Ping Jim Thompson Sine Shaper question ????

ah, got ya :)

Zeners would do it, is that what you mean?

NT

No. A "catcher" loop that stops the OpAmp back-swing when it isn't in "clamp" mode. ...Jim Thompson

I'm not imagining it. Maybe show us some time when your deadlines are dead.

NT

Interesting and pretty idea. I wonder how many degrees of freedom you need to allow for the not quite ideal tanh function of real components.

I gave it a try in a spreadsheet and optimising for |x| < pi/2 got a surprisingly good fit to a sine wave using least squares.

pseudosine(x) = (tanh(ax)-bx)*c

where a = 0.3374, b = 0.2708, c = 9.386

A slightly lucky solution got an even better fit requiring the residuals to be multiplied by 1000 to be viisible on the same graph. Leading error term is about 0.02% of third harmonic judging by eye.

a = 0.3813, b = 0.27166, c = 9.125

Insisting on having absolutely no discontinuity in gradient at the peak ties down the solution but might not be minimum THD.

I can't quite reproduce your results exactly but the conclusion is that it ought to work rather well with a bit of fine tuning. It is a bit tetchy about the parameter tuning > 3 sig fig needed ideally.

Quicky...

In the sine shaper case this "R" will affect the sizing of the PWL step resistors. ...Jim Thompson

Right, cheers. Not come across that one. My remaining braincell was trying to think in terms of adding feedback when output got within a diode drop of a rail.

NT

Oh, I like it. Thanks.

George H. (A thread on simple (1) opamp circuits would be fun for me... right, I'll start on when I have more time.)

NT

-- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | STV, Queen Creek, AZ 85142 Skype: skypeanalog | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at

I'm looking for work... see my website.

Thinking outside the box...producing elegant & economic solutions.

are you allowed 2 votes? I won't tell though.

NT

Accumulating ;-) ...Jim Thompson

Yep, I can see how to do that with a pair of opamps. One monostable to set pulse width, 2nd integrator to add the incoming pulses analoguely.

NT