LM6171: High Speed_High current Amplifier

Well Nithin, I don't think TI has anything with 1.5A output, 

You could consider paralleling 4 of the THS3122 devices (it is a dual, so two packages) with load sharing ballast resistors. This is done a lot in AWG outputs where the THS3491 product folder has an example design showing considerations. 

The THS3122 is a CFA so it has the speed and slew rate for your 20Vpp, 10MHz signal. 4 of them in theory would have the 1.5A output capability.

There is no claw curve in this old datasheet, so not sure if you can get by with +/-12V supplies or would need more like +/-15V due to IR headroom increase at max output current. 

Hello Nithin,

Below is the load sharing reference design Michael is referring to:

Best,

Hasan Babiker

Well Hasan, that is quite the document. Rohit covered about everything - on the scale of a full datasheet effort. Lot's of good stuff inthere, it does in fact help to parallel devices to reduce HD. Took awhile to get this out, but is there now for the AWG community. 

Hi Michael,

Thanks for your valuable points,

As per your comment we decided to go ahead with THS3122  (2 nos)

we have input voltage in the range of 1- 9 v 

can you please suggest the best gain option (Low gain or high gain ) for obtain  constant response for our required bandwidth (100Hz - 10 MHz) with output voltage of 20 Vpp.

considered power supply = +/-15V with  non-inverting configuration.

It will be great help if you suggest  gain setting resistor 

Ok Nithin

Apparently you can be AC coupled - that is very helpful, not sure what your load is but here I used 6.66ohm to create a 1.5A swing on +/-10V. Those ballast resistors do cause an IR drop adding to the output swing so they cannot be too high and introduce a loss to the load through their combined 1.05ohm out. I really worry about mismatched output DC offsets creating high DC currents through those resistors. I might consider stage by stage servo amps to hold each output at 0V DC using a precision servo amp. 

Simulates nicely, 

4 parallel THS3122 driver.TSC

Here is small signal AC

And then +/-10V output, the other waveform is the required input - but this could be offset as you said 

And Nithin, 

This is what I mean by a servo amp, this will hold each THS3122 output at 0V eliminating DC current due to output offset mismatch. Essentially another high pass filter at lower frequency than the input side. The DC probe is on the OPA197 output to show it is introducing an offsetting signal, 

And tthe AC response is unchanged, 

And this single stage TINA file,

single THS3122 stage with OPA197 integrator.TSC

Hi Michael,

Thanks for the valuable support ,

One more doubt.

Whether each channel of THS3122 will carry 450 mA load current at 20Vpp?

Conformation on above doubt will be a great help.

Thanks

Nithin