can you have a look at my eagle board layout - the Guru's mic amp

[danrogers]

Hi Bonuk, thanks for that advice!

Ok so, I will need a diode in series with the op amp output, and also a non polarizing cap in series?

Then the output of the cap (with a high value? resistor connected to gnd) could be connected to a mosfet, which would then act as my switch?

does that sound right?

I think I will have to look into how to calculate capacitance!

 

[audioguru]

The filter capacitor at the output of the rectifier diode must have its negative wire connected to ground so it filters the pulses out of the rectified audio. Then the positive wire of the capacitor and the cathode of the rectifier diode produce a slowly changing DC voltage with the peak voltage almost equal to the peak voltage of the audio signal.

 

[Boncuk]

Hi Dan,

here is a modified circuit acting as sound activated switch.

The amplifier output is rectified by D1 (BAT85 Schottky diode). The output voltage charges C3 and if the voltage level is high enough the MosFet transistor (T1) conducts and switches on the load. If no more sound is present R6 slowly discharges C3 and the transistor turns off.

For C3 and R6 experiment with different values to suit your needs.

T1 is located at a position to fit a heat sink if necessary.

Boncuk

 

[danrogers]

Hi guys. Its making sense now how the output stage will work

Bonuk, thanks for drawing that up! You are very skilled at routing, I dont seem to have that creative streak.

One thing i'm unsure of, is that I need to switch from 5v to an input on the encoder im using. From your schematic I don't think that would be possible would it?

But could I simply connect T1 between my 5v+ supply and the pin in question on the ic, and when its activated, the connection would be made?

once again thanks all, you are great

 

[audioguru]

The Mosfet output goes to ground with a high current possible but the encoder you want to drive needs a +5V input at a low current. So the Mosfet is not needed. You need to limit the output of the rectifier to no more than +5V to feed to the input of the encoder. With a 9V supply and a schottky diode, the rectified output will be about 7.4V which is too high for the encoder. This voltage will drop as the 9V battery runs down.

What will you do with the encoder/decoder without sending data through them? Do you have a transmitter and a receiver? Without data then I don't think there will be an output.

 

[danrogers]

I see, thanks guru. The encoder im using needs 5v+ aswel so I will need to get a 5v supply in there somewhere.

Yes I have a transmitter and reciever to use with them.

I dont quite understand the question what will I be doing with the encoder/decoder without sending data?

 

[audioguru]

The encoder receives the first pulse group on the4 inputs that are the encrypted data to turn on the transmitter and receiver. There will be no output from the decoder yet. Then the four data pins on the encoder send parallel data through the transmitter and receiver then the decoder outputs serial data.
I think.

 

[danrogers]

From looking at the data sheet, the decoder converts the encoded serial data stream into parallel.

I just hope it all works lol. Hopefully going to etch the board and do some testing over the weekend.

Considering I will need to have 5v+ for the IC, would it still not be best for me to use a transistor to switch 5v on the input pin rather than trying to stabilize the op amp output to 5v?

 

[audioguru]

A transistor as a switch inverts the signal then you must invert it twice with two transistors.
Simply use two resistors as a voltage divider to reduce the voltage from the rectifier.

 

[danrogers]

Guru, could you kindly explain what you mean when you say a transistor as a switch would invert the signal?

Is there not a way to have the input signal terminating at the transistor or making sure it doesnt go above 5v, but when high enough making the emitter and collector pins go short?

its the schematic of a PNP transistor that kind of makes me think that

 

[audioguru]

an NPN transistor as a switch has its emitter connected to 0V and a resistor from its collector to the positive supply. Its base has a series current-limiting resistor.

When the base resistor has a 0V signal then the transistor is turned off and the collectore is at the positive supply by the collectore resistor.
When the base resistor has a positive signal then the transistor conducts and its output at the collector goes to almost 0V which is the input signal but inverted.

You can use an NPN transistor as an emitter-follower that does not invert the signal. But its emitter voltage is about 0.7V less than its base voltage.

 

[audioguru]

an NPN transistor as a switch has its emitter connected to 0V and a resistor from its collector to the positive supply. Its base has a series current-limiting resistor.

When the base resistor has a 0V signal then the transistor is turned off and the collectore is at the positive supply by the collectore resistor.
When the base resistor has a positive signal then the transistor conducts and its output at the collector goes to almost 0V which is the input signal but inverted.

You can use an NPN transistor as an emitter-follower that does not invert the signal. But its emitter voltage is about 0.7V less than its base voltage.

EDIT:
A PNP transistor also inverts. when the base voltage is high then the output is pulled low by the collector resistor.
when the base voltage is low then the transistor turns on and makes the output high.

 

[danrogers]

Hi there, sorry I think i was editing my post as you where writing the reply.

Thanks very much for that explaination. I think I understand how I could not use a PNP transistor in the method that I suggested now.

However, could I use a NPN? From your explaination, I conclude that I could use a one, rectfy the amp output and curent limit, then a high signal on the base would flow through to the emitter , but 700mv less than the volatge on the base.

Is that anywhere near correct?

 

[audioguru]

You don't need an NPN emitter-follower transistor. The output from the rectifier does what you want but its voltage needs two resistors to reduce it.

 

[danrogers]

Ah right ok, sorry for barking at the wrong tree there Guru!

So, one diode should do the trick and only let through positive side of the ac waveform correct?

As for the voltage divider, would it be set up like the image ive atatched, with the v in coming from the diode, and vout going to my encoder?

thanks again!

 

[audioguru]

Your circuit might make a pulsing DC voltage because I don't see a filter capacitor to ground at its input. It might filter better if the filter capacitor is at its output.

 

[danrogers]

ok thank you. i try one on the output to smooth the spikes then. something like a .33uF or something?

 

[audioguru]

You need to calculate the value of parts, not guess.

 

[danrogers]

I'm not sure how to implement formlue to this circuit really

I guess I need to get the amp built and see what it's outputting and go from there?

 

[danrogers]

Guru, I've just been reading back through all the advice you have given me while I wait for the components to arrive and you had said the following...

''With a 9V supply, a TL071 opamp clips (severe distortion) when its output is about 3.3V peak or 2.3V RMS.''

Surely that means that I'm not going to reach the 5v needed to trigger my encoders input or am I in-correct?

Thanks again!