From what I have read it seems that FPAA could provide a way to design a "programmable" analog computer, but are there any online resources on how to implement such a system?

  • \$\begingroup\$ Ask a single question, please. Otherwise expect it to be closed. \$\endgroup\$ – Brian Carlton Mar 16 '11 at 20:49

Don't expect to build analog computers.

They are mainly useful for custom reconfigurable analog filters & some schemes with opamps, that's it.

Can't suggest any resources except datasheeds on your chips :-)

  • \$\begingroup\$ BarsMonsters: Thanks for your reply,I asked as I have seen some abstracts of research papers on use of FPAAs in solving diff eqns. I suppose should look at more conventional designs. \$\endgroup\$ – Alexdagre8 Mar 19 '11 at 9:14
  • \$\begingroup\$ Analog computation has a rich history and potential, but it's fallen out of favor due to the relative difficulty of constructing such systems vs. digital "equivalents". This is what I've gathered from casual research over the past few years. @BarsMonster exaggerates a bit – analog computation is useful for many kinds of systems as is chemical computation, but he's right, don't expect to be doing anything useful anytime soon. \$\endgroup\$ – terrace Mar 19 '11 at 20:22
  • \$\begingroup\$ Well, I am not against analog computations. I am just saying that FPAA would not do the same for analog computations, as FPGA does for digital. \$\endgroup\$ – BarsMonster Mar 19 '11 at 20:28
  • \$\begingroup\$ Agreed, though theoretically it could ^_^ \$\endgroup\$ – terrace Mar 19 '11 at 21:24
  • 1
    \$\begingroup\$ @msutherl My aim is to build an analog system capable of simulating simple physical systems like damped oscillations,similar to the bouncing ball system shown here. youtube.com/watch?v=qt6RVrmvh-o \$\endgroup\$ – Alexdagre8 Mar 19 '11 at 22:37

As someone who actually worked on designing a programmable analogue computer in the first year of my PhD, I'd say that there are interesting applications for such machines. However, like what some of the others have said, it's rather niche.

You need to start to think of analogue as analogous instead of continuous. Once you do that, you should be able to design analogue computation circuits with FPAA. However, there is limited modern literature on this matter. You would need to go further back to the 1950s or so to find books on the fundamentals of designing analogue computers.

You can find some newer papers from this academic.


I have a Cypress PSoC series 1 programmer w/ 2 chips that I bought for around $50. I haven't gotten around to using them myself, but I know somebody who his using them on the data acquisition side of a prototype digital musical instrument. He uses it to amplify and digitize signals from fiber-optic sensors.

There's also a project in the book iPhone Hacks that uses a series 1 PSoC to make a sensor input to the iPhone using the microphone jack.


Last I checked, the companies that used to make pure PGA's went out of business.

EDIT: This post at SparkFun has some information: http://forum.sparkfun.com/viewtopic.php?f=5&t=577

FPAA's are a rare breed. Ther have been three other companies that tried to create these. One went under (they also wanted $20 a chip!), one was reorganized and the FPAA "disappeared" from their product lineup, and the last one, Zetex, still exists, but have one product and it doesn't appear to be all that well supported.

EDIT: Oh, I forgot about this guy: http://servenger.com/products/index.htm – programmable analog signal processing board.

  • \$\begingroup\$ Thanks for your response, I am sorry about changing the question. \$\endgroup\$ – Alexdagre8 Mar 16 '11 at 21:19
  • \$\begingroup\$ Now my answer doesn't answer the question anymore. Might be useful anyway, so I suggest keeping it. \$\endgroup\$ – terrace Mar 16 '11 at 22:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.