60mV to 10V voltage amplifier

A couple of points: It's not clear what your overall goal is here, but with an overall gain of

167 you are probably going to need to be concerned with offset voltage and drift in the op-amps.

I don't know what the final load is supposed to be, but I presume this is for some sort of measurement. If so, and offset and drift are problems, you could consider a "chopper" circuit: Extend the front fork... oops, wrong chopper! Seriously, the way to beat offset/drift in high gain circuits is to "chop" the input signal to make it into a square wave, then amplify it with AC gain stages and measure the peak-to-peak output. Lots of options here.

Note that this approach eliminates problems due to intrinsic op-amp offset and drift, but does nothing for gain drift due to resistor thermal effects, etc.

Regarding the power supply, you should consider using one op-amp as a "ground splitter". You should consider this whether you go with your original cicuit, or something fancier with a chopper. Instead of the circuit you show, use 2 equal-valued resistors (10k is OK) to divide the

24V as shown, then feed that to an op-amp buffer that drives the "ground" for the rest of the circuit. You probably don't need the 100uF caps on the divider, just a single smaller cap from the 12V midpoint to the true input ground... maybe 0.1uF, just to keep noise down.

Best regards,

Bob Masta DAQARTA v5.10 Data AcQuisition And Real-Time Analysis

Scope, Spectrum, Spectrogram, Sound Level Meter Frequency Counter, FREE Signal Generator Pitch Track, Pitch-to-MIDI DaqMusic - FREE MUSIC, Forever! (Some assembly required) Science (and fun!) with your sound card!

What is the frequency of your signal?

=A0 =A0 =A0 TL084

=A0|\ =A0 =A0 =A0 0-10V

=A0 =A0 =A0output

=A0 =A0| =A0>----o----o

=A0+--|-/ =A0 =A0 | =A0 =A0|

/ =A0 =A0 =A0| =A0 /+\

=A0 =A0| =A0( =A0 )

=A0 +--------- + =A0 \-/

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 |

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 |

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =3D=3D=3D

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0GND

-----o--------- +24V

=A0 |

=A0.-. =A0 =A0|+

| =A0 =3D=3D=3D

| =A0 /-\ 100uF

=A0'-' =A0 =A0|

=A0 | =A0 =A0 |

-o

=A0 | =A0 =A0 |

=A0.-. =A0 =A0|+

| =A0 =3D=3D=3D

| =A0 /-\ 100uF

=A0'-' =A0 =A0|

=A0 |

-----o---------0V

*Generally* it's a good idea to put all the gain in one stage. The only reason for not doing this is if you run into gain-bandwidth problems with one stage.

Shouldn't be an issue, though inputs and outputs shouldn't be routed together. Depending on the bandwidth of your signal, you should put a cap across the feedback resistors. It doesn't take much to kill any chance of oscillation.

No, that's a particularly bad choice for your circuit. The input offset voltage on that part is 20mV! That's 1/3 of your full scale signal. Not good. You should easily be able to find something with one or two millivolts of input offset.

It's going to be noisy, but it should work. You might add another opamp in a voltage follower configuration to buffer this point. Also, add a couple of .1uf caps (and maybe even lower values also) in parallel with the 100uF caps. Another step would be to use this additional opamp as an active filter for your bias network.

=A0 =A0 =A0 TL084

=A0|\ =A0 =A0 =A0 0-10V

=A0 =A0 =A0output

=A0 =A0| =A0>----o----o

=A0+--|-/ =A0 =A0 | =A0 =A0|

|/ =A0 =A0 =A0| =A0 /+\

=A0 =A0| =A0( =A0 )

=A0 +--------- + =A0 \-/

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 |

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 |

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =3D=3D=3D

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0GND

o-----o--------- +24V

=A0 |

.-. =A0 =A0|+

| =A0 =3D=3D=3D

=A0 /-\ 100uF

'-' =A0 =A0|

=A0| =A0 =A0 |

--o

=A0| =A0 =A0 |

.-. =A0 =A0|+

| =A0 =3D=3D=3D

=A0 /-\ 100uF

'-' =A0 =A0|

=A0 |

o-----o---------0V

Hi Bob,

Thanks for your reply and recommendations. The input of the circuit is a current sensor which outputs 60mV DC when the current is maximum. The output of the circuit goes to a 10V volt meter. Yes it is used for some sort of measurment.

