.
If I use for example an 8-bit ADC that accepts inputs levels from 0 to +5 only, the silence level would be at +2.5 volts (128 binary). However, my input signal level could be more that 5 volts (destroying the ADC) or could swing negative. So I am using a capacitor at the input to block dc and then I am routing the signal to a unity-gain opamp that it is supplied with a positive only power supply (5 volts and gnd for the +vcc and -vcc power pins of the opamp, and generating a virtual ground of 2.5 volts with a voltage divisor made from resistors). May I suppose that the blocking capacitor is enough to prevent my opamp and the ADC to see any negative swing of the input signal? Do I need another capacitor between the opamp and the ADC? A sine signal has the mid-point as required by the ADC, but what about if the input signal is distorted some way that appear like a square wave with duty cycle? What would be the mid-point then?
I do not have an oscilloscope right now so I am wondering how all those questions will be affecting my signal path. And the tricky point is that when I would be using an 3.3 volt ADC (audio input only from 0 to 3.3 V max), how I could apply to it a signal that comes from an +/-15 volts opamp?
Thank you very much for any help.
I'm not shure wat you are trying to do with that Capacitor.
But blocking DC won't prevent negative AC voltages.
You should try this
R1 and R2 are dividing the 15V to 5V.
But the most important components are the diods.
They will protect the adc. The maximum voltage on the ADC will be about +5.5 or -0.5V. When using shottky diods this voltages can be lowered to -0.2 +5.2.
But most of the adcs alow a little overshoot above the supply.
If you don't need any signal scalling. Lose R2! R1 must stay.
The signal lifting (virtual ground @ 2.5V ) should be buffered with an opamp.
But blocking DC won't prevent negative AC voltages.
You should try this
An externally hosted image should be here but it was not working when we last tested it.
R1 and R2 are dividing the 15V to 5V.
But the most important components are the diods.
They will protect the adc. The maximum voltage on the ADC will be about +5.5 or -0.5V. When using shottky diods this voltages can be lowered to -0.2 +5.2.
But most of the adcs alow a little overshoot above the supply.
If you don't need any signal scalling. Lose R2! R1 must stay.
The signal lifting (virtual ground @ 2.5V ) should be buffered with an opamp.
Thank you very much for the information about the diodes and the schematic. I am reviewing some schematics of single-supply opamps and all they have their audio input connected to system ground and the opamp input itself. I am wondering why it is not a common practice in those devices to connect the audio input to virtual ground (vcc/2) and the opamp input itself. It appears more logic to me since the input signal would then swing directly above and under vcc/2. ¿or maybe I am wrong?
Thank you very much for anyl help.
Thank you very much for anyl help.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.