Author Topic: Simple amplifier circuit, how to handle power and isolation?  (Read 2885 times)

0 Members and 1 Guest are viewing this topic.

Offline DukeOfEarlTopic starter

  • Newbie
  • Posts: 9
  • Country: au
Simple amplifier circuit, how to handle power and isolation?
« on: April 25, 2018, 12:59:48 pm »
Hello,

I am building a "remote" for handheld radios, i.e. a microphone / speaker / PTT box to connect to the accessory connector of the radio.

I know nothing about audio electronics.

I want to have an amplified speaker output that can be driven from the radio's accessory power or an additional battery / power supply, switched via a ganged switch or a relay.

I chose to go for a simple single-transistor Class-A amplifier because it's simple and I don't have a high power or low idle power need. So this is the basic circuit I am going for:



I am a bit stumped on how to design the circuit for external power.

Here is where I'm at right now:



Now this can't work with the external (battery) power because there's no way for the base current to get back to its ground level. So I need to connect / couple GND and PA_GND somehow. Just how do I do that? Do I need to put an audio transformer in? Can I just connect the two grounds (radio and battery) together, with two power sources sharing ground?

Here is my whole circuit:
 

Offline Audioguru

  • Super Contributor
  • ***
  • Posts: 1507
  • Country: ca
Re: Simple amplifier circuit, how to handle power and isolation?
« Reply #1 on: April 25, 2018, 04:10:32 pm »
Your electret mic is not biased and the pots are connected backwards.
Do you want the transistor circuit to drive a loudspeaker? It is missing an output coupling capacitor but if it has one then its output power will be very close to zero with severe distortion.
 

Offline DukeOfEarlTopic starter

  • Newbie
  • Posts: 9
  • Country: au
Re: Simple amplifier circuit, how to handle power and isolation?
« Reply #2 on: April 25, 2018, 11:21:43 pm »
Your electret mic is not biased

Are you saying that my biasing circuit is wrong? I built it after doing some research, it is essentially modelled after this: Electret condenser microphone schematic.

Do you want the transistor circuit to drive a loudspeaker?

Yes.

It is missing an output coupling capacitor but if it has one then its output power will be very close to zero with severe distortion.

So is this "single-stage class-A amplifier" circuit schema I posted wrong / just a toy example? Should I use a Darlington transistor instead? I modeled the circuit after the information here: Class A Amplifier
 

Offline Audioguru

  • Super Contributor
  • ***
  • Posts: 1507
  • Country: ca
Re: Simple amplifier circuit, how to handle power and isolation?
« Reply #3 on: April 25, 2018, 11:52:42 pm »
Are you saying that my biasing circuit is wrong?
Since your mic circuit does not have a mic connected then I guessed it is connected to MIC and you are biasing the capacitor at MIC EXT.

For a class-A circuit to drive a loudspeaker then the currents must be VERY HIGH with VERY LOW resistor values and a transistor that can conduct a high current and be able to get rid of a lot of heat. I simulated your amplifier and it can drive a 10000 ohms load well but has a problem driving 1000 ohms or less.

The circuit you copied and the random 2.2k resistor value you used creates voltage divider. The collector and emitter values are the same so 1/3rd of the 5V can be across the collector resistor in series with the low resistance of the loudspeaker. The maximum peak current is (0.33 x 5V)/2208 ohms= 0.75mA. Then the peak power in an 8 ohm speaker is 0.75mA squared/8 ohms= 0.07 thousandths of a Watt and the average power is half. A cheap clock radio produces 0.5W.
 

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 12807
Re: Simple amplifier circuit, how to handle power and isolation?
« Reply #4 on: April 26, 2018, 12:12:24 am »
Its quite common to have several supplies sharing a ground.   The supplies may be derived from each other or truly independent.   However when you add a ground interconnection between two more complex circuits, you need to be absolutely sure that the points you want to join are either at the same potential or free to float relative to each other.   External connections often introduce a ground, so, in your case, be cautious if the radio has a DC power jack separate from its accessory connector.

Of the circuits on the page you linked, only the 'Transformer-coupled Amplifier Circuit' is anywhere close to being viable as a single transistor speaker driver.  Audioguru has already pointed out the impedance mismatch problems with your implementation of the first one, and even if you go for the second (Darlington) one, the load resistor required to get usable power into an eight ohm speaker (presumably capacitively coupled) is inconveniently low ant the quiescent current and dissipation unpleasantly high.   Frankly, you are insane if you want to use a pure class A audio amplifier for speaker output in a battery powered application.  The quiescent current consumption will require the batteries to be much larger and more expensive than they otherwise would be, and even so their life is likely to be undesirably short. 

