FPGA, CPLD, ASIC, and ASM Design

Key Points

  • Instrumentation amplifiers, which are analog circuits, are your window to the world. Do one poorly, and you'll be looking through a fogged window.
  • Properly characterizing what you're measuring is critical to getting your analog front end and data collection system defined correctly.
  • Signals can be filtered with an analog front end or manipulated later in software. Knowing where to do one or the other depends on a solid knowledge of both analog and digital signal processing theory.

Instrumentation Design

Without correctly designed instrumentation circuitry, the smartest embedded microprocessors and microcontrollers on the planet would never have a window into the very real (and complex) analog world in which we all live.
Whether it’s a biometric device for a security application, an extremely low-noise amplifier for a photomultiplier tube that measures light by counting individual photons, or an electroencephalogram looking at the millionth of a volt required to fire a neuron; getting the measurement taken accurately – before it’s digitized and operated on by an embedded processor – can mean the difference between having the information to complete a critical calculation correctly and having a buffer full of bad data.
Obviously biological reactions are one place where bandwidth requirements can be very low but electronic noise in the instrument itself can quickly swamp the actual measured signal.  So keeping analog and digital electronics separate from each other is critical, and proper printed circuit board layout can easily mean the difference between success and failure.  But even in the world of medical equipment design, the difference between a correct diagnosis and a poor one may be the availability of a few data points resolved to higher granularity – so there’s no guarantee the process of new product design and development won’t also involve the balancing of bandwidth (in order to get enough readings taken within a period of time) and the absolute accuracy of a single measurement.
In short, having a properly designed analog front end is critical to being able to make the most of all of the mathematical tools (Discrete Fourier Transforms, Fast Fourier Transforms, digital filtering, general DSP algorithms, Laplace Transforms, Poisson Statistics, Root Locus of Evans and Routh-Hurwitz for stability, etc.) at your disposal.  Knowledge of digital signal processing (and what can and can’t be done on the back end) will inform the specifications of the analog circuits that interface with the world beyond the printed circuit board.  And sometimes, with remote telemetry systems, knowledge of how the communications channel works – again a signal processing question – will determine how the captured data from the instrumentation system is packed (and perhaps compressed) before being sent on.
When you need instrumentation (or telemetry systems) designed properly, you need the design engineering services of a company that can design across the analog and digital domains as well as do the “heavy math” to determine what data should be taken and how it should be sorted.
If you need instrumentation designed, you need Focus Embedded.