OTA and OP: circuit-wise,how to spot the difference?

hello guys.

there has been so many theories and explanations regarding OP and OTA.
circuit-wise, please tell me the similarities and differences between these two.

Please tell me also if my question does make any sense at all.

Thank you......

Basically, they have the same structure. The main differences are :
1. The output of OTA, acronym of Operational Transconductence Amplifier , is current. So it has very high output impedence. While the output of Opamp is voltage ,so it's output impedence is small. Therefore, the Opamp can drive small resistences while the
OTA can not . For driving capacitors, if they have the same output current capability, the output impedence make no difference.
2. OTAs are often used in open loop(no negtive feedback) ,while Opams in closed loop.
Conclusion: From the circuit perspective, you can discern these two kinds by finding whether the circuit has negtive feedback and does the circuit have a high output impedence ! Usually, Opams contain an output buffer stage!

hope this helps!

thank you very much EngEntEco...

If you are looking at a schematic, the output stage is likely
to be common-source (current mirror) style in an OTA and
source-follower / emitter-follower (Class B or AB) in a
voltage-mode op amp.

In an application schematic, if there is a resistive load it
is likely to be an op amp (or else likely to have trouble).

One supplement to EngEntEco's contribution.
The OTA is a rather versatile unit because - in most cases - the transfer characteristic (transconductance gm) can be controlled externally by a control voltage/current. Thus, it's very easy to realize steerable filter and oscillators.

dick_freebird said:

If you are looking at a schematic, the output stage is likely
to be common-source (current mirror) style in an OTA and
source-follower / emitter-follower (Class B or AB) in a
voltage-mode op amp.

In an application schematic, if there is a resistive load it
is likely to be an op amp (or else likely to have trouble).

Click to expand...

so you mean that having a common emitter/source output stage means that the circuit is an OTA ?
i have seen this kind of structure at Gray book on NE5234 op-amp an
correct me if Im wrong but it was sayed the purpose of putting a common emitter at output is to prevent out put stage from going off and as a result increasing speed
and also NE5234 is an Op-amp not an OTA

I would say an OTA is an "amplifier" which has only one high impedance node and that is its output (it could "look" low impedance by just having large transistors and current in the output stage) but the output impedance would be rdsNmos||rdsPmos ( or what ever is the output transistors). Since it only has one high impedance node it thus has a voltage input and current output. (correct me if I am wrong here)

An Op amp has two or more high impedance nodes and needs compensation to be stable. Now the output doesn't need to be an source followers( ~1/gm resistance) but can also be a common source (Class A or class AB) output rdsNmos||rdsPmos. I would say the most common in IC design is using the common source Class A, AB output, and if you need to drive a large load (small resistance) you would sacrifice the gain in the last stage, or spend the current to have the low resistance. In all of my amplifiers I have designed, I have never used a source follower, its just a waste of current and power (Class A)!

Does anyone else agree or disagree?

Jgk

Emitter follower output stages have been a typical characteristic of classical OP designs, e.g. 741, and are still used for most bipolar OPs, including high speed types. But as correctly mentioned by sina_extreme and jgk2004, an output follower is not essential for an OP. Obviously, it can't work for rail-to-rail output stages, neither bipolar nor MOS. A low open loop output impedance in a several 10 to 100 ohm order of magnitude is nevertheless a characteristic of general purpose OP, although not strictly essential. Without output followers, it can be achieved by the internal compensation feedback.

An OTA is basically a linear gm element, possibly controllable. It's linearity is essential for applications, where it's operated fully or partly in open loop, e.g. controllable amplifier or filters. For an OP, a linear transfer characteristic is less strictly required, but high linearity in closed loop can't be achieved without it.