LM358 voltage converter

Hi there I need an advice please! I'm doing a very simple voltage
converter ie, I have an output signal 0-20v from a external detector,
and I want to analize that signal in an ADC, but the input of the ADC
is 0-2.5v. I did an voltage divider and then I puted an LM358 like
this

0-20V------
|
__ +---------+
|| | +15v |
56k || | |-\| |
-- +-|- \ |
| | )---+--- Output (0 - 2.5V)
+----|+ /
-- |_/| LM358
|| |
8k || |
-- gnd
|
gnd

This simple circuit is powered by a tranformer wich has 1 output 0V
and +15V.
I've tested and I cannot get the desired output. I've tested with a
breadboard, then the 0-20V input I simulated with a function generator
(sine wave), I powered the AMPOP with vcc=+15v and vee=0v. When I
watch the output on the osciloscope I don't get get sine wave reduce
in amplitude, what is wrong????
Please help me I need this to a project, and when I studied
electronics class a looong time ago, I only did with and 741

PS: I've cheked the pads of the 358 and its all ok

Thanks I advanced

 

Subject: LM358 voltage converter

Hi, Ricardo. The LM358 is a good choice here, if you don't mind a couple of
millivolts offset, and not being able to get within more than a couple of
millivolts of GND. Given that...

1) Check your pinouts:

LM358 Pinout
1. Out 1
2. Inverting Input 1 (-)
3. Non-invering input 1 (+)
4. GND
5. Non-inverting input 2 (+)
6. INverting input 2 (-)
7. Out 2
8. Vcc

2) Recheck your wiring

3) Make sure you have the positive side of the input signal connected to the
voltage divider, and the negative connected to GND

4) Make sure there are no "sneak paths" for current to GND or other voltage
problems. Look at the

5) Check your wiring

If none of these work, you may have a dud LM358. You're on the right track,
and once you work out the problem, you should be OK. (By the way, set up the
other LM358 as a voltage follower off of the output signal of the first one, if
you're not using it for anything else. By the way, if you're using a PC ADC
card, your signal probably has to be GND referenced. That may also be a source
of problems.)

Good luck
Chris

 

Subject: LM358 voltage converter

Hi, Ricardo. The LM358 is a good choice here, if you don't mind a couple
ofymillivolts offset, and not being able to get within more than a couple
ofymillivolts of GND. Given that...

1) Check your pinouts:

LM358 Pinout
1. Out 1
2. Inverting Input 1 (-)
3. Non-invering input 1 (+)
4. GND
5. Non-inverting input 2 (+)
6. INverting input 2 (-)
7. Out 2
8. Vcc

2) Recheck your wiring

3) Make sure you have the positive side of the input signal connected to the
voltage divider, and the negative connected to GND

4) Make sure there are no "sneak paths" for current to GND or other voltage
problems.

5) Check your wiring

If none of these work, you may have a dud LM358. You're on the right track,
and once you work out the problem, you should be OK. (By the way, set up the
other LM358 as a voltage follower off of the output signal of the first one, if
you're not using it for anything else. By the way, if you're using a PC ADC
card, your signal probably has to be GND referenced. That may also be a source
of problems. See if your circuit works if it's not connected to the ADC.)

Good luck
Chris

 

Greetings.

Although the LM358 claims to be able to swing the output all the way to
ground level, this is not the whole picture. For output voltages less than
around one diode drop the output cannot sink much current at all (look at
the graph titled "Output Characteristics Current Sinking" in the datasheet).
IIRC the device is limited to sinking about the same amount of current as a
roughly 10k ohm pull down resistor. This is obviously pitiful when the
voltage gets too low.

On method of substantially improving its current sinking capability at low
output voltage would be to place a low value pull down resistor directly on
the output pin to ground. Something like 470 ohms might be a good choice in
your application.

You say you are inputting 0-20V sine wave into the circuit with your
function generator. Did make sure to include a large DC offset on your sine
wave so that the input voltage (to the resistive divider) only swings from
0-20V? Ordinary sine waves of course swing both positive and negative, and
the op-amp will not to be able to properly handle the negative voltage
excursion. This could certainly cause unanticipated op-amp output.

Also, what frequency is your input signal. Hopefully your signal is well
under one megahertz to avoid problems with the op-amps's gain-bandwidth
product and maximum slew rate capability.

What kind of load do you have hooked up to the output? Op-amps in general
don't usually like having significantly capacitive loads hooked directly to
their outputs. In these cases a simple resistor in series with the
capacitor will often help. The resistor doesn't usually need to be too
large, something like 100 ohms is probably adequate in many instances. I
mention this since some ADCs have significantly capacitive inputs. So if
the load is more than say a few picofarads try adding that 100 ohm or so
output resistor in series.

Otherwise your basic circuit concept looks fine and doing what you propose
should work.

 

Ricardo,

Are you talking about AC signals or DC voltages? You are running your
op-amp with a single polarity supply and your function generator test
producing the sine will pull the + input of the op amp below its
negative supply which is at ground. Please clarify so that we can help
you.

Fred.

 

test the voltage on the + input.
it should be what your looking for at the output.

 

Hi There and many thanks to all of you that gave me a little of your
time, yes the circuit is allright now, in fact the problem was with
the function generator it did not supply any dc offset so the voltage
was from -20 to +20 suppid me, and as I'm working with low freq 0 to
10 Hz I 've read the data sheet, I think I'll not have any further
problem. Many thanks again.

By the way this is a common problem in Portugal, we all learn
theorical electronics and stuff but when we get to the real world,
these kind of things happen. Thanks again

R. Fernandes

 

By the way this is a common problem in Portugal, we all learn

I don't think it has anything to do with Portugal, it is a worldwide
phenomenon (not really a problem) - we all need a period of
"apprenticeship
training" after doing the theoretical stuff at university to help the
ideas sink in.

Fred.

 

Hi Ricardo,

I just wonder... Why use anything more than the potential divider ?
K.I.S.S.