AIC3204 sporadic input bias loss

Hi, Wolfgang,

Welcome to e2e, and thanks for your interest in our products!

It sounds to me like maybe you are seeing an ESD event causing the IC to misbehave. Can you scope the pins going to the AIC3204 and try to see if there is a glitch on them when the chip gets into this mode?

Unfortunately, the image you posted didn't come thru. I assume it's a schematic. Can you try again?

-d2

Hi Don!

Here's the Schematic in JPEG Format (you can drag them larger for better reading)

I hope the schematics work this time.  VPH is disconnected, so that cannot be the cause.  Also I connected TN and RN on the receptacle to ground. That didn't make a difference. I am using a dynamic microphone and just from pluging in and out I can make the BIAS of the Chip go 0V. Often takes a few or many tries but it happens sure enough. I cannot see any glitches on the scope.  So do you know of any special sensitivity of the Bias or can you figure from the schemtics that we missed out on something?

2. Question: We also figured that the protection circuit as in the schematic show above is not very efficient in protecting against 48V spikes when switching Mics when 48 V phantom power is enabled (this is a different case than the one shown above).  What would be an efficient way to protect the inputs of the AIC3204 from spikes on the input? 

Many thanks

Wolfgang

pixel2media GmbH

Wolfgang,

Looks like there is an issue with the images. Could you post them as zip?

Thanks,

J-

Dear J. Arbona!

I attached the schematic as .rar and as PDF -  I hope that is ok.

The microphone inputs to this circuit are connected to a XLR connector.  pin 2 and 3 of the XLR connector are connected to 48V via a 6.8k Ohm resistor. Pin 1 is connected to Analog GND. GNDA and GND (digital ground) is connected via an EMI bead. 

I have asked for help on two different problems:

1. Even with phantom power 48V diabled  (= 0V)  frequently the 0.9 V Bias on Pin  13/14 or Pin 15/16 goes 0 when we connect/disconnect the Microphone plug. We use a dynamic microphone. This phanomenon  "disables" one stereo channel as it puts a negative DC level on the digital output also puts a lot of noise on that channel. The other stereo channel remains in working order. After power-up reset the faulty channel is back working fine.

2. When phantom power is enabled, we have already two cases where we distroyed the corresponding input after plugging in a condenser microphone. Here ist the general question, if you can provide a good app note in how to apply phantom power to a microphone amplifier using the AIC3204. I did in the past design good working microphone amplifiers with phantom power but they always operated on much higher VCC and BIAS voltages than the AIC3204.With the AIC3204 it seems to be difficult to find the right protection diodes without introducing exessive noise into the high impedance inputs of the chip.

Thanks a lot and best regards Wolfgang

Hi Wolfgang,

For dynamic mic case, please try the following experiments:

1. Change affected inputs from differential to single ended using device registers by setting M side of MicPGA to internal common mode.

2. If the above did not make a difference, try increasing input series resistors (R92, etc.) to a higher value to limit the current.

Let me know how this goes and I'll respond back on the phantom power case.

Thanks,

J-