Help with amplifying a 0-5V analog signal

[eichler]

Hi.

I'm using a NI USB-6009 data acquisition card to send a signal to a proportional pressure controller which varies the supply pressure to a pneumatic system. Unfortunately, the pressure controller requires a 0-10V analogue control signal and the 6009 can only do 0-5V (very low current).

So I thought I would build a little circuit with a gain of 2 to amplify the signal. The 741 op-amp is powered by a 24V DC supply. 24V is a little on the high side for the op-amp but I have this supply already built in and (using a lab bench supply) the circuit works exactly the same at 12 and 18V. See links below (or attachments for pics)

**broken link removed**

**broken link removed**

Bascially, with input voltages from 5 down to 2 volts, the circuit behaves perfectly. The output is just the input multiplied by two.

However, the output never goes below 3.9 (ish) volts, no matter what the input voltage is.

Can anyone suggest a solution (or know where else to ask)? I really need to be able to control my output within the range of 1-10V. Is this a featurewith the op-amp? If so, is there a more suitable one out there that can handle low input voltages?

Many thanks. J

 

[crutschow]

As has been often noted in this forum, the old 741 input can not go to zero in or out with a single supply. For that to occur, the common-mode input range and output swing must go the negative supply voltage, which is this case is 0V, and the 741 can't.


You need a rail-to-rail or single-supply type op amp such as one of these:

• MC34074 (OnSemi) - V=3/44, GBW=4.5MHz, Bipolar (no R-R), 0.71€
• MC33174 (OnSemi) - V=3/44, GBW=1.8MHz, Bipolar (no R-R), 0.70€
• MC33204 (OnSemi) - V=2/12, GBW=2.2MHz, Bipolar, 0.96€
• TLV2374 (TI) - V=3/16, GBW=3MHz, CMOS, 0.81€
• TLC274 (TI) - V=3/16, GBW=1.7MHz, CMOS, 0.46€
• TLV2464 (TI) - V=2.7/6, GBW=6.4MHz, CMOS, 1.48€
• LM6134 (NSC) - V=2.7/12, GBW=10MHz, Bipolar, 2.82€
• LMC6484 (NSC) - V=3/15, GBW=1.5MHz, CMOS, 2.55€
• MCP604 (Microchip) - V=2.7/6, GBW=2.8MHz, CMOS, 0.78€

 

[MikeMl]

Even a crummy old LM358 has a common-mode input range down to -0.5V when the negative supply is 0V.

If lightly loaded, or if the load that it is driving is returned to 0V, it will swing within a few mV of 0V.

btw, how long do you suppose the LED will last if you apply 24V to it without having a resistor in series with it?

 

[Reloadron]

As was mentioned the LM741 isn't a good choice for what you want to do. You can try about the same circuit using one of these as Mike mentioned.

Most transducers that I have used to convert voltage or current to pressure are generally 0 to 10 volts, 2 to 10 volts, 0 to 5 volts, 1 to 5 volts, 0 to 20 mA or 4 to 20 mA = 0 to 15 PSI or 3 to 15 PSI. So in your case a few mV won't matter much coming off a 12 bit DAC.

Additionally as mentioned looking at your posted circuit your LED will have a very short life unless it is an LED bought for 24 VDC. It will need a series current limiting resistor.

Ron

 

[eichler]

Thanks for the replies (and sorry for my late reply)!

I have followed your suggestions and replaced the op-amp for something more suitable. I removed the LED and changed the circuit slightly to suit the new op-amp. The circuit works perfectly.

cheers for you help! J

 

[MikeMl]

Is the top "0V" terminal tied to the bottom "0V" terminals external to the circuit? Some how, they have to be tied together to guarantee that the inputs remain within the allowed common-mode range of the amplifer.

 

[eichler]

The two 0 volt inputs are not tied together. However, both are actually at 0 with respect to ground and always have a potential of 0 volts between them. One is connected to the ground of a DAQ card and the other is connected to the zero volt terminal of the psu. I have tried connecting them together and it makes no difference to the function of the circuit (in this case). Cheers. J

 

[MikeMl]

As long as they are "connected" externally, then there is no problem.