The sound amplifier: Wow! it works and works well! (video)

The sound amplifier: Wow! it works and works well! (video)

This is a discussion on The sound amplifier: Wow! it works and works well! (video) within the Off Topic & Humor Discussion forums, part of the The Back Porch category; This is just a demo video of where I am with the sound amplifier power amp. In the demo it's powered by 2.00 volts and ...

Results 1 to 10 of 10
Like Tree19Likes
  • 6 Post By Tangle
  • 3 Post By ETXhiker
  • 4 Post By Tangle
  • 2 Post By ETXhiker
  • 1 Post By Osprey
  • 2 Post By Tangle
  • 1 Post By Tangle

Thread: The sound amplifier: Wow! it works and works well! (video)

  1. #1
    Lead Moderator
    Array Tangle's Avatar
    Join Date
    Dec 2004
    Location
    Chattanooga
    Posts
    12,979

    The sound amplifier: Wow! it works and works well! (video)

    This is just a demo video of where I am with the sound amplifier power amp. In the demo it's powered by 2.00 volts and it's driving two 16 ohm earbuds in parallel.

    It does show some crossover distortion which I will likely make some adjustments to the circuit to minimize crossover. It's a delicate thing to do though - you can get it just right at 3 and it's not right at 2.0 volts. Typically, to minimize crossover, you have to have some conduction through the output transistors without a signal present. But, that "idle" courrent can grow very large if the temperature changes so there's definitely a compromise between temperature stability and crossover minimization.

    There's more in the description part of the video, so you can read that if you like.



    I forgot to add the revised circuit, so here it is. In order to reduce power supply voltage variation sensitivity, I replaced several passive loads (resistors) with active loads (constant current sources):

    We don't have a gun problem in the US, We have a people problem.
    The problem we have is people that want to kill large numbers innocent people
    in Gun Free Zones.

  2. #2
    Distinguished Member Array ETXhiker's Avatar
    Join Date
    Jun 2013
    Location
    East Texas
    Posts
    1,483
    Excellent engineering, Tangle! I have a degree in electronics and I like to build guitar pedals, but the circuits are very easy compared to that one. I usually work with 9 volts and distortion is often the goal, rather than something to be avoided. Trying to build clean circuit with minimal power drain at 2 volts is a difficult task. Good job!

    (Edit to add: What CAD program did you use for the schematic?)
    "There are some ideas that are so wrong, only a very intelligent person could believe them." - George Orwell

  3. #3
    Lead Moderator
    Array Tangle's Avatar
    Join Date
    Dec 2004
    Location
    Chattanooga
    Posts
    12,979
    Quote Originally Posted by ETXhiker View Post
    Excellent engineering, Tangle! I have a degree in electronics and I like to build guitar pedals, but the circuits are very easy compared to that one. I usually work with 9 volts and distortion is often the goal, rather than something to be avoided. Trying to build clean circuit with minimal power drain at 2 volts is a difficult task. Good job!

    (Edit to add: What CAD program did you use for the schematic?)
    That's what my BS degree is too, Electrical/Electronics.

    Well, with regard to the guitar pedals, I'm not sure "...very easy..." is that accurate, maybe "less difficult"

    But yeah, this was tough! With the goal of making it as easy as possible/feasible on my brother in law, I think I stretched my brain to come up with something feasible. That low voltage requirement really makes it tough!

    The circuit was designed and developed on National Instruments MultiSim. It not only lets you draw the circuit, it simulates it pretty close - waveforms and all. Saved me a ton of time and effort breadboarding.

    I got this free gratis from National Instruments because I was teaching electronics technology at a community college. I had the foresight to install it on my laptop and when I retired, I got to keep it.

    If I can get this thing off my hands/mind, and I've still got a way to go, my next electronics project is going to be a circuit that allows one send speech, i.e. talk, over a laser beam. I've already done that twice but this will be an updated version. The power amp circuit for the sound amplifier will come in handy for that project.

