simple op-amp woes

[Chet Hosey]

I'm trying to construct a virtual earth mixer using an op-amp. In
order to test my understanding of the way op-amps work, I threw
together a quick test circuit on a breadboard. I'm having trouble
getting the circuit response I expect, so hopefully somebody here can
point out the source of my troubles.

It's been a long while since I've dabbled in things electronic, so
I'll try to describe my circuit and expectations as briefly yet
completely as possible.

I'm using a 120 VAC-to-12.6VAC stepdown transformer through a bridge
rectifier. I'm using a 100uF electrolytic capacitor to smooth the
ripples, and using it to feed a TL082 op-amp. I've measured the
difference between positive and negative supply voltages at about
18.5V. Using two 220K resistors to obtain ~9.25V, I set the
non-inverting input to 9.25V and jumpered the non-inverting input to
the positive supply.

As the non-inverting input was at a lower voltage than the inverting
input, I expected the output to swing down to 0V. However, the
measured difference between the op-amp's output and ground was about
18.1V. Thinking that I'd been mistaken in my expectations, I reversed
the jumpers connecting the two inputs and again measured 18.1V between
the op-amp's output and ground.

Without a feedback resistor, shouldn't the output be near the negative
supply voltage when the inverting input is at a lower voltage than the
non-inverting input? I'm doing something wrong here, but I'm at a loss
as to what's wrong.

I've posted a schematic at
http://www.nauticom.net/www/chosey/opamp.gif. My eventual goal is to
make an audio mixer as a gift for a friend, so of course I would be
most appreciative of any input anyone cares to offer.

 

[John Popelish]

I'm trying to construct a virtual earth mixer using an op-amp. In
order to test my understanding of the way op-amps work, I threw
together a quick test circuit on a breadboard. I'm having trouble
getting the circuit response I expect, so hopefully somebody here can
point out the source of my troubles.

It's been a long while since I've dabbled in things electronic, so
I'll try to describe my circuit and expectations as briefly yet
completely as possible.

I'm using a 120 VAC-to-12.6VAC stepdown transformer through a bridge
rectifier. I'm using a 100uF electrolytic capacitor to smooth the
ripples, and using it to feed a TL082 op-amp. I've measured the
difference between positive and negative supply voltages at about
18.5V. Using two 220K resistors to obtain ~9.25V, I set the
non-inverting input to 9.25V and jumpered the non-inverting input to
the positive supply.

You have confused yourself. You have the noninverting input connected
to two things. It may be simpler to refer to the inputs as + and -.

As the non-inverting input was at a lower voltage than the inverting
input, I expected the output to swing down to 0V.

Agreed, if the opamp were ideal.

However, the
measured difference between the op-amp's output and ground was about
18.1V. Thinking that I'd been mistaken in my expectations, I reversed
the jumpers connecting the two inputs and again measured 18.1V between
the op-amp's output and ground.

So reversing the inputs does not change the output voltage?

Without a feedback resistor, shouldn't the output be near the negative
supply voltage when the inverting input is at a lower voltage than the
non-inverting input? I'm doing something wrong here, but I'm at a loss
as to what's wrong.

Do you have the data sheet handy to make sure of the pinout?
http://focus.ti.com/lit/ds/symlink/tl082a.pdf

I've posted a schematic at
http://www.nauticom.net/www/chosey/opamp.gif. My eventual goal is to
make an audio mixer as a gift for a friend, so of course I would be
most appreciative of any input anyone cares to offer.

Take a look at page 8 of the data sheet for the specification of the
common mode input voltage range. Many opamps act weird if either
input voltage is outside this range. This design needs to have both
inputs a few volts from the negative supply rail for everything to
work as expected. It is not obvious to me that taking either output
to the negative rail would lock the output negative, but to make sure
this is not happening, change your input divider to 3 equal resistors
in series, and try connecting the inputs to the two 1/3 supply points
alternately, to see if your unit acts more reasonably.

 

[Jamie]

when ever the voltage on the - input over comes the + voltage you get
negitive effects, the results should be aprox 0.8 volts.
just think of the + input giving you the expected positive effects.
when voltage on the + input exceeds the voltage on the - input it
the output goes high..
do that the same for the - input and it simple does the oposite..
unless you have something not correctly wired?
are you using a single supply or a +/- common ?

 

[Skeptic]

Your description does not match your schematic. According to your
schematic the non-inverting input (+) is connected to the negative
supply or 0V. The inverting input (-) is connected to the junction of
R1 & R2 or 9.25V. Since the non-inverting input is higher than the
inverting input, the output should be high. This is what you observed.

If you're looking for a mixer circuit, you might try this one.

http://www.aaroncake.net/circuits/mixer2.htm

 

[Chet Hosey]

You have confused yourself. You have the noninverting input connected
to two things. It may be simpler to refer to the inputs as + and -.

Aye, I have. One of those should have been a reference to the
inverting input. I'll stick to + and -.

So reversing the inputs does not change the output voltage?

No, it wasn't affecting the output voltage -- either orientation would
result in the output running very near the positive supply voltage.

