using instruementation amplifier as microphone amplifier

use amplifier as microphone

hi
i am using INA163UA instruementation amplifier to detect output of dynamic microphone of 5 ohm impedance. after detecting the signal i amplify it and feed it in summing amplifier and than using audio amplifier ssm2211 i drive a headphone of 8 ohms. the INA163U isbeing with single supply. the 15 v supply of INA163UA is being controlled by on/off switch. the problem is when the switch is ON and the microphone is unplugged , INA163UA generate excesisve noise that is amplifeied and fed into the headphone, moreover the current increases also. when the microphone is plugge din the noise is low.
i have tried everything but the problem is same. i hope some one must have solution
waqas

micrpophone with single supply amplifier

waqas5304,
I see 2 potential causes. When the microphone is unplugged, the IA provides a very high impedance to the outside world 60 Megohms), making it very susceptible to capacitively coupled noise pickup. You can solve this problem by connecting resistors from both differential inputs to signal ground. These ristors can be high compared to the microphone output impedance, so that loading is not a problem. I suggest this approach rather than connecting a single resistor accross the IA differential input for the following reason: The IA must have a path to ground for the bias currents to flow. If this path is not provided, the IA could saturate due to the high input voltage provided by the bias current flowing thru the high input resistance. This could also be the reason that the Power supply current increases when the microphone is unplugged. The power supply current spec only applies when the device is operated in the linear (non-saturated) region.
Regards,
Kral

ina163u

Hello,

I would explain somewhat different, although input open crcuit is actually a keyword. The other keyword is current noise. INA163 has input stages optimzed for low source impedance (noise matching is for 200 Ω). As a result it has considerably current noise at the input, requiring low impedance termination of both input and also Rg terminals.

With microphone disconnected, some extra noise can't be avoided, unless not shorting the input with a switching connector. Depending on the microphone's specification, I would choose an appropriate low resistance parallel termination, may be a few 100 ohms to 1 or 2k.
.

Regards,
Frank

thanx for replies
i have tried to connet two resistors from high and low input , but of no use, i tried by varying the value to resistances. than i placed resistor across the high and low and the noise got less . now i am connecting 47 k ohms resistance than the noise is considerably low but it is there ,do you have any other suggestion. is this noise can be caused by the single supply operation?

Added after 34 seconds:

thanx for replies
i have tried to connet two resistors from high and low input , but of no use, i tried by varying the value to resistances. than i placed resistor across the high and low and the noise got less . now i am connecting 47 k ohms resistance than the noise is considerably low but it is there ,do you have any other suggestion. is this noise can be caused by the single supply operation?

Hello,

single supply operation with 15 V should be possibly without deteriorating noise perfomance. But how did you bias the inputs? They need to be biased at ≈ 7.5 V then. From what you reported with different resistor connections at the input, I fear the bias could be incorrect (either biasing inputs at negative supply or inputs floating).

Furthermore, the bias circuit should be designed in a way that doesn't couple interfering voltage from supply to the input, this is obviously a disadvantage of signal supply operation.

You didn't tell, if you use a symmetrical or asymmetrical connection of signal source (microphone), thus I can't suggest an exact bias circuit.

Regards,
Frank