Project for my Kids - I'm stuck!

[tedsdad]

Hi all,

I'm new to this forum.

I've just started dipping my toes into the electron stream and it's not as easy as I thought it was.

I'm a technical guy (web designer) but I've gotten lost on my project. Can you help?

I've got a small mono amp operating on 9 volts. It has a sound input which is powered by the amp – about 1.2v. It then outputs to a 1 watt 8 ohm speaker. This all works perfectly.

Now, what I want to do is connect the output of the amp to a voice recorder circuit. The voice recorder is designed to play a sound only (we use the VR because we can record funny sounds on it). It has a momentary switch which plays the digitised sound. I hooked up the amp output to the momentary switch input and got the voice recorder to make the digitised sound based on the amount of voltage being generated by the amp. The DC voltage is 0.00 and the AC runs between 0.00 and 600mV depending on it the output from the amp.

Here's what we want to do:

When the sound plays (active AC voltage) we don't want the voice recorder circuit to play any sounds.

When the amp stops outputting sound (no AC voltage) or the voltage drops below a certain threshold, we want it to trigger the voice recorder and play it's digitised sound (probably a fart sound or something). The VR sample will run for about 1 to 2 seconds.

I know this should be simple but I just can't get my head around it. Not sure if the voltage needs converting to DC and then storing in a capacitor to release when the voltage drops (no sound from amp). That way could it make a momentary circuit?

So, to recap. There are two circuits:

1 watt mono amp – operates between 4.5v and 9v.
20 second voice recorder which has a built in mic and speaker – operates between 4.5v and 6v.

Then there is the external speaker for the amp and an input – either a mic or an mp3 player.

My electronics knowledge is basic so any help would be most welcome. Treat me gently though because my knowledge does not extend very far. This is a project for my kids – I promised I would do it!

Here's hoping,
Rees.

 

[mstovenour]

Hi all,
So, to recap. There are two circuits:

1 watt mono amp – operates between 4.5v and 9v.
20 second voice recorder which has a built in mic and speaker – operates between 4.5v and 6v.

Then there is the external speaker for the amp and an input – either a mic or an mp3 player.

Rees.

Hi tedsdad,
Unfortunately, I'm going to start with more questions than answers. Let me make sure I understand exactly what you want to do. I attached a drawing that shows two ways to connect the amp to the digital recorder. Does the drawing look like what you want to achieve? And do you already have the Amp and Digital Recorder?

If it is drawing is reasonable then the project has two design tasks and I need to ask a couple of other questions.
Design task #1 - Sound input for the digital recorder.
Let me just ask, have you figured out how to drive the amp with sound and record the sound at the same time? There are a bunch of ways to do this. The simplest is to just speak into both microphones at the same time; Option 1 in the drawing. The most architecturally elegant way is to disable the microphone on the digital recorder and drive the amp output directly into the voice recorder; Option 2. Either way will work; but the approach for Option 2 is somewhat harder depending upon the features of the voice recorder and amp. For instance does the amp already have an "output" jack? and does the voice recorder already have in "input" jack?

Design Task #2 - Create a sound activated switch that triggers the digital voice recorder.
There are a couple of ways to do this and you can probably find all kinds of example circuits. The simplest method would involve a circuit that drives an electromechanical relay to short across the digital recorder trigger switch. With more effort it might be possible to use a simple transistor or MOSFET on the output of the voice activated switch in place of the relay. The reason this is harder is because you will need to reverse engineer the trigger circuit in the voice recorder to see exactly how it works. If that circuit is something simple like shorting the trigger to battery minus or plus, then a transistor based solution would be very easy. Otherwise the relay might be easier.

Voice activated switch - Circuit claims to use a detector that only activates on voice and ignores background noise. Might needs a small addition to drive a relay.

Ramsey VS-1 VOX kit - Complete kit. Even shows a hookup for a radio (similar to a voice recorder with trigger switch). They show that their circuit has enough gain so that you could eliminate the amp all together. This is probably your best bet. I saw it on ebay for ~$7.

 

[tedsdad]

Thanks so much getting back to me. Did you get my user name – Ted's Dad! As it's for Ted I thought it was fitting to choose this username – any enough of my eccentricities!

Okay, I'll try and work through your questions.

Yes, I have both the 1w mono amp and the 20 second digital voice recorder / player (I'll refer to it as DVR). Presently they have their own speakers and battery packs.

You're diagrams are very close to what I want to do – impressive. I've attached images of the modules I'm trying to use.

More info:
The amp, has a jack input / output for mic (I've added a jack plug) and speakers.

The DVR has a 40mm mylar speaker attached, button for record, button for play, an LED to indicated playback and a battery pack. I plan to keep the mic and the record button so Ted can change the sound. This will all be made much more compact.

Ideally, what I'd to do is run both modules from a 9v battery. This would require a voltage regulator to step down to the DVR. The amp can still run at 9v.

Next, I'd like the speaker output leads from the DVR to be connected to the output jack of the amp (maybe with an inline pot to control gain). I then just have one speaker. I can then independently control the gain of the amp and the DVR. Why don't I want to route sound through the DVR? Well, the sound quality is not that great in the circuit. The amp is a really nice design and has a very clean input and output.

So, to recap. When sound plays through the amp, nothing happens with the DVR playback. When the amp becomes silent or music / speech stops playing, the DVR is triggered on the play contacts to make the sound. It would need to timed so it didn't last too long – maybe 300ms and it would then have to stop.

