Telescopic OTA DC Gain in 130 nm

Hello,

I need to design an OTA in 0.13 µm CMOS (1.2 V) technology. My specs are:

Av=80 db
GBW>250 MHz.
Slew rate > 100v/µs

It is possible with a single telescopic stage to achieve this results?
Any idea about the power need for achieve this specs?

Regards

I think its possible (not 100%) but you will have to gain boost your cascodes to get above 80dB and can NOT use minimum channel length transistors. I would say LminX2 atleast maybe X4. As for the power I don't want to do your homework for you. It can be found in any textbook start out with ....GBW= gm/2piCL, go to gm, which leads to Id! ~~power =supplyXIdX2 since you have a diff amp hint hint

Hope this helps,
JGK

Thanks for the reply, JGK.
Don't worry about the ""homework" of the power dissipassion calculation or even the use of minimum channel length transistors =).
I just created the topic to have an overview about the possibilites of this design and "get a sense" of the power dissipation from experienced people.
Well, i will use regulated cascodes to boost the gain and I expect achieve 80 dB.
Best Regards!

pcca said:

Hello,

I need to design an OTA in 0.13 µm CMOS (1.2 V) technology. My specs are:

Av=80 db
GBW>250 MHz.
Slew rate > 100v/µs

It is possible with a single telescopic stage to achieve this results?
Any idea about the power need for achieve this specs?

Regards

Click to expand...

With a telescopic cascode and 1.2 V supply, what kind of output swing are you expecting?

Even if you use 0.15V as your VDS for each transistor, you will exhaust 0.6V in just the cascodes. That leaves you with at best 0.6V of o/p swing.

Is that ok for your design?

Hello

It is important to define your load condtion, CL and RL if it will be. especially the capacitor will play the first step in finding the GBW and the slew rate. Second , as it is 1.2 V, the telescopic OTA is not at all suitable, however it depend on the output range you need. so please provide us with the output swing as well.

pcca said:

Hello,

I need to design an OTA in 0.13 µm CMOS (1.2 V) technology. My specs are:

Av=80 db
GBW>250 MHz.
Slew rate > 100v/µs

It is possible with a single telescopic stage to achieve this results?
Any idea about the power need for achieve this specs?

Regards

Click to expand...

Hello,

I would suggest instead of a Telescopic OTA a folded cascode OTA, which has slightly lower GAIN but leaves you with more headroom, especially whit 1.2 V supply. If you use gain boosting you should have no problems achieving 80dB.

Of course in using the folded over the telescopic you have more output range, but at the cost of more power. Therefore, if you dont need the range stay with a telescopic Additionally, why would the folded have more gain then a telescopic? I would say when looking at output impedance its the same gain = -gminput*(rds^2gmoutN||rds^2gmoutP).

JGK

Hello,

While you are correct regarding the power consumption folded vs. telescopic, in my limited experience it is difficult to design a telescopic OTA with only 1.2 V.
But as JKG stated if voltage headroom is not a concern stay with the telescopic OTA, otherwise check the folded amp.

Dear friends, I can see that 80% of the design is implemented by the folded topology. this topology compromise most of the requirements of the op-amp but with a littile more power. thye second good choice is the current mirror OTA but it has a little limited ICMR.

However, both the folded and the current mirror OTA are self-compensated OTAs, theses amplifiers compensate itself from the capacitor load, the more capacitive load, the more stable it will be. therefore it has a problem if the load capacitance is very small unless if another capacitor is connected with the load in parallel.

in the latter case I would suggest the use of the conventional two stage op-amp as the compensation capacitor will be also small, the gain is very high as well as the input and output swing

I

Hello

Perhaps pcca can provide the output load value, noise and power consumption for his project and we can have a discussion regarding the optimal choice in this case.

Thanks everybody for the replies! and sorry for reply this just now.
I have to change my high level design and know I am working on it again.
As JGK said, I need do define correctly my output swing before to define the OTA topology. Since folded cascode OTA presents higher power consumption it will be employed only If the output swing desired won't be achieved with a telescopic ota.
Best regards.