[Solved] Non-inverting Op-Amp configuration not working

[wobbly]

I'm scratching my head about this 741 op-amp circuit I'm trying to build on a solderless breadboard.  (I imagine this is already raising red flags in people's minds!)


The final section does not work: the op-amp that I hoped to set up as a non-inverting 10x amplifier.  Instead it outputs nothing at all, just GND (0V) all the time.  Does not respond to the input at all.

The first op-amp does indeed split the 16V bench-top PCU into +8 and -8 volts and is stable.

The microphone section also works.  I can see very clear audio signals on my scope at 200mV/div.

I tried removing the RC filter, but that didn't help.
I tried changing the op-amp to a fresh, unused one (also an ST branded ua741), this didn't help either.
Checked every single connection in the circuit and even tore the whole thing up and rebuilt it on a brand new breadboard - no change.

Measured all the resistors and caps carefully, they are all correct.

One odd thing that I notice is that the circuit only draws 1.5 mA.  That seems a bit low to me, given that the datasheet for the ua741 suggests that the typical is 1.7 mA for a single device - and I'm using two in parallel!  But it doesn't specify a minimum, so maybe that's right? 

I'm starting to lose my marbles with this .  Every book I've consulted basically gives this exact circuit for a 10x non-inverting amplifier, and generally they cite 741s as a suitable part for such a project.  What's wrong with this circuit?

[Manul]

Well, what's wrong... Non inverting input has no DC path, that will not work, cause you must give bias current a path, otherwise the input will drift into the woods. I'm not saying that's your problem, cause in that case output would not be at zero, but nonetheless you must fix that. Driving capacitive load with 741 is not good idea. You say it's stable, but I would not use such design. Again it's likely not your problem.

Have you tried to measure how much supply current your second op amp takes? By removing and inserting it? Most likely some blind spot of using breadboard, so check all connectivity by multimeter. If still does not work, post voltages of every pin of op amp: 1,2,3,4...

[Phil1977]

From my point of view you should pull out C4. It shorts out the AC coming from your mic - and that's the signal you want to amplify.

If this doesn't help, I´d start tracing the signal. The microphone should be sensitive enough that you see some signal if you connect the scope to its output and blow on it. If you have the signal there you can check it on the non-inverting input of the opamp. If it´s there but the output stays on 0V<->GND then the opamp probably is broken or not well supplied with power.

In general I would use the "transimpedance amplifier" circuit for a microphone capsule. It´s inverting, but in general it´s more stable if you amplify the current that´s coming from the microphone through the capacitors compared to building a "voltage to voltage" amplifier.

EDIT: You should replace C4 by some high resistor value (1MOhm or similar). As Manul has written, else the DC level there is undefined and can be anywhere.

[Xena E]

^What they said^ 

Then suspect wiring mistakes and shite breadboard connections in equal measure.

[wobbly]

Driving capacitive load with 741 is not good idea. You say it's stable, but I would not use such design. Again it's likely not your problem.

It is a bit noisy to be honest, but this isn't a serious design at all I'm just learning how these things work.
I have removed C4 (thanks @Phil) and also added a pair of 67K resistors between the + input and the +/-8 V rails to provide a DC path.

It does appear to be working now.  Pic shows some music playing into the mic and looks like I'm getting 10x gain (note 50mv/div on yellow and 500mv/div on green).



I really want to enjoy op-amps as much as everyone else seems to, but it's just so ****ing hard to understand

Thanks guys. 

[Terry Bites]

Its a bad idea to use a DC coupled amplifier to get AC signal gain.
Adding C4 lets you set the DC gain to unity and to whatever value you want for the AC signals.
It stops you multiplying the offset error by Av too.
The RC time constant R2,C4 sets the low frequency cut off and Rin,Cin the high pass frequency.
Use 18k/2k for 9:1 its just easier.