    Oops - gotta run - it's range day! I try to hit the range on days that end in "y"
    We don't have a gun problem in the US, We have a people problem.
    The problem we have is people that want to kill large numbers innocent people
    in Gun Free Zones.

  4. #4
    Distinguished Member Array ETXhiker's Avatar
    Join Date
    Jun 2013
    Location
    East Texas
    Posts
    1,483
    Cool stuff, Tangle. Keep us informed on the laser project, sounds awesome. Have fun at the range!
    airslot and Tangle like this.
    "There are some ideas that are so wrong, only a very intelligent person could believe them." - George Orwell

  5. #5
    Member Array Osprey's Avatar
    Join Date
    Jan 2020
    Location
    Texas
    Posts
    230
    Nice progress. Pretty clean signal. the crossover isn't too bad. Are you going to pre-bias the base voltages to get rid of the crossover?

    I am not trying to mimic yours but I have been working on a similar one this week (after your thread inspired me). I measured it this morning with 3.7 volts to 2.2 volts (it fell off at 2.1 which I thought was amazing for the LM358) and .7 amps. Output current was .25 mA which was pretty darn close to Multisim when I ran the simulation (.221mA).

    I changed the feedback resistors from 1M to 560K Ohms and the hissing is about gone.

    I am ovedriving the circuit since my cheapo waveform generator sucks (15 dollar cheapo). I have a Siglent SDG1032X that I meant to use today but got busy.

    Again: great work!

    Chn 1 is output (yellow). Chn 2 is input (150 mv p=p) (blue)



    The sound amplifier: Wow! it works and works well! (video)-20200202_091527.jpg
    ETXhiker likes this.

  6. #6
    Lead Moderator
    Array Tangle's Avatar
    Join Date
    Dec 2004
    Location
    Chattanooga
    Posts
    12,979
    Quote Originally Posted by Osprey View Post
    Nice progress. Pretty clean signal. the crossover isn't too bad. Are you going to pre-bias the base voltages to get rid of the crossover?
    The bias is pretty close. Biasing transistors without temperature compensation is pretty tricky. Since this thing will have to operate in temps anywhere from around 30 to 90, you have to pick the worst case, 90 and bias it there and just take what you get at 30.

    A silicon bipolar transistor's Vbe changes about -2mV/C. So if the bias is set too close, when the temperature goes up, Vbe will decrease causing the transistor to conduct more. That may or may not be a problem, but it can lead to thermal runaway that bipolar transistors are known to be susceptible to. For example, suppose the current through a transistor causes its temperature rise slightly. That rise in temperature causes the transistor to conduct more, which heats it up more, which causes it to conduct more....well that's thermal run away.

    I think in my case, the worst that's gonna happen is the transistors just conduct more at higher temps and drain the battery a bit faster, but the battery has plenty of capacity, so I think it won't be a problem.

    I'm about to re-run the temperature sweep and see how it behaves. I ran it once before I had the lower constant current source in the circuit. I'm hoping that because all the constant currents are controlled by one single Vbe, as the temperature changes, that one Vbe will change in harmony with the output transistors and provide some temperature compensation.

    Quote Originally Posted by Osprey View Post
    ...I am not trying to mimic yours but I have been working on a similar one this week (after your thread inspired me). I measured it this morning with 3.7 volts to 2.2 volts (it fell off at 2.1 which I thought was amazing for the LM358) and .7 amps. Output current was .25 mA which was pretty darn close to Multisim when I ran the simulation (.221mA).
    Yeah, the LM358 op amp uses an emitter follower type output which is likely driven by a constant current source as a part of the IC. The characteristic of this type of output circuit is the output is only spec'd to Vplus - 1.5 volts. So if you're using a 3.7 volt supply the maximum output can only be 3.7 - 1.5 or 2.2 volts on the positive output swing.