Probably your best bet for comms quality fractional wattage speaker output would be a single chip audio amplifier.   If 0.1W is sufficient, consider the LM386.
« Last Edit: April 26, 2018, 12:19:41 am by Ian.M »
 

Offline DukeOfEarlTopic starter

  • Newbie
  • Posts: 9
  • Country: au
Re: Simple amplifier circuit, how to handle power and isolation?
« Reply #5 on: April 26, 2018, 01:14:29 am »
Thank you both for your replies.

Sorry for not adding more explanations to my circuit - I've been working on it off-and-on for over a month, so everything's very obvious to me... The "Accessory Connector" in the top left section is supposed to be connected up to the actual connector for the radio that you see in the insert picture. So, the MIC net is the Microphone (TX) input to the radio, and the SPK net is the speaker (RX) output from the radio. So SPK is directly connected to the Headphones and Headset connector (which I just noticed I assigned the wrong net to, SPK should be MIC_EXT) for monitoring the radio RX audio without additional amplification, and to the amplification circuit for driving a loudspeaker. The MIC net is connected to the Microphone biasing circuit, which is directly connected to an electret capsule (i.e. headset / microphone "MIC" connector) at MIC_EXT.

I hope that clarifies your confusion about the MIC biasing, AudioGuru - again, sorry for not explaining that up-front!

External connections often introduce a ground, so, in your case, be cautious if the radio has a DC power jack separate from its accessory connector.

Hi, any external (mains-derived) power to radio or my circuit would come from independent isolated power supplies.

Maybe this is something I need to research / study more. I attached a quick toy circuit of having two voltage sources driving two loads sharing a ground:



Is this a common "nothing can go wrong here" arrangement? Where would the "external connections introduce a ground" problem come into play here? Would I just need to make sure that the external (third) ground is also connected to my common ground to avoid a ground loop? Is a ground loop the specific problem I need to avoid here?

For a class-A circuit to drive a loudspeaker then the currents must be VERY HIGH with VERY LOW resistor values and a transistor that can conduct a high current and be able to get rid of a lot of heat. I simulated your amplifier and it can drive a 10000 ohms load well but has a problem driving 1000 ohms or less.

The circuit you copied and the random 2.2k resistor value you used creates voltage divider. The collector and emitter values are the same so 1/3rd of the 5V can be across the collector resistor in series with the low resistance of the loudspeaker. The maximum peak current is (0.33 x 5V)/2208 ohms= 0.75mA. Then the peak power in an 8 ohm speaker is 0.75mA squared/8 ohms= 0.07 thousandths of a Watt and the average power is half. A cheap clock radio produces 0.5W.

I haven't made any calculations for the actual resistor values I would need to use for this! I'm still at the "pick an amplifier design" stage. It sounds like my simple circuit really is pretty much useless except for demonstration purposes and I should go with an amplifier chip like the LM386 suggested by Ian.M.
 

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 12807
Re: Simple amplifier circuit, how to handle power and isolation?
« Reply #6 on: April 26, 2018, 02:07:12 am »
The big issue is when there's a ground reference for one of the circuits that you don't know about.    E.g. if L1 was actually an audio circuit of some sort, it would be quite common to have the input signal ground connected to the junction of B1 and B3 so the circuit effectively has positive and negative supply rails.   As long as whatever is plugged into the input is floating, no problem, but if you connect something that is truly grounded, and you've made extra ground connections as in your example above, you'll have shorted B3 through the input signal wiring shield, the signal source, and back through the external ground, and the odds are something expensive will let the magic smoke out. 

Also, you need to be wary of ground loops: Each wire or track in a circuit has some inductance and resistance, so depending on the voltage drops between the low side of each bulb and its associated battery(s), you may get current flow through the wire joining them, and back through the external ground.    Stray voltages on the external ground system and external EM fields coupled to the area of the ground loop inducing currents in it can get coupled into audio (and other) circuits that way and cause major problems.
 

Offline DukeOfEarlTopic starter

  • Newbie
  • Posts: 9
  • Country: au
Re: Simple amplifier circuit, how to handle power and isolation?
« Reply #7 on: April 26, 2018, 04:32:30 am »
I see what you mean, so I would be getting problems if the Accessory GND isn't actually the device GND and I managed to connect a circuit referenced to one with a circuit referenced to the other via non-isolated power supplies or similar.
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 19345
  • Country: gb
  • 0999
Re: Simple amplifier circuit, how to handle power and isolation?
« Reply #8 on: April 26, 2018, 11:19:30 am »
Why not use the LM386?
 

Offline DukeOfEarlTopic starter

  • Newbie
  • Posts: 9
  • Country: au
Re: Simple amplifier circuit, how to handle power and isolation?
« Reply #9 on: May 09, 2018, 08:22:10 am »
I came to the conclusion that an LM386 would be much better, yes :-)

So I'm back with an LM386 based design.

How's this?

Amplifier circuit:



The whole circuit is attached.
« Last Edit: May 09, 2018, 08:23:58 am by DukeOfEarl »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf