Circuits to assess the performance of single vs. Dual power supply in an OP- Amp

[Samarpan]

Hi, I am going to write a research paper on OP AMP and i found the topic of power supply very intriguing hence, i looked for resources that would give me more knowledge on the Performance between these two supply modes, but i couldn't find much information about it. I was wondering if you could show me circuit that I can use to assess the performance (in various levels) of the OP AMP doing the same task.

Thanks in advance.
Samarpan

 

[crutschow]

The only significant difference between one supply versus two is the input and output voltage range near zero volts. Obviously an op amp with a single positive supply cannot handle input or output DC voltages below zero volts. And if it is not a rail-rail type op amp it can't go down to zero volts input or output with a single supply.

 

[Samarpan]

And what are the implications, by that i mean, If i have a dual supply op amp doing a particular function which lies in the domain of single supply op amp too, then i do think we can see many performance changes. Can you advice me circuit(s) that i can use to assess these factors?

 

[MrAl]

Hello there,

Well do be fair, if you use a power supply of plus and minus 10 volts for example then you have to use a single supply of 20 volts. That means the op amp gets almost the same bias anyway.

But i dont think you'll see too much difference, except in some areas which i'll mention here.

First, there is usually a limit on the output voltage as to the max and the min that it can put out. This means the upper range can be severely limited on a plus and minus supply while it will not seem to be as much with a single supply. For example the LM358 is spec'd to output up to Vcc-1.5v, which means for a dual 10v supply with the output ranging from 0 to +10 we'll only get 0 to 8.5v, and that's 85 percent of the positive rail. But with a single supply, we'll get about 0.1 to 18.5, and that's 92.5 percent of the positive rail.

Second, the same goes for the lower end, but is less severe in some cases. For an op amp that can put out 0.1v up from ground, that means the dual 10v setup can actually put out 0v while the 20v setup can only output 0.1v. This actually becomes a big problem in some single supply setups.
Also, some op amps cant even get that close to either rail, so there the output range is limited even more so the choice of supply could make a difference whether the circuit works at all or not as intended.

We can also say roughly the same thing about the inputs. The input ranges is usually limited too so we have to choose carefully sometimes.

If you are interested in the actual running characteristics then set up an op amp of your choice in both circuits to act as an amplifier. See if the bandwidth changes at all for either power supply choice. The internal circuits are usually biased with constant current circuits so there probably wont be too much difference, except maybe for power op amps. Do some measurements and compare.

ADDED LATER:

Oh yeah one more thing, since the output slew rate is more or less constant that means an op amp that has to slew 20 volts will take twice as much time to reach max (or min) as that which only has to slew 10 volts (to zero or max). Of course if the dual 10v supply op amp circuit has to slew from max minus to max plus supply then it will take the same time.

The output voltage and input voltage ranges become very important with op amps that have a range that is something like +4v more than the negative rail and -4v less than the positive rail obviously because the range is very limited for lower voltage power supplies. They wont ever work right with a single supply of +5v only.
The other type which have their output go close to ground and not too close to the positive rail are still a problem sometimes with that single supply of +5v. That's because the output may be limited to 0.1 to 3.5v which sometimes isnt enough in a 5 volt system.

 

[Samarpan]

Thank you very much for your reply. Wow, i am just noob in hardware. And the information were overwhelming. I did understood the following (maybe) :
1. Voltage clamping in single supply
...and thats basically it.

I have done some of my own research of how i can use it and i was interested to learn about 'slew rate'. Can you guide me further? Moreover, for the analytical part, can you suggest me circuits that i can usec to inspect, record and analyse the difference in characteristics shown. .

Thanks again

 

[Samarpan]

Or can u point out me some factors i need to examine to measure the performance?

 

[MrAl]

Hello again,

Well, the voltage mode op amp is usually used in a circuit where the feedback keeps the inverting terminal voltage very close to the non inverting terminal voltage. But during various transitions, the two voltages will be very different. That means the op amp goes into open loop mode and the output voltage starts to ramp up (or down) until it can satisfy the feedback again and get the two voltages close to one another again. That's normal linear operation, but the transition period takes the op amp into it's open loop mode so it behaves as if it were a ramp generator that tries to ramp up as fast as possible. It turns out that various internal mechanisms limit this ramp time so it ends up being somewhat slow, but still fast enough to meet the demands of the application. The maximum rate at which it can ramp up is known as the "slew rate" of the op amp and is usually measured in volts per microsecond. So a slew rate of 10v/us means that the output can ramp up at a maximum of 10 volts every microsecond. If the output has to jump up very quickly, this can seriously limit the ability of the op amp to follow a fast changing input signal. Because of this, it also limits the ability of the op amp to pass an undistorted sinusoidal wave even within the bandwidth of the op amp.

For your test circuits, i would suggest an ordinary inverting amplifier where you can change the feedback resistor to change the gain from 1 to 2, and maybe 5 and 10. You can then measure various things like slew rate, bandwidth, distortion too if you wish.

What kind of test equipment do you have or can use?

 

[Samarpan]

Thank you very much for the information on slew rate. I will keep that in mind.

Well i have osilloscope at my school,
Digital Multimeter and i play around arduino and i can make function generator out of it (crude one though) or other transducers like temperature sensor etc. So can i use the process you specified to test op amp in their two power supply modes? Can you add more experiments that i can do?

- samarpan

 

[KeepItSimpleStupid]

I will also add that you can group into:
1) dual supply - tracking
2) dual supply - non tracking
3) single supply

So, here's your paper **broken link removed** You just have to do your own research.

 

[MrAl]

Thank you very much for the information on slew rate. I will keep that in mind.

Well i have osilloscope at my school,
Digital Multimeter and i play around arduino and i can make function generator out of it (crude one though) or other transducers like temperature sensor etc. So can i use the process you specified to test op amp in their two power supply modes? Can you add more experiments that i can do?

- samarpan

Hi,

Well if you have a scope then you can use the Arduino to generate a square wave and look at the output of the op amp amplifier to check the slew rate. That should be simple enough. Again the op amp is connected as an inverting amplifier with various gains.