EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: m98 on July 18, 2015, 06:32:56 pm
-
Hi,
I'm developing kind of an headphone amplifier, can't reveal the application, but at least it's not intended for audiophoolery ;)
As I want to have the device free of any buttons or switches because that would increase the complexity of mechanical design, there must be a different solution to switch the thing on or off. Both input and output are 3.5mm TRRS sockets wired according the CITA-standard, the functionality of the special contact must be preserved. Using just a jack with switching contacts would give the need of disconnecting the device every time from the headphones
Basically the circuit I'm looking for should reliably(!) trigger the shutdown-input of a voltage regulator while there is line-level audio input on at least one of the two channels, while being able to be powered by the unregulated power output of a single LiPo cell.
Thanks for your replies
m9898
-
The control pin on your regulator, should it be at 0V or Vcc to enable the output?
-
Sorry, forgot saying that. It should be on Vcc to enable the output.
-
Thanks for your answer. Just as you mentioned op-amps, what about using a comparator with an RC-Lowpass?
-
How about a dedicated squelch chip as used in two-way radios?
-
Please see the following schematic diagram:
http://electricaltechnology.org/wp-content/uploads/2014/10/Clap-Switch-2.png (http://electricaltechnology.org/wp-content/uploads/2014/10/Clap-Switch-2.png)
As you are feeding in a signal at line level you do not need the first transistor so feed your signal into the base of the second BC547. If you want perfection then duplicate the transistor amp, one for each channel, and feed the signals through two capacitors into the 555 trigger pin
The Q output pin will go high for a period set by the 47 K resistor and the 100uF capacitor, this can drive the control pin of your regulator.
-
The squelch chip sounds interesting, but I don't think that there's a chip which is only dedicated to squelch. Also I didn't find one which gives something like a logic signal when there's audio input above the threshold level. I also like the 555 based circuit, but it has higher part count than my current solution.
Done a simulation on the comparator circuit, looks quite promising. Only flaw is that the voltage tends to get to the half of Vcc because, as assumed before, on a sinewave input the comparator just outputs a square wave with 50% duty cycle (ok, that should've been obvious...). So it needs another comparator after the lowpass to get a reliable logic high. Does anyone know an extremely low power and small package quad channel comparator?
-
Looks like you are trying to build your own version of relay control over audio or something similar: http://www.fmsystems-inc.com/cw/Details.cfm?ProdID=273&category=4 (http://www.fmsystems-inc.com/cw/Details.cfm?ProdID=273&category=4).
Problem is so common that it looks like a godzillion of solutions were designed over years. In almost all cases I have seen and had a chance to poke my fingers into they use high frequency channel like 20kHz+ and then filter it out from audio. Frequently Manchester encoding is used so that you have stable high frequency (20kHz+) tone regardless of data bits transmitted.
What can complicate things now is if "F" class (PWM basically) amplifier is used, also many sound cards have sampling rates in 192KHz range so it is all going to be recorded etc.
If you need only one command - pickup/hangup signal on many cell phones is sent from headset to a phone by grounding one of channels for short pulse-like period.
-
ham radio vox switch.
-
The VOX is what I proposed using the 555 timer but the OP thinks that the parts count is too high. After eliminating the first transistor stage I see six components including the 555.
-
The VOX is what I proposed using the 555 timer but the OP thinks that the parts count is too high. After eliminating the first transistor stage I see six components including the 555.
Ok, sorry for being wrong on that. I'll build up both on a breadboard to see how they work.
-
What if there is simply a couple of minutes of absolute silence as audio input?
-
Well, if there's no signal, then the analog components must not be supplied with power, as there is nothing to filter and amplify. As I said, It's just a special kind of headphone amplifier for normal music played by a smartphone, mp3-player or whatever regular audio source.
-
It's always going to take some time for everything to get going again ... but I guess a couple 10s of ms don't matter for audio.
Any way, so what you want is a wake up detector with a sensitivity of <100 uV over a bandwidth of 1 Hz to 20 kHz ... and I assume no more than a couple uA of current consumption. Part of a solution could be the TS12011 from touchstone, a nice ultra low power opamp+latching comparator. Purpose designed for wake up detectors, but the GBW of the opamp is not nearly enough for the sensitivity you need. Probably best to use it as an envelope detector without gain.
So you'll need a discrete pre-amp and it needs to not get ridiculously noise the moment the input goes high impedance, so the first stage needs to be a JFET. Dunno if discrete JFETs have much amplification/bandwidth at such low drain currents (SPICE says yes, for the BF862 ... but can't trust SPICE in these corner cases). You'd have to breadboard.