CURRENT ISSUE

Contests

bottom corner

Feature Article



Issue #204 July 2007
Pyxos Power
by Tom Cantrell

Start | LAN-In-the-Box |Down to the Wire | Variable Interest | Hook 'Em Danno |Build it and they will come | Match point | Sources and PDF

MATCH POINT

Before finishing up, I want to complement Echelon on the quality of the kit and tools. This is complicated stuff, but for the most part, everything worked as advertised right out of the box. 

The only glitch I encountered was that the network activity-logging feature of the demo application didn’t appear to work. It took a call to Echelon to find out that the feature is only enabled if your display resolution is greater than 1024 × 768 since otherwise there’s just not enough room on the screen to show everything.

Between the various manuals for the chip, kit, and tools, the documentation is exceptionally complete and well-written. For example, the chip databook includes a very comprehensive 10-page itemized “check list” for your design. Follow it and you will find and correct 99% of the potential hardware problems before you power up the first time.

With the kudos out of the way, let’s get to the bottom line. Is Pyxos FT right for your application?

Of course, for existing LonWorks applications the answer is clearly yes. Pyxos FT gives those designs the ability to cost-effectively extend the network’s reach from the building down to the room and box level while using familiar techniques and tools. 

On the other hand, there are clearly simple applications that you can get by without. For a physically small and self-contained device with relatively few nodes, key Pyxos FT features (up to 32 nodes over tens to hundreds of meters including the link-power option) go unutilized. At the same time, the costs (less than $2 for the Pyxos FT chip in volume, MCU code-space required, etc.), little they may be, loom relatively larger. For the simplest applications, hacking a one-shot hardwired solution, or even just sticking with the old wiring harness approach, remain viable options.

Somewhere in the middle is the sweet spot for Pyxos FT. Ask yourself these questions. Are there enough nodes and protocol complexity that a dedicated resource (i.e., the Pilot) is called for? Are there enough wiring concerns (e.g., range, noise, and immunity) that a transceiver chip of some sort is required in any case? Is the link-power capability a must-have that justifies the possible extra cost of dealing with a 24-V supply and the need to derive power locally at each node? Is your application a one-shot project, or will it grow and evolve to encompass a range of devices, which could benefit by using a single standard network? Does rolling your own networking solution represent a development chore that you’d just as well leave to Echelon?

There are many world-beyond-wiring-harness applications for which at least one of the answers is “yes.” If yours is one of them, Pyxos FT is definitely an option worth considering.

Previous | Next

 


bottom corner