    Quote Originally Posted by Osprey View Post
    ...I changed the feedback resistors from 1M to 560K Ohms and the hissing is about gone.
    That makes sense, it halved the gain hence lower output including lower noise. But, you shouldn't be hearing/seeing hiss if the circuit is working properly. While larger value resistors do produce more thermal noise (thermal noise is noise produced by an object being above absolute zero) than lower values, but we're talking microvolts of noise, you should be hearing hissing - well, unless something is overheating

    Quote Originally Posted by Osprey View Post
    ...I am overdriving the circuit since my cheapo waveform generator sucks (15 dollar cheapo). I have a Siglent SDG1032X that I meant to use today but got busy.
    You can always use a two resistor voltage divider to cut the signal down so it won't overdrive the op amp.
    Osprey and ETXhiker like this.
    We don't have a gun problem in the US, We have a people problem.
    The problem we have is people that want to kill large numbers innocent people
    in Gun Free Zones.

  7. #7
    Lead Moderator
    Array Tangle's Avatar
    Join Date
    Dec 2004
    Location
    Chattanooga
    Posts
    12,979
    Quote Originally Posted by Tangle
    Quote Originally Posted by Osprey View Post
    ...I am overdriving the circuit since my cheapo waveform generator sucks (15 dollar cheapo). I have a Siglent SDG1032X that I meant to use today but got busy.
    You can always use a two resistor voltage divider to cut the signal down so it won't overdrive the op amp.
    Shoot @Osprey I forgot to include this. If you overdrive the op amp, it goes out of its controlled linear mode and into a switching mode which introduces all kines of distortion.
    Osprey likes this.
    We don't have a gun problem in the US, We have a people problem.
    The problem we have is people that want to kill large numbers innocent people
    in Gun Free Zones.

  8. #8
    Member Array Osprey's Avatar
    Join Date
    Jan 2020
    Location
    Texas
    Posts
    230
    Quote Originally Posted by Tangle View Post
    Shoot @Osprey I forgot to include this. If you overdrive the op amp, it goes out of its controlled linear mode and into a switching mode which introduces all kines of distortion.
    Heck, I know that LOL! I am going to use my Siglent tomorrow so I can get the input down and not overdrive the op amps.

  9. #9
    Lead Moderator
    Array Tangle's Avatar
    Join Date
    Dec 2004
    Location
    Chattanooga
    Posts
    12,979
    Quote Originally Posted by Osprey View Post
    Heck, I know that LOL! I am going to use my Siglent tomorrow so I can get the input down and not overdrive the op amps.
    I guess I'm not quite following your application. Are you trying to amplify an analog signal accurately or just make the output of an op amp switch from high to low? The o'scope pic you posted and from what you've said, i.e., you are simply "...overdriving the op amps...", indicates you making it switch rather than amplify.

    The o'scope pic shows severe distortion due to switching rather than amplifying, so I'm unclear as to what the purpose of your circuit is???
    We don't have a gun problem in the US, We have a people problem.
    The problem we have is people that want to kill large numbers innocent people
    in Gun Free Zones.

  10. #10
    Member Array Osprey's Avatar
    Join Date
    Jan 2020
    Location
    Texas
    Posts
    230
    Quote Originally Posted by Tangle View Post
    I guess I'm not quite following your application. Are you trying to amplify an analog signal accurately or just make the output of an op amp switch from high to low? The o'scope pic you posted and from what you've said, i.e., you are simply "...overdriving the op amps...", indicates you making it switch rather than amplify.

    The o'scope pic shows severe distortion due to switching rather than amplifying, so I'm unclear as to what the purpose of your circuit is???
    It should amplify. It is switching because it is overdriven which is showing the distortion. It was working OK when I used 1Khz signal from a speaker. When I use a sig gen, the signal level is way too high.

    Hope that makes sense. I need to put in about 10 mv p-p to really check it. My Siglent 1032 lets me do the levels I want. I was just too busy to use it :).

Sponsored Links

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •