finding arcsin using opampa

Thread Starter

ranga

Joined Feb 9, 2007
1
hi friends

i am doing project, i need help, how to implement arcsin function using opamps and othere devices.
 

beenthere

Joined Apr 20, 2004
15,819
Hi,

You may need some extra components for this. Are you actually trying to make an analog computer that will output the arcsine function?. How are you planning to implement the input?
 

kender

Joined Jan 17, 2007
264
i am doing project, i need help, how to implement arcsin function using opamps and other [analog?] devices.
I would approximate the arcsin with a polynomial and then do the polynomial in analog.

What forces you to build an analog computer instead of using DSP?
 

kpm

Joined Feb 9, 2007
3
An alternative way would be the following:
Have a stable sine wave source, say 10kHz. Put this sine wave on the minus input terminal of a comparator (such as LM393) and put the input voltage of the arc sin circuit on the plus input terminal.
At the output a Pulse Width Modulated version of the arc sin of the input voltage will be available. A simple low pass filter will convert it back to dc (or rather continuous form).
Accuracy depends mainly on the amplitude stability of the sine wave generator( or oscllator) and also the PWM signal should not be influenced by the load of the low pass filter.
 

kpm

Joined Feb 9, 2007
3
To clarify what I discussed on the previous post, I prepared a simulation.

In the attached file 'arc sine 1.gif' the principle schematic is shown
In the attached file 'arc sine 2.gif' the wave form of the circuit's output versus input is shown, as reference also the actual arcsin function is superimposed.
The attached file 'arc sine.zip' contains the LTSpice file to run this simulation.

Greetings KM
 

Attachments

hgmjr

Joined Jan 28, 2005
9,027
Kubeek's astute observation has pointed out a flaw in my logic. Reply withdrawn awaiting further consideration.

Thanks, Kubeek.

hgmjr
 

kubeek

Joined Sep 20, 2005
5,794
Another technique that just might work to solve your problem would be to exploit the log/anti-log function...
Are you really sure that arcsin(x)= 1/sin(x)?

I know arcsin is sometimes noted as sin^-1, but that just means inverse function and not 1/( ).
 

hgmjr

Joined Jan 28, 2005
9,027
Are you really sure that arcsin(x)= 1/sin(x)?

I know arcsin is sometimes noted as sin^-1, but that just means inverse function and not 1/( ).
You know, I believe you are correct. I had my trig identities goofed up. Thanks for the correction.

hgmjr
 

Sparky

Joined Aug 1, 2005
75
The acrsin() function solves for the angle for which the sine(angle) = (opposite / hypotenuse).

Arcsine(opp/hyp) = angle.

This doesn't help with the electronics yet

-Sparky
 

Sparky

Joined Aug 1, 2005
75
Greetings,

I found that the acrsine function can be approximated with a Taylor series.

Does this help? Can it be used with the electronics side?

I've attached a document showing the Taylor series.

-Sparky
 

Attachments

kubeek

Joined Sep 20, 2005
5,794
That is another possibility, but I think it is easier to implement the PWM-style rather than making three or more powers. I would use taylor series only with DSP.

Ranga, what is your desired precision?
 
Top