Difference Amplifer Design with Single Supply Voltage

[akhurash]

Hello, I need some help with a difference amplifier design I'm working on. The amplifier has to be work from a positive voltage supply only. I'm having a hard time getting it work with single supply voltage (it works when I use both negative and positive supply voltages). Both the inputs have a peak-to-peak voltage of about 2.2V (with the 2.5V DC bias). Also, the input voltages is not hitting the rail voltage (5V).

When I use a single supply the curve below 2.5V is railed and the positive voltage is not railed BUT the amplitude decreases dramatically. I think the single supply isn’t working because when the subtraction takes take it negates the DC bias voltage. Am I correct in this assumption? If so, how can I design a difference amplifier with just a single supply?

Attached is the difference amplifier circuit design.

 

[berkeman]

Hello, I need some help with a difference amplifier design I'm working on. The amplifier has to be work from a positive voltage supply only. I'm having a hard time getting it work with single supply voltage (it works when I use both negative and positive supply voltages). Both the inputs have a peak-to-peak voltage of about 2.2V (with the 2.5V DC bias). Also, the input voltages is not hitting the rail voltage (5V).

When I use a single supply the curve below 2.5V is railed and the positive voltage is not railed BUT the amplitude decreases dramatically. I think the single supply isn’t working because when the subtraction takes take it negates the DC bias voltage. Am I correct in this assumption? If so, how can I design a difference amplifier with just a single supply?

Attached is the difference amplifier circuit design.

The TL071 opamp is not designed for use as a single-supply opamp. Use a different opamp -- one specifically designed to be used with a single-supply. What do you find when you use Google to search for such an opamp...?

 

[akhurash]

I asked about single supply op-amps here actually and I was told TL071 would be a good start. I tried other op-amps and when I googled for them I was lead to TI and Analog websites. I used the part search on Digikey to try few other single supply op-amps, such as LMH6609, LT1006, and TL084. I also tried a difference amplifer (can't remember the name) and that also didn't work when using single supply; although I don't think the difference amplifer was meant for single supply applications.

 

[akhurash]

Right now I'm using the LMH6609 because I need an op-amp with high slew rate.

 

[berkeman]

I asked about single supply op-amps here actually and I was told TL071 would be a good start. I tried other op-amps and when I googled for them I was lead to TI and Analog websites. I used the part search on Digikey to try few other single supply op-amps, such as LMH6609, LT1006, and TL084. I also tried a difference amplifer (can't remember the name) and that also didn't work when using single supply; although I don't think the difference amplifer was meant for single supply applications.

Right now I'm using the LMH6609 because I need an op-amp with high slew rate.

I'm not sure why anybody would suggest the TL071 opamp for single supply operation -- it is meant for low-noise split supply operation.

The LM324 is the classic single supply opamp, although there are others. It is not all that fast, though, so it may not fit your application.

You can look at CMOS rail-to-rail opamps... They work fine for single supply operation, although they are a bit more expensive than the classic LM324.

 

[akhurash]

If anyone is interested in knowing how I got this to work please see the attachment. I'm sorry but I don't' remember where I got that info from or else I would've given credit but that is how to do it.

You have add a DC bias to the inputs and in your gain you will always have a positive component which will ensure you are never hitting the negative rail (in my case the negative rail is 0V and I used a DC bias of 2.5V).

 

[NascentOxygen]

It's not clear — have you solved what you were having trouble with? Are you wanting the diff amp to operate right down to DC?

Your second attachment looks like it should work, but be aware that it doesn't present a fixed input impedance; Z_in varies with the amplitude of the output, so this potentially introduces some non-linear distortion.

 

[akhurash]

I have solved what I was having trobule with. But I didn't think about the input impedance, :(. Although this shouldn't matter as much for my application.

 

[NascentOxygen]

I have solved what I was having trobule with.

So you've discarded the arrangement in your opening post, it being something that won't work?

If using down to DC, be aware that whereas your input signal sources might be designed to, say, source current at 1.8v, when connected to your biasing arrangement, that signal source will find itself needing to sink current when delivering 1.8v.