The ADAU1401A data sheet Rev A doesn't specify the ADC inputs current in its Absolute Maximum Rating section. Could ADI provide this missing information?
The maximum input current to achieve 0dBFS from the converter is in the datasheet. If you exceed this the ADC will obviously be clipped. After that it is the voltage at the pin that will be important. It cannot exceed the AVDD voltage. AVDD+0.3V will start to forward bias the ESD protection circuits. This is to be avoided.
Thanks for your answer.
Usually, either the input voltage or the input current must be limited to avoid damage to the chip. And those two values are given in the Absolute Maximum Ratings section of a data sheet.
For example for the ADAU1978, Analog Devices specifies the maximum input voltage and the input maximum current of +/-20mA.
It’s my understanding that if the input voltage is not limited and the protection diodes are activated the current should be limited to a safe smaller value by a series resistor.
You stated that the activation of the protection diodes is to be avoided. Could you elaborate more on this? Is the analog input stage of the ADAU1401A much different to the input stage used in the ADAU1978?
Looking at the schematics for Analog Devices evaluation boards for ADAU1701 it can be seen that there is nothing there to prevent the analog inputs from being driven beyond its 0V and +3.3V supply rails (when the audio source is powered from +/-15V for example). The protection diodes will be activated and the input current will be limited to a few mA by the inputs series resistors.
In our case we have an operational amplifier based input stage powered from +/-15V (to be able to handle large input voltages) which through a potentiometer feeds a voltage follower based on the second opamp of the dual amplifier chip. Hence, both are powered from +/-15V. The voltage follower feeds through 6.8k series resistor analog input of the ADAU1401A. In normal operation the pot would be set as not to exceed the ADAU1401A full scale voltage, but there is no way the user could not over drive the ADAU1401A analog input if they wish to do so (yes, the audio would be clipped by the ADC).
The only way we could prevent the ADAU1401A analog input protection diodes from being activated would be to use as the voltage follower a separate opamp powered from the same +3.3V voltage as the ADAU1401A. As this would increase the cost and prevent flexibility of different build options (different number of channels) we would rather try to avoid this solution if not necessary. With 15V and 6.8k resistors the maximum input current would not exceed 2.2mA.
It would be appreciated if you could clarify the following:
1. Can driving the analog inputs of the ADAU1401A supplied with +3.3V AVDD from a source providing up to 30Vpp audio signal with the inputs sinking and sourcing up to 2.2mA peak currents cause permanent damage to the device or reduce its reliability?
2. If the ADAU1401A two audio inputs are used as a separate mono audio channels can over-driving one input with such a signal significantly affect the performance of the other input or the device as the whole?
Both ADAU1401A audio inputs (ADC0, ADC1) have 6.8k external series resistors for 0.88Vrms 0dBFS.
Q1, Yes, it will degrade the reliability and eventually damage the part.
Q2: Yes, I would expect the performance of the other channel to suffer.
Is there a reason why you do not want to pad down the input to handle a 30Vpp signal and stay within the AVDD voltage?
I did just read your earlier post. You have me thinking about something. I will check on something when I get a chance and get back to you.
I had to check with one of the designers. The ESD circuit is right at the input to the die as expected. So the internal 2k ohms of resistance is after the ESD circuit. So the voltage on the 2K resistor cannot exceed the AVDD.
So roughly if the pin is sitting at the CM voltage of 1.5V then roughly sinking or sourcing current through the 2K ohms to add or subtract 1.5V would be the limit. Yes, you can go another 0.3V above that but to stay safe I would hold this to +- 1.5V.
So that means that the pin cannot take more than 0.75mA of current. before the voltage at the resistor starts to exceed 3V or 0V. So you would need to hold the input down to that level.
If you use the 18k ohm input resistor we list then with 15V input the current would be limited to 0.68mA. So it seems that you will be fine using an 18K resistor but not the 6K8 you are planning to use.
Thanks for your answers to my two questions and the following explanations.
I’m still a bit puzzled about the analog inputs protection fragility (by the way neither the analog inputs maximum voltage or current are specified in the AD1401A/ADAU1701 Absolute Maximum Ratings sections of their data sheet). To me normal working conditions and absolute maximum ratings are two different sets of requirements. We are not talking about continuously sinking/sourcing 2.2mA of current through the ADC inputs only about an unusual situation which if and when occur should not cause a permanent damage to the device.
But, if the maximum analog input current of the ADAU1401A is recommended not to ever exceed 0.75mA then our circuit needs to be modified.
As our equipment may be driven by various third-party pre-amps or mixers which may be +/-15V powered it needs to be able to handle input voltages up to around 10Vrms. At the same time the expected sensitivity should be around 1Vrms (hence the potentiometer before the ADC input). Increasing the ADC series resistors value to 18k would change the ADC FS voltage to 2Vrms.
For this reason a better option seems to be to change the opamp driving the ADC to a rail-to-rail one powered from +3.3V (the same which powers the ADC AVDD). In this way, the ADC inputs will never go above or below the supply rails, but we will still have some flexibility (limited by the 3.3Vpp range) of adjusting the FS voltage by changing the series resistors value.
Retrieving data ...