My son runs a recording studio, so he can't use a doorbell that generates sound while he is recording. I decided to hack a wireless doorbell to provide him a way control the ringer volume volume when he is at his mixing desk, and that would flash a bright LED whenever the doorbell rings. I came up with the circuit below:
Unfortunately, when this circuit is put in place, it destroys the output amplifier that drives the ringer's speaker I am puzzled as to why this happens. Testing with just the 10k ohm variable resistor attached across the speaker terminals (and attaching nothing else) works as expected, with sound from the speaker being attenuated as the pot is rotated. If I substitute a signal generator (peak 3.5 V sinusoid) for the doorbell ringer amp, and feed the output from the signal generator into the full circuit, every thing works as it should, and the Arduino blinks the LED (not shown in the circuit diagram) whenever it sees the sine waves from the signal generator. My tiny oscilloscope shows no negative voltage at the A0 pin on the Arduino when the diode is in place. Maximum positive voltage of the half-wave at the pin is around 3 V.
I wonder if the problem is that the diode that I have put into the circuit to prevent negative voltages from the amp being sent to the analog read pin A0 on the Arduino is causing a DC voltage to be fed back through the Arduino onto the return leg to the amplifier, and the amp can't tolerate that. I would have thought that the internal resistance of the Arduino would be high enough to prevent that from being problematic, but what do I know? If that is the problem, would substituting a 1M ohm for the 1K ohm resistor cut any DC voltage and current down enough to prevent release of the magic smoke?
Being something of an electronics newby, I realize I could could be completely wrong in my guess at the cause of the problem. No doubt the entire approach could be garbage. But if my approach is unworkable, do you have any suggestions for another approach?
Oh as soon as I posted, I saw the REAL problem. When the pot is roptated one way, is effectively a dead short... I will fix that, and post again. It will take a couple of days as now I need to order another one of the little $7.00 wireless doorbells. To replace the two that I have killed.
You'll also need a "pull down" resistor on the Arduino's analog input. Otherwise the input can "float up" and you can get a false trigger. Use something bigger than the 1K resistor, maybe 10K or 100K so it doesn't load the circuit too much.
I'd be surprised if a wireless doorbell has an 8 ohm speaker. If you want to be able to control the sound level you should use a pot in series with it, or with the speaker connected where the diagram shows an 8 ohm resistor.
a 1M pull-down may expose the arduino adc to noise - so I'd try around 10k.
I'd also remove the direct connection between the cathode of the diode and pin A0 and replace it with a 10k resistor and I would consider putting a 5v zener diode between A0 and GND (cathode on A0) in case that amp can output in excess of 5v.
In the first drawing the speaker is connected across the Amps output , and the lower connection of the speaker goes to common GND between the Amp and the Arduino.
The second drawing: The speaker is connected to the Amp through the Pot and the Pot will not vary the level to A0.
The Amp and the Arduino does not share a common GND anymore.
Thanks johnerrington. I need to understand your suggestions better. The way that I understand that this circuit to operate in terms of current is that current flow through everything to the right of the pot in the diagram would be neglible, and essentially equivalent to
As a speaker is current driven this seems logical to me. So in terms of current, the speaker is in series with the pot. The speaker's resistance measures at about 7 ohms, so I assume that its impedence is probably 8 ohms. So current flow through the speaker will vary from near zero when pot is turned to the left to a peak of about 300 milliamps when the pot is turned all the way to the right as total resistance is 16 ohms. This should provide volume control I would think. Or do I have it wrong? I understand your point on changing from a 1M pull-down to 10k, and will give that a try.
6v6gt: I will add a 10K resistor between the diode and pin A0 and add a zener diode from pin A0 to ground.
mikedb: You comment that the Amp and the Arduino do not share a common GND anymore. That is true, but in this case, GND is just a reference voltage on one side of a resistor that is compared to the voltage on the other side of the resistor where A0 is attached, isn't it? I don't need to know the voltage relative to a common ground. No doubt my ignorance is showing again -- could you please explain a bit more?
noobmastha and Paul_KD7HB: You ask why I am working on this project when I could buy a doorbell for the deaf, and why I am using an Arduino. There are actually a number of reasons:
I want it to be possible to turn the ringer on and off easily with a control that is mounted on the recording studio's mixing desk console. You can mute and unmute most ringers that I have seen but you have to push buttons on the ringer to do so. This is klunky, and I do not want the ringer near the mixing desk.
I eventually hope to have the studio's recording DAW send midi or OSC messages to the Arduino so that the ringer is automatically muted and unmuted when the 'record button' on the DAW is toggled.
The wireless doorbell I am working with costs $8 and the Arduino Pro Mini clone that I am using costs about $1.50. So it is a cheap solution that I can play with without worrying about the cost of damaging something.
My knowledge of electronics is somewhat rudimentary, whereas I have spent quite a bit of time writing computer code, and I find it much easier to think up software solutions using microcontrollers than to come up with circuit solutions. Basically my thought process is often something like "please, please, just let me get the bloody signal to the microcontroller so that I can get something done".
I can modify the behavior of my DIY device or add capabilities more easily using the Arduino than I can by changing circuitry.
There is no fun in buying an off the shelf component, and there is no learning involved. The feeling of accomplishment from completing any DIY project is often the main reason for doing it.
I modified the circuit as suggested, except that I substituted a 10 ohm for the 8 ohm resistor in the schematic as was the closest to 8 ohms that I had at hand. Since I am still waiting for a doorbell to replace the one I destroyed, I decided to test the circuit do some testing using my using my signal generator to simulate amp output of the ringer. Also, the speaker itself was not in the circuit, as I substituted a multimeter for it so that I could measure current passing through the circuit. I then tested the circuit, and all appears to be working as expected.
I set my signal generator up to send 1 second bursts of 6V peak-to-peak sinusoidal waves and connected it in place of the ringer amp. When the pot was rotated fully clock-wise, 196 mA of current was passed through the device. In other words, sufficient current would have been passed to the speaker to operate if it had been in circuit. No blinking of the LED occurred.
When the pot was rotated fully counterclock-wise, about 15mA of current was passed through the device, which I believe will be insufficient to drive the speaker to make audible sound. The LED blinked as expected.
Based on these tests, I would think that there should be no problem when I connect the doorbell ringer amp instead of the signal generator. Am I missing anything?
