Microchip’s PIC16(L)F170X - Embedded.com

Microchip’s PIC16(L)F170X

At last week’s EELive! (née ESC) event in San Jose Microchip announced a new family of 8-bit MCUs.

8 bits? Isn’t that so 20 minutes ago?

Apparently not. Microchip continues in both growth and profitability despite resisting the ARMslaught of 32-bit microcontrollers. Fully three quarters of Digi-Key’s MCU part numbers are non-ARM parts, and a third of those are PIC devices. They continue to garner design wins through their low pin count, low power, low price, and huge range of available on-board resources.

A couple of features of the new family caught my eye, though the 451 page datasheet covers many other peripherals and functionality.

First, there’s a zero-crossing detector. Now, that’s not terribly innovative, and we’ve been doing zero crossing for decades. But these parts can be connected directly to the AC mains! Check out the required circuit from their app note:

220 or 110 right into the chip. Be careful probing the board!

No doubt this would be unwanted in some applications like medical circuits where mains isolation is a must. And the thought of working on a PCB with high-voltage AC lurking is a bit scary.

Like some of Microchip’s other parts these new MCUs include a “puddle of gates” (their term) called Configurable Logic Cells (CLCs). It’s a poor person’s FPGA which takes up to 32 inputs and reduces them to four outputs via a logic design set by control registers.

At first glance this is a nice but unexciting feature. However, all of the logic runs without CPU intervention. Even if the processor is asleep it can continue to function. One application ties the zero-crossing input through the logic to handle a relay. Using logic, rather than software, means microsecond or better response times without the jitter one expects from code-implemented solutions. Firmware engineers, prepare to learn some logic design!

Configurable logic cells put a bit of random logic on-chip.

But it’s not all about ones and zeroes (“Every idiot can count to one,” said Bob Widlar, famous analog designer who detested the digital world). A pair of configurable op amps can take inputs from pins and/or a DAC or on-board fixed voltage reference. It is an analog world, after all. The amps’ specs are nothing to get excited about (CMMR 55 dB min, gain-bandwidth product 2 MHz typical), but are more than sufficient for lots of MCU applications.

Firmware engineers, prepare to learn a little analog design! Which is, of course, a ton of fun.

Configurable op amps bring analog design to the MCU world.

The press documents claim 30 uA/Mhz operation, which is an almost unheard of number. I can’t pry that out of the datasheet, and suspect it’s a typical rather than worst case result. So, as they say, your mileage may vary.

A “Complementary Output Generator” (with the mandatory TLA COG) can use a single PWM to drive two or four outputs in a variety of modes (lots, and each gets too little documentation in the datasheet), like push pull. That’s ideal for generating high-efficiency switched-mode drive for all sorts of loads. Everything, like deadbands, is fully programmable.

A typical COG in push-pull mode.

A peripheral pin select feature is new for Microchip. It allows the software to remap the pins used by each digital I/O. That’s useful for easing PCB routing, reducing cross-pin coupling, and, most specially, for fixing those all-too-common board layout goofs.

To complement all of these features the company is releasing their Code Configurator, a plugin for the MPLAB IDE. You specify how the peripherals are to be used and it generates the code.

Code Configurator – Notice pinouts reflect the selected peripheral pin select setup.

According to Digi-Key these microcontrollers are under a buck in quantity. At the moment they are listed as non-stock, but Microchip is good at getting new parts to market roughly at the same time they’re announced.

6 thoughts on “Microchip’s PIC16(L)F170X

  1. I don't know why you would compare ARMs to 8-bitters. It is like comparing trucks to bicycles.

    Both have their place.

    Sure, ARMs are taking business from what used to be “top end” 8-bitters and 16 bitters, but the same forces that give us cheap

    Log in to Reply
  2. You're right that the reason to use 8-bitters is some superior 'analog' features, such as power supply brownout or HV resistance such as the above-mentioned PIC 220VAC sensing. Price it ain't: the entry level ARMs are around $0.50 (Kinetis K02, etc).

    Log in to Reply
  3. Wouldn't 8-bitters typically be less vulnerable than 32-bitters to random bit flips or permanent damage caused by cosmic radiation and/or an EMP event, due to their larger process geometries and simpler circuitry?

    Log in to Reply
  4. In large quantity production runs, the 8 bit machines will ALWAYS be less expensive. Not just the smaller number of transistors but the shorter test times ( and lower contact counts) all make their production much less expensive.

    Log in to Reply
  5. This is increasingly becoming the lot of the 8-bitter: some non-CPU circuit needs a small amount of processing power, so put in an 8-bitter.

    This PIC is basically an analogue circuit with a small amount of digital processing, the same can be said of some

    Log in to Reply

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.