Nuts and Volts - May 2015

READER FEEDBACK Continued from page 77

I've tried to stick with the heart of the Arduino — the ATmega328P with all of its limitations and benefits — and keep the concepts as true to the microcontroller as possible. By doing this, I hope that readers may more easily transition to working with raw microcontrollers, learn from the path that I struggled along, and be able to put the learning to use in exploring the broader options out there.

We've got much ground to cover, and I hope that you'll keep reading and keep the ideas and comments coming. Please do drop me a line with any thoughts — I enjoy these kinds of discussions as they broaden horizons all round.

Andrew Retallack

“Beyond” Fan

I just wanted to thank Andrew Retallack for his article "Beyond the Arduino." As someone who prefers individual components in projects rather than reusing a new development board each time (e.g., MSP430 vs. Launchpad; PIC16F84 vs. BASIC Stamp; etc.), this was a refreshing change of pace. With little to go on in the way of designing with ATmega ICs instead of an Arduino board, maybe now I too can try my hand at using this little processor in a few of my own projects.

Maybe others can relegate the Arduino back to the task for which it is best suited — a reusable prototype board for developing projects which can then be migrated to their own ATmega-based circuits.

Derek Tombrello

Propeller Preference

The frequency counter article by Jim Teixeira in the March issue is similar to something I have been planning to replace; something I built about 40 years ago.

However, I would suggest replacing most of the hardware and the PIC micoprocessor with a Parallax Propeller. It will accept inputs greater than 20 MHz, and its eight processors each have two built-in 32-bit multipurpose counters — along with the clock counter — that will provide all of the needed counting functions (and a lot more, if needed). Programming these functions is almost (but not quite) trivial.

Jerry Nicholson

PICAXEd Apart

Though I'm not involved in the PICAXE platform, I've been following Ron Hackett’s PICAXE Primer column lately. Although the circuits probably end up being functional, his explanation of transistor basics couldn't be more wrong. I refer to the "explanation" of the functioning of an NPN transistor.

Since the articles are targeting novice hobbyists, it’s important to get the fundamentals right, rather than causing confusion.

1. There are TWO junctions in a bipolar transistor, not three. They are the base-emitter junction and the collector-base junction. Perhaps you meant "three terminals."

"The switch will turn on (i.e., current will flow from the collector to the emitter) whenever the base is 'taken toward' (whatever that means) the collector, i.e., whenever the base is connected to the same voltage level as the collector."

2. A bipolar transistor is a current operated device, not voltage, i.e., the collector voltage has little control over how much current flows from the collector to the base. Current flowing into the base-emitter junction permits current to flow from the collector to the emitter, given almost any collector voltage.

The ultimate collector voltage is dependent on the load resistor and the collector CURRENT, V = IR.

Approximately, Ice = β Ibe — the collector current is equal to the base current times the beta (or current gain) of the transistor.

In order for this equation to be met, there must be limited current flowing into the base which, in general terms, requires the voltage of the base to be above approx. 0.6V for silicon. Also, the collector voltage must be above zero. The equation sets the upper limit for collector current once Vc is above Vce(sat); typically, a few tenths of a volt under these operating conditions.

Later in his article, "When the base of the NPN transistor is at + 5, the collector is (almost) at ground." If the base were at + 5, the transistor would act as a fuse, shorting the B-E junction and melting the base bond wire. He qualifies it earlier by stating R1 limits current, but still puts the base at + 5 — which it is not. Maybe you should measure Vbe.

I'm sure a tutorial on the operation of a bipolar transistor or circuit is not necessary, having built many successful projects, and smoked his share of parts (as have I). However, I feel this project is full of misconceptions.

I honestly don't understand why Mr. Hackett chose not to use a CMOS hex inverter for the adapter. It's simpler, uses fewer components, saves power, and the operation of FETs is easier to explain in voltage terms than a BJT. It is certainly as space efficient as two transistors and six resistors. Even using the stripboard technique, this project could be built one row narrower and shorter. Or, not even bother with hardware and do it all in firmware with the XOR instruction.

Steve McChrystal

Thanks for your feedback, Steve, on the March 2015 PICAXE Primer Column. It has certainly made me aware of the fact that I should have been more careful (and more precise) in my explanation of the functioning