I have googled on chopper circuit and came across this link

It says that a good low drift preamp like LM121 would be an alternative for a chopper solution as you suggested. But I cant seem to find the datasheet for LM121 even in my 1982 NS linear databook. I would like to know the gain of this device to see if I can make use of it.

Googling "ground splitter" finally get me to "Virtual ground circuits"

I never knew that there are so many ways to do it even with transistors. I picked the one with op-amp as you suggested.

Thanks and regards.

Allen

Why not use an instrumentation op amp configuration?

You can buy single packages that do it all for you.

=A0 =A0 =A0 TL084

=A0 =A0|\ =A0 =A0 =A0 0-10V

=A0 =A0 =A0output

=A0 =A0| =A0>----o----o

=A0+--|-/ =A0 =A0 | =A0 =A0|

=A0|/ =A0 =A0 =A0| =A0 /+\

=A0 =A0| =A0( =A0 )

=A0 =A0 +--------- + =A0 \-/

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 |

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 |

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =3D=3D=3D

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0GND

-o-----o--------- +24V

=A0 |

=A0.-. =A0 =A0|+

| | =A0 =3D=3D=3D

| | =A0 /-\ 100uF

=A0'-' =A0 =A0|

=A0 | =A0 =A0 |

---o

=A0 | =A0 =A0 |

=A0.-. =A0 =A0|+

| | =A0 =3D=3D=3D

| | =A0 /-\ 100uF

=A0'-' =A0 =A0|

=A0 |

-o-----o---------0V

The input signal is DC. So it's ok to amplify the signal by 167 in one stage as it is DC.

I'll keep that in mind and put a cap across each feedback resistor. Is 10nF ok for my application?

You're right about this and I have found a lot of opamps with low offset input voltage. I'll write them and do some shopping tomorrow. But most of them are in single and dual packages. Thanks for your good advice!

text -

I'll take care of this by putting a virtual ground circuit using one of the opamps and connecting a few .1uF for power decoupling.

Regards,

Allen

=A0 =A0 =A0 TL084

=A0|\ =A0 =A0 =A0 0-10V

=A0 =A0 =A0output

=A0 =A0| =A0>----o----o

=A0+--|-/ =A0 =A0 | =A0 =A0|

|/ =A0 =A0 =A0| =A0 /+\

=A0 =A0| =A0( =A0 )

=A0 +--------- + =A0 \-/

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 |

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 |

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =3D=3D=3D

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0GND

o-----o--------- +24V

=A0 |

.-. =A0 =A0|+

| =A0 =3D=3D=3D

=A0 /-\ 100uF

'-' =A0 =A0|

=A0| =A0 =A0 |

--
> > =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0=
 =A0| =A0 =A0 |
> > =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0=
 .-. =A0 =A0|+
> > =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0R2 |=
 | =A0 =3D=3D=3D
> > =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 10K | |=
 =A0 /-\ 100uF
> > =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0=
 '-' =A0 =A0|
> > =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0Vss op amp =A0 =A0| =A0 =
=A0 |
> > =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0----=
o-----o---------0V
>
> > (created by AACircuit v1.28.6 beta 04/19/05www.tech-chat.de)
>
> > The first op-amp amplifies the signal by 20 and the second op-amp
> > amplifies the output by 5-15 times. =A0The third stage is a buffer.
> > The op-amp used is a TI TL084. =A0I was wondering if it is a good
> > idea to just amplify the 60mV by 166 in one stage followed by the
> > buffer. =A0Would that cause any unstabilities? =A0In that case I may
> > use TL082 instead.
>
> > I also have other op-amps like mc1458, lm748, lm358, lm324 and
> > TL071. =A0Is TL084 a good choice in this particular circuit?
>
> > As for the power supply. =A0I have only +24V DC from a battery.
> > Would the value of R1 and R2 be OK for the circuit to work?
>
> > Thanks,
>
> > Allen
>
> Why not use an instrumentation op amp configuration?
>
> http://en.wikipedia.org/wiki/Instrumentation_amplifier
>
> You can buy single packages that do it all for you.- Hide quoted text -
>
> - Show quoted text -

