A Fully Functioning Prototype
Student experience 16th March 2016
It’s finally there! This week, seven months of development finally resulted in a fully functioning completed prototype. Comprising of hardware, software, and mechanical aspects, WILFRED is now a completely usable wireless data logger.
HARDWARE
Unfortunately, this completion is about a month or two behind schedule. An essential component order being shipped from the US was hit by delay after delay, out of the project’s control. This meant that only now am I able to complete the initial testing and debugging that should have been conducted weeks ago. And with the project report deadline looming, there’s not that much time left.
Short of blowing a voltage regulator along the way, the hardware was pretty much perfect. This second PCB revision appears to have done the job, although has some room for improvement. The clearly visible free space on the PCB shows that the size could be reduced, bringing down the overall cost. In a final design, it would also be sensible to use surface mount components (instead of through hole) to further reduce cost.
SOFTWARE
Most of the software had been previously tested, but that didn’t mean it was bug free. Only by testing it in a fully functional system did the bugs become apparent. Issues ranging from standby mode to button pressing resulted in hours of scratching heads.
By far the biggest issue though was the wireless communication. As always, wireless isn’t straight forward, and a pain to debug. With time running out, it seemed at one point possible that the wireless may not even be functional for the final prototype. Fortunately, however, all the issues were resolved with a day or two to spare, with a big sigh of relief.
I was even able to throw a program together which runs on a laptop, displaying the real-time measurements from the data logger on a graph. Although far from the ‘final product’, it’s acts really well as a proof-of-concept, showing off the device functionality.
MECHANICAL
Probably the part of the project I’m proudest of is the 3D printed enclosure. It turns an amalgamation of hardware and software into a potentially marketable product.
It is evident, however, that there’s much room for improvement in the design. For instance, the back-lit sensor symbols aren’t that clear. What was intended to be a clearly recognisable symbol is more of an ominous glow. The solution to this I believe is illuminated Perspex insets, however that doesn’t seem straightforward in this case considering the accuracy of 3D printing.
Overall, the months of work have paid off. I’ve spoken to course mates in the last few days who haven’t got their project as far as they’d liked, so I’m really happy with the outcome for the data logger. Even with improvements apparent, the device has all the desired functionality initially set out, and could go into a classroom tomorrow. Talking of which, it will be! I’m about to take WILFRED into a primary school for some road-testing, I’ll let you know how it goes next post!
Leave a Reply