Any audio experts? Need help configuring audio input circuit 4

[auterik]

Hello Everyone-

Thanks for reading this post as I really hope someone out there can help me in figuring out this small audio configuration problem out or lead me in the right direction.
I'm working on a stereo bluetooth transmitter project with stereo inputs (differential: L+/L- & R+/R-) but plan to use a single ended configuration for a seperate music player device to source the audio.
In the the current design the datasheet for the audio section of the chip it clearly states that the full scale audio input cannot exceed 400mV rms even when setting the input gain at the absolute minimum. So the problem is that the audio source we plan to use has a line out with an audio output of max 1V rms. So how would I have to modify the existing simple circuit to safely input the 1V rms audio signal without saturating or distorting the audio signal?

These are the things that are known for the current design that can safely input 400mV rms on the L and R differential inputs.
1. Input impedance is 130k between L+ and L- also the same for R+ and R-
2. L+ and R+ is connected in series with a 47nF cap to the respective L and R audio line
3. L- and R- is each grounded (analog) in series with a 47nF capacitor.

I have also seen other designs that were using 1uF instead of 47nF for the cap values that are in series with the audio input. What is the real effect?


Thanks again for all that read this post!

Michael

 

[itsmoked]

If the 400mV input has high impedance what about a simple voltage divider?

 

[auterik]

I understand how a voltage divider would work in that the simple circuit will make the signal smaller? The question is how would I correctly implement that into the circuit?

 

[elfgriper]

Auterik,
I remember that the output impedance of most line outs is 1k. This can be proved by installing a 1k from the output to gnd. This will give a .5v output with a 1v programmed line out. If that correct then adding a 666 ohm resister from the line out to ground will give .4v (400mv) max out. Putting 47nf cap in series with a output will only allow high freq to pass (filter) thru ie blocking the lower freq's. Adding 1uf will allow lower freq's to come thru.

 

[jimkirk]

Loading the source is not the best way to attenuate the signal. Not all sources are 1K, and some might distort with such a low impedance load. Use a resistor divider, values something between the 130K input resistance and the 1K or so output impedance. A 10K and 6.8K (6.65K or 6.81K if you're using 1% parts) resistor should do it. This won't load the source too much, and will provide a relatively low source resitance to the + side of your differential input. (You might want to look into a differential resistor divider, but you may well degrade the common mode rejection if some of the resistors aren't well matched.)

The two 47nF caps, effectively in series, and the 130K input resistance will roll off the low frequencies, will be about -3 dB at 50Hz, fine for voice but not "HiFi". A 1uF cap will lower this to about 2.4 Hz, well below the audio range.

 

[auterik]

Thanks! You guys are all so quick!
I'm more of a rf and layout engineer and really have no idea how all this audio stuff works so all your feedback is much appreciated! That's a lot of stuff, I really cant thank you guys enough. I promise when I see something I can answer in this forum I'll be right on it.

Okay here is a little more detail to what I have setup so far. My layout is complete but still trying to figure out the resistor values. I left room open so I can work with the configuration if change is needed by using 0 ohm resistors for shorts or nothing for opens. So unfortunately I won't really know for now what the results are by simply chnaging the values. But I would really like to know in figuring this out for optimal hi-fi input using conventional parts before adding the changes. (all my components are 0402 so not so easy to just change around)

This reference circuit is already laid out on the prototype design for an input connection of a mic and a line input.
So I would really like to understand what all these values mean since half aren't probaly needed especially since I'm looking to do only a audio line input. From the output of the audio source (normal headphone jack, powers 32 ohm speakers) on the board it goes through a 1k resistor in series to a 15nH inductor which is then tied to p-input which is also than in series to another 2.2K resistor tied to the n-input which is also connected analog ground.

There is another 3 resistors of 100K from P-input in series with 1M to analog ground. Also a 150K resistor is connected in series from the N-input to the connection between 100k and the 1M resistor on the P-input side.
Hope that wasn't confusing but a fairly simple circuit. IF you need to see an illsutration please let me know.

The 2.2K resistor I beleive is to set the impedance of the mic and not sure what function the inductor has. Probably can get rid of them?

What would be the disadvantage of going for a cap like 1uF and having a freq. of 2.4hz?
So what would be a good value of the capacitor for hi-fi application?
And for future reference how did you calculate that low frequency?



mk

 

[ScottI2R]

Auterik,
(PIE)=3.1415...
f=frequency
C=capacitance
Xc=Capacitive reactance

The formula can be easily transposed to figure frequency as opposed to capacitive reactance. (AC resistance in a capacitive circuit.)

I believe the formula used was a transposition of 1/2*(PIE)*f*C. Or basically, the transposition was f=1/2*(PIE)*Xc*C.

I hope this helps. AND, if you want to thank these guys enough, vote them a star! They ARE great and very helpful. They have helped me before!

Scott

In a hundred years, it isn't going to matter anyway.

 

[auterik]

Thanks everyone for your feedback!!