Hi George,

Thanks for letting me know about this type of opamp.  I'll take a good
look at it and see if they are available locally.  Looks like MAX4194
is a good candidate.  Does Maxim give away free samples?

Allen

The LM121 has gone the way of the dodo.

Try this device -

LMC6084 which should get you close to what you need. It's in a PDIP or SOIC package so you shouldn't have any trouble soldering it.

It never changes? Well, there's no reason to monitor it then. ;-) Seriously, it matters how fast the signal is expected to change.

Is 1600Hz operation OK? I was thinking more along the lines of 1nF or even

100pF, but it varies by application.

Unless you chopper stabilize it, good idea. Add a chopper and the input offset isn't all that important.

-

Virtual ground? You want a "non-inverting unity-gain", "voltage-follower", or "buffer" configuration.

An instrumentation amp isn't going to buy anything in this application. The input offset will likely eat him alive.

Maxim doesn't even sell parts. They do dream up some fantastic specs, though. (That's a hint: stay *far* away from maxim. Don't even use their name, except on stage at the Comedy Club).

=A0 =A0 =A0 =A0 TL084

=A0 =A0|\ =A0 =A0 =A0 0-10V

|+\ =A0 =A0 =A0output

=A0 =A0 =A0| =A0>----o----o

=A0 =A0+--|-/ =A0 =A0 | =A0 =A0|

=A0|/ =A0 =A0 =A0| =A0 /+\

=A0 =A0 =A0| =A0( =A0 )

=A0 =A0 +--------- + =A0 \-/

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 |

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 |

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =3D=3D=3D

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0GND

----o-----o--------- +24V

=A0 =A0 |

=A0 =A0.-. =A0 =A0|+

R1 | | =A0 =3D=3D=3D

0k | | =A0 /-\ 100uF

=A0 =A0'-' =A0 =A0|

=A0 =A0 | =A0 =A0 |

o-----o

=A0 =A0 | =A0 =A0 |

=A0 =A0.-. =A0 =A0|+

R2 | | =A0 =3D=3D=3D

0K | | =A0 /-\ 100uF

=A0 =A0'-' =A0 =A0|

=A0 =A0 |

----o-----o---------0V

=A0

Oops, yes it does change of course. But may be in the realm of 1 to

10 cycles per second.

a

even

More likely 10 Hz or else the DC voltmeter wouldn't be able to get a steady reading!

Allen

th

t

o,

ed text -

er", or

Huh? instrumentation amps have very low dc offsets and are simply more ideal op amps. It can do no worse than any other basic op amp configuration.

eal

No, they really don't. Instrumentation amps are built using three conventional opamps, the first two in a non-inverting configuration (each of the two non-inverting inputs for each instrumentation amp input) followed by a differential amp. They're useful for high impedance differential sensing, but do not have a particularly low input offset voltage. Of course, IOS can be trimmed and low IOS opamps do exist (the AD8605/6/8 have an IOS of 20uV typical - 1000 times better than the opamp discussed above).

n.

ideal

Yup, Inst amps are great when you need a differential input with high input impedance. But it's wrong to think of them as 'better' opamps. They can't be used to make active filters, oscillators, rectifiers.... and all the other cool opamp circuits.

George H.