Do you have the data sheet handy to make sure of the pinout?
http://focus.ti.com/lit/ds/symlink/tl082a.pdf

Take a look at page 8 of the data sheet for the specification of the
common mode input voltage range. Many opamps act weird if either
input voltage is outside this range. This design needs to have both
inputs a few volts from the negative supply rail for everything to
work as expected. It is not obvious to me that taking either output
to the negative rail would lock the output negative, but to make sure
this is not happening, change your input divider to 3 equal resistors
in series, and try connecting the inputs to the two 1/3 supply points
alternately, to see if your unit acts more reasonably.

I've verified the pinout both against the data sheet (thanks for the
link, by the way) and against the packaging in which it came. I had
much better luck with your suggestion of taking inputs which weren't
so near the supply voltages -- with the + input at halfway between the
positive and negative supply voltages I was getting the results I
expected (ie, a virtual ground amplifier).

This leads me to what was probably the cause of my misunderstanding --
I've been treating the negative supply rail as "ground", and jumpered
the + input straight to the adjacent negative supply pin (pins 3 and 4
on the TL082).

It was a big forehead slapper, and reminds me of the time back in 1993
when I wondered why op-amp schematics always showed +VDC and -VDC
instead of just calling it +VDC and ground. Unfortunately I'd moved on
to less costly hobbies before learning the answer.

I really appreciate your response. I'm no longer thinking that trying
to build a homebrew audio mixer was an ill-conceived gift idea, and
it's really nice to rediscover an old interest.

 

[Chet Hosey]

when ever the voltage on the - input over comes the + voltage you get
negitive effects, the results should be aprox 0.8 volts.
just think of the + input giving you the expected positive effects.
when voltage on the + input exceeds the voltage on the - input it
the output goes high..
do that the same for the - input and it simple does the oposite..
unless you have something not correctly wired?
are you using a single supply or a +/- common ?

I think I discovered the source of my misunderstanding -- I'd jumpered
the + input to the negative supply rail, when it seems that it should
be kept halfway between the positive and negative supply voltages (0V,
not -VDC).

Thanks for the response!

 

[John Popelish]

Chet Hosey wrote:
(snip)

I really appreciate your response. I'm no longer thinking that trying
to build a homebrew audio mixer was an ill-conceived gift idea, and
it's really nice to rediscover an old interest.

You are welcome. Opamps are fun and easy to use compared to almost
any other amplifier mechanism, once you get your mind wrapped around
their limitations.

Good luck on your project.

 

[Outrider 141]

How many inputs do you want to use, how many outputs (mono or stereo)?
There are schematics on the web for multi input mixers, Craig Anderton
has an 8 in 1 out mixer that I use quite often. Cheap and easy to
build, and can be modified for high or low level inputs. It was from
the Electronic Projects for Musicians book, and I think PC boards are
available from PAIA.com - could be wrong on that one, tho.

I'm trying to construct a virtual earth mixer using an op-amp. In
order to test my understanding of the way op-amps work, I threw
together a quick test circuit on a breadboard. I'm having trouble
getting the circuit response I expect, so hopefully somebody here can
point out the source of my troubles.

It's been a long while since I've dabbled in things electronic, so
I'll try to describe my circuit and expectations as briefly yet
completely as possible.

I'm using a 120 VAC-to-12.6VAC stepdown transformer through a bridge
rectifier. I'm using a 100uF electrolytic capacitor to smooth the
ripples, and using it to feed a TL082 op-amp. I've measured the
difference between positive and negative supply voltages at about
18.5V. Using two 220K resistors to obtain ~9.25V, I set the
non-inverting input to 9.25V and jumpered the non-inverting input to
the positive supply.

As the non-inverting input was at a lower voltage than the inverting
input, I expected the output to swing down to 0V. However, the
measured difference between the op-amp's output and ground was about
18.1V. Thinking that I'd been mistaken in my expectations, I reversed
the jumpers connecting the two inputs and again measured 18.1V between
the op-amp's output and ground.

Without a feedback resistor, shouldn't the output be near the negative
supply voltage when the inverting input is at a lower voltage than the
non-inverting input? I'm doing something wrong here, but I'm at a loss
as to what's wrong.

I've posted a schematic at
http://www.nauticom.net/www/chosey/opamp.gif. My eventual goal is to
make an audio mixer as a gift for a friend, so of course I would be
most appreciative of any input anyone cares to offer.

Use the usual techniques to reply via email.

Molon Labe!

 

[Chet Hosey]

My description was a tad off, although the confusing part was that
reversing the inputs had no effect on the output voltage.

As I discovered last night after reading an earlier response to my
post, attaching the + input to the negative supply rail was not what
I'd actually wanted to do. I appreciate your response, and had I seen
a schematic such as the one to which you'd posted a link when
prototyping my design I'd have noticed that what I actually wanted to
do was to split the difference between the positive and negative
supply voltages and feed it to the + input.

Thanks for your response! The circuit you pointed me towards is
actually pretty close to what I'd had in mind, and will serve as a
handy reference.

-chet