I did get the DVR pulsing sound by connecting the jack output contacts from the amp to the DVR play button contacts. Unfortunately, this is the reverse of my goal.

I thought it would be as simple as a resistor and a capacitor connected to the DVR play switch from the output jack terminals of the amp. It didn't work though. My idea was that the cap would store a charge and the release it when the amp output voltage ceased. What I forgot was that it is all AC voltage outputting from the amp (measured DC and nothing outputting from the amp). So, the cap wouldn't work for this, it would only smooth the voltage and allow it to pass through to the play switch of the DVR.

Does this all make sense? I'm trying to add just a couple of small components to do this – well in my world anyway. The real world maybe different. I need to keep it all very small for the application.

I thought about transistors to do it but I have no experience with them. Also, don't they need a DC voltage to trigger them?

The schematic is a bit beyond my present skill – I am learning and will continue to do so. BTW, a 555 chip came to mind – would this work with a transistor or miniature relay? I like the idea of a transistor as it is so small.

The plot thickens. Any more suggestions would be most welcome.

Cheers,
Rees.

 

[mstovenour]

Next, I'd like the speaker output leads from the DVR to be connected to the output jack of the amp (maybe with an inline pot to control gain). I then just have one speaker. I can then independently control the gain of the amp and the DVR. Why don't I want to route sound through the DVR? Well, the sound quality is not that great in the circuit. The amp is a really nice design and has a very clean input and output.

I think it will be difficult to drive one speaker with two separate output stages. It is possible but I've only seen it done with coupling transformers. Maybe you can use a pot to attenuate the DVR output and then bring it into the amp input along with the mic. Summing at the input to an amp doesn't normally require exotic impedance matching.

So, to recap. When sound plays through the amp, nothing happens with the DVR playback. When the amp becomes silent or music / speech stops playing, the DVR is triggered on the play contacts to make the sound. It would need to timed so it didn't last too long – maybe 300ms and it would then have to stop.

I didn't understand the application before but I think that I'm starting to get it. You want to:

1. Record a sound using the DVR's mic and existing record button
2. Play music or speak through the amp using it's speaker
3. When the sound source stops, generate a short pulse on the DVR play button

I thought it would be as simple as a resistor and a capacitor connected to the DVR play switch from the output jack terminals of the amp. It didn't work though. My idea was that the cap would store a charge and the release it when the amp output voltage ceased. What I forgot was that it is all AC voltage outputting from the amp (measured DC and nothing outputting from the amp). So, the cap wouldn't work for this, it would only smooth the voltage and allow it to pass through to the play switch of the DVR.

Ah, yes. That idea is called a peak holding circuit but you forgot the diode. If you place a diode between the amp output and the capacitor, then the current will only flow into the cap, and not back out through the amp.

At first I thought you needed a simple VOX and wanted to find a circuit that was already fully baked since you are new to electronics. Now that I understand exactly what you want, I can see why you tried to use a peek hold and why you are suggesting a 555. I think you are going down the right track. I think that a peek hold circuit should be used to disable the detector output while the music is playing. A relatively long discharge time constant on the peek hold can be configured that will prevent triggering during the music and give a configurable delay before the DVR is activated. The 555 is basically a level triggered flip-flop. A flip-flop is a good idea for a one-shot pulse circuit needed to pulse the play button.

I still think a relay will be more likely to work because you don't know how the buttons are connected to the DVR circuit. Now if you can determine that one side of the play button is connected to the battery plus or minus then we have other options besides the relay.

I suggest that you hold off on the power supply and single speaker requirements. You can get a functional prototype working if you just get the trigger working. I've attached a new drawing that shows two independent amp and DVR modules where the DVR is triggered by the transition from an AC signal to silence at the amp. That should be a reasonable place to start.

Theory:
Amp output is A/C coupled to the detector by C1; this ensures there is no DC current path into the detector. This may not be needed if the amp output is a simple transistor output or already has an AC coupling capacitor. The AC coupled amp signal is then rectified by D1. R2 and C2 form the peak hold circuit where R2 determines how fast the circuit responds to silence. Raising the value of R2 will make the circuit less sensitive to short silence intervals in speech or music. Initially an AC signal from the amp will charge C2. When that signal stops (silence) R2 discharges C2. The falling edge is AC coupled through C3. C3 is necessary to ensure that the one-shot does not continue to trigger. The 555 will continuously trigger as long as pin 2 is held low. R3 will eventually charge C3 enough to keep from triggering the 555. R4 and C4 form the timebase for the 555 one shot. The forumla is 1.1 * C4 * R4 in seconds (~100ms). You can adjust those values to make the trigger longer if needed. The output from the 555 is feed to a 5V relay. If you figure out exactly how the trigger input is wired internally in the DVR then it might be possible to make this work without the relay. I've also seen circuit examples where the relay is driven directly from the 555. You can try that if the relay is very small.

 

[tedsdad]

Whoa, what a circuit. My sincere thanks for providing a schematic. I now have to get my head around it and get some help to understand how to build it. I understand (in principle) how this works.

I'll try my best to build this on a breadboard and get it working. I would rather use a PNP transistor instead of a relay due to size. This gadget needs to fit in a small box!

Sometimes, do you wish you ever agreed to do things! Oh well, I have to see it through.

Again, your input is much appreciated. I'll feedback to you on how I get on.

Cheers,
Rees.