[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/diy/ - Do It Yourself

View post   
File: 89 KB, 720x1280, 1430448262767.png [View same] [iqdb] [saucenao] [google]
806452 No.806452 [Reply] [Original]

So I come from /ohm/, and I've decided to make a thread about this issue because I don't want to bloat the thread, since this is making my head hurt and I can't stop bumping into new problems.

If you're new to OP amps like me, you can lurk to learn more about them.

The thing is, I have two OP amps, and I want to connect them in series so they share the total gain. The first problem I encountered is the thing must run only with positive voltage. I've found a way to avoid cutting the waveform by rising the reference voltage, so the signal can move up and down in the positive region without reaching saturation (0 and 9 volts), and by using a capacitor at the output I can keep the constant part of the current at the output equal to 0, so any earphones connected won't fry.

The purpose of this is to build a "bionic ear" with hundreds of times the sensitivity of the human hearing.

Now I'm currently stuck with this problem, I have set the two OP amps to act inverting amplifiers, but only the second one is doing that, what is going on here?
The reference voltage has been set to 4.5 for both non inverting inputs to bias the output signal as I said before.

>> No.806468
File: 95 KB, 720x1280, Screenshot_2015-05-01-05-44-28.png [View same] [iqdb] [saucenao] [google]
806468

The problem got solved after I decreased the resistance of the resistors in the voltage divider, the small current flowing through them was causing the voltage to vary due the function generator (simulating a 10mV microphone) drawing too much current.

I still think there are thinks that could be improved, the divider seems a bit crappy and I'm not sure if there's a better way to set a reference voltage, like using a zener.
I also would like to know how good this circuit is in terms of noise, how could it be reduced...

>> No.806471
File: 33 KB, 772x260, 2015-05-01-140348_772x260_scrot.png [View same] [iqdb] [saucenao] [google]
806471

>>806452
the first one is doing as what you'd expect.

look at the picture and look at your schematic.

besides that there's a few other points about your schematic.

but really I might be able to sum it up by just saying put the positive connection to ground rather than mid point. that might be enough

op amps are sorta "tricky" to first timers but really all they do is compare "whats the voltage on +/-, which one is bigger? I'll go to that supply"

so if you have 3v on + and 2v on -, it'll shoot to the positive rail. but the feedback changes the amount of voltage on the terminal and so it re-adjusts itself.

>> No.806473
File: 7 KB, 1024x768, shit.png [View same] [iqdb] [saucenao] [google]
806473

also pic related.

your circuit is a bit "difficult" ( not, but harder than it has to be) to actually figure out.

>> No.806483
File: 80 KB, 720x1280, Screenshot_2015-05-01-06-25-14.png [View same] [iqdb] [saucenao] [google]
806483

>>806473
I see, the voltage variation at the divider was causing it to behave like a non inverting.
I tried to simulate that circuit without good results, I'll keep trying tomorrow, now I need to get some sleep. Setting a capacitor in series with the generator stops current from going though, but it still doesn't works properly.

Thanks for the help

>> No.806709
File: 70 KB, 720x1280, Screenshot_2015-05-01-19-22-56.png [View same] [iqdb] [saucenao] [google]
806709

>>806473
>that pic
I just realized that's probably wrong. Those are inverting amplifiers, if you connect the divider output at the inverting input the OP amp will try to bring it to 0, since the reference voltage is ground.
The simulation did just that the reference must be at the non inverting input.
I tried >>806468 IRL and it worked, but I'm not sure if the gain was less than what I expected. I could just hear the electret microphone, I tried with a piezo and it did better. The gain was 100 100 for both. Maybe it's because the output current isn't big enough, I'll add a biased transistor to compensate this.