DC voltage follower (buffer) with single power supply

I want to convert a PWM signal (from my micro-controller) to a DC voltage (varying between 0 and 3.3V). A simple RC circuit does the trick. However, as I am not too sure what's the impedance of the next circuit (a motor controller), I would like to add a buffer. Using a LM324 with +/- 12V supply it works perfectly on my breadboard. Now, on my robot I don't have a negative voltage, I only have 3.3V, 5V and 12V (from an arduino). Although the 324 doc says it accepts single-supply, I understand that this is only for ac operation (using vcc/2 as a reference voltage). A simple voltage follower montage with 0/12V supply gives 0V at the output regardless of the input voltage...

- Is there any solution besides getting a negative voltage supply?
- If not, how can I get a negative voltage?

In the worst case I can get a AC-DC (12V) converter to create the -12V but since I only need a negative voltage only for that small problem, I'd prefer a more elegant solution, that would only rely on the 3.3V, 5V and 12V available sources.

bristou said:

Although the 324 doc says it accepts single-supply, I understand that this is only for ac operation (using vcc/2 as a reference voltage)

Click to expand...

The LM124 series consists of four independent, high gain,
internally frequency compensated operational amplifiers
which were designed specifically to operate from a single
power supply over a wide range of voltages. Operation from
split power supplies is also possible and the low power
supply current drain is independent of the magnitude of the
power supply voltage.

Click to expand...

what you describe (Vcc/2) is the bias you have to use in order to use an AC signal input while the opamp works with single rail supply.
It works fine with single supply and DC input.

Alex

“Single-voltage supplied” LM324 has BIG-BIG problems when working with output voltages close to 0V and close to Vcc ..
You have to select an Opamp that has rail-to-rail output ..

:wink:
IanP

IanP said:

“Single-voltage supplied” LM324 has BIG-BIG problems when working with output voltages close to 0V and close to Vcc ..
You have to select an Opamp that has rail-to-rail output ..

Click to expand...

But according to the datasheet

supply voltage Large output voltage swing 0V to V+ − 1.5V

Click to expand...

EDIT: the above is written in the features but in the actual electrical data it says 5mV-20mV as minimum output with 5v supply

Input Common-Mode Voltage Range (Note 10) 0 to (V+)−1.5

Note 10: The input common-mode voltage of either input signal voltage should not be allowed to go negative by more than 0.3V (at 25˚C). The upper end of the common-mode voltage range is V+ − 1.5V (at 25˚C), but either or both inputs can go to +32V without damage (+26V for LM2902), independent of the magnitude of V+.

Click to expand...

With a power supply of 5v or even better 12v it will be fine.

Alex

I'm also planning to use a lm324 to buffer a dc signal that goes from 0.6 v to 1.4 v.....I've tested it from 0v to 1.2 v so far and it seems to be working fine.
@IanP : Can you please tell us about the "BIG BIG" problems...it might just save me from a lot of problems!

The LM324 (358) will never produce 0V output ..
If you use 5V to power the LM324, the output voltage swing will be restricted to somewhere between a low of 0.3 volts and a high of 3.5 volts ..
The low of 0.3V will also be present when the supply voltage rises to 12V (and more) volts ..

So if your application can tolerate this restriction - you can use the LM324, otherwise, as I suggested before, go for a rail-to-rail (output) opamp ..

:wink:
IanP

Alright ! In my case,this shouldn't be a problem since the voltage never goes below 0.5v (or above 3v) but I guess the OP might be in trouble
Hey bristou, it should be possible to split the 12v you have into +-6v by a pot divider for the lm324...how about that?

IanP said:

The LM324 (358) will never produce 0V output ..
If you use 5V to power the LM324, the output voltage swing will be restricted to somewhere between a low of 0.3 volts and a high of 3.5 volts ..
The low of 0.3V will also be present when the supply voltage rises to 12V (and more) volts ..

So if your application can tolerate this restriction - you can use the LM324, otherwise, as I suggested before, go for a rail-to-rail (output) opamp ..

:wink:
IanP

Click to expand...

I don't understand, so you are saying that the datasheet is lying when it says the output minimum is 5-20mV (sorry not 0 as I said in the previous post), this is at least 15 time lower that your specification of 0.3v.
Is the specification of the datasheet wrong in both LM324 and LM358?

Alex

for one thing: what is the exact circuit that you will be driving with that signal. In most driver/amplifier/etc. chips, the signal input has relatively high impedance, so maybe it's possible to run the whole thing with no buffer. There has to be that parameter in the datasheet somewhere (look for input impedance or input resistance).

If it's not high impedance (let's say <500k) then use some precision rail to rail amplifier. You don't need high bandwidth, which makes things much simpler. I wouldn't use the LM324 in the first place, as it's not a precision component. I'd rather look for something like OP07 but r-t-r.

As for the negative voltage source it's also easy to build one. You would need a simple inverting charge pump. The most basic circuit of that kind comprises of 4 BJT's, 2 small ceramic caps, 4 resistors, 2 electrolytic caps and 2 diodes (and optionally a voltage regulator). You can also base that on 555 timer.
(LINK).

alexan_e said:

I don't understand, so you are saying that the datasheet is lying when it says the output minimum is 5-20mV (sorry not 0 as I said in the previous post), this is at least 15 time lower that your specification of 0.3v.Is the specification of the datasheet wrong in both LM324 and LM358?Alex

Click to expand...

I don't know why data sheets claim 5mV, physically it's impossible (see internal configuration of the output stage) ..
Models should take a pessimistic value of worst case, rather than showing values that are not achievable ..
See also:
LM324 Output Won't Zero

:wink:
IanP

I think you missed this in my original point: "A simple voltage follower montage with 0/12V supply gives 0V at the output regardless of the input voltage"

I can live with 0.3V instead of 0V btw...