Opamp input terminals always need a dc path to some fixed reference point in the circuit. Else, as Manul says, the operating point is not defined and could wander anywhere.
Its a good idea to use simulators for sanity checking. eg www.qorvo.com/design-hub/design-tools/interactive/qspice its free and free from oldLTpsice.

[Xena E]

I really want to enjoy op-amps as much as everyone else seems to, but it's just so ****ing hard to understand

Thanks guys. 

A good book to cover the basics is E A Parr 'How To Use OP Amps'.

Non mathematical introduction, and several example circuits...

Free PDF download at World Radio History.

https://www.google.com/url?sa=t&source=web&rct=j&opi=89978449&url=https://www.worldradiohistory.com/UK/Bernards-And-Babani/Babani/BP-88-How-to-Use-Op-Amps-1982.pdf&ved=2ahUKEwiK9qzQr8WIAxVZSUEAHby1DTkQFnoECBAQAQ&usg=AOvVaw0-BknRUI4coRd6ggakllMh
 

[wobbly]

Its a bad idea to use a DC coupled amplifier to get AC signal gain.

This kind of experienced insight is entirely absent from most op-amp tutorials.  Books, online, whatever.  All they say is "Op-Amps are great, look at this...!" and then they give you a minimal feasible circuit and I'm forced to provide my own brick wall to bang my head against

[Phil1977]

Its a bad idea to use a DC coupled amplifier to get AC signal gain.

This kind of experienced insight is what is entirely absent from most op-amp tutorials.  Books, online, whatever.  All they say is "Op-Amps are great, look at this...!" and then they give you a minimal feasible circuit and I'm forced to provide my own brick wall to bang my head against

But that´s just the type of learning that everybody needs in the beginning. You can read as many books or do as many courses as you want, but the real understanding happens when you use the knowledge to solve problems. And one really great aspect of playing with opamps is that their basic principles are the same for thousands of applications. Understanding opamps makes understanding of all amplification circuits so much easier.

[wobbly]

Actually, this makes me wonder.  How much of my frustration is caused by the 741 op-amp in particular?  Despite being a very popular part and very old, everyone seems to have a collective loathing of it.  H&H recommend the LF411 in TAoE.  Will I find that chip easier or more forgiving?  I doubt it, else the 741 wouldn't be so popular.  I will not be beaten; I will learn to love op-amps.  But jeese it's hard work.

[Phil1977]

As long as you have bipolar voltage supply the 741 is not a bad choice to play with. It´s slow enough for breadboard assemblies and it´s "non-rail-to-rail" feature makes you care for some margin between its operating point and the supply voltage - which always is good, even if you switch to rail-to-rail chips some day.

I personally did a lot of projects with the LM324. It´s also slow enough for dirty wiring, and the presence of 4 amps in one chips makes you use them in a generous way. If you need a flipflop? Take an opamp. Need a comparator? Take an opamp. Need an oscillator? Take an opamp. Need some timer? Take an opamp.

Experts will teach you all these functions can and should be done by something better suited. But if you have the chip available it´s very tempting to use it. And it often just works.

[ledtester]

Actually, this makes me wonder.  How much of my frustration is caused by the 741 op-amp in particular?  Despite being a very popular part and very old, everyone seems to have a collective loathing of it.  H&H recommend the LF411 in TAoE.  Will I find that chip easier or more forgiving?  I doubt it, else the 741 wouldn't be so popular.  I will not be beaten; I will learn to love op-amps.  But jeese it's hard work.

The 741 requires a large supply voltage (like 9V between V+ and V-) in order to be useful. On the other hand, an electret microphone can work down to 3V (some even lower). Since you are interfacing with a microcontroller it would make sense to use an op-amp that can work with a lower supply voltage.

This shows an electret amplifier using a single-sided +5V supply and a LM358:

https://www.ee-diary.com/2021/11/how-to-make-electret-microphone.html

I'm sure you could adapt this design by powering the electret with +5V and the rest of circuit with your +18V battery. (Most electrets have a max supply voltage of around 10V.)