One Sentence Sound Byte:
“Grillzilla - A wireless meat grilling thermometer which alerts the user whether their entrée is cooked according to USDA recommendations via LCD and voice feedback.”
Summary of what we did:
As the weather starts to get warmer a common type of social gathering that occurs is the barbeque. Barbeques are all about the variety of drinks, good weather, interesting company and the grill. Meats of all types are cooked to each person’s preference on the grill, and it is of the utmost importance that the host fulfills his guest’s requests. Unfortunately, the person who is hosting the party is often obligated to stay near the grill and oversee the progress of his work. This can seriously hamper the griller’s social interaction and also makes him smell like raw beef until long after the party.
As a solution we built a device that allows the griller to detach him or herself from the grill and mingle with the people that he or she is trying so hard to impress with his stellar cooking skills. Such interaction could include a sporting game of volleyball, or a game of badminton. Either way such activities are not possible when one is tied up in cooking for the entire barbeque.
We came across this concept because of experience with a similar device that Brookstone sells – the “Grill Alert Talking Remote Thermometer” (here). However, we did not merely copy Brookstone’s design. Instead, we recreated the functionality of the Grill Alert system, and improved upon several of its features that users have complained in reviews (see amazon.com and cnet.com). In addition, we designed our device with cost in mind, as Brookstone’s Grill Alert is quite expensive for the average consumer ($75.00 US).
Our transmitter and receiver for the device was built on the foundation of Meghan Desai’s work in Fall ’05 (“Wireless Protocol”) and Spring ’04 (“Wireless Telemetry”) with the Radiotronix RCT-433 and RCR-433 RF chips . This was the first system that we built, and it took us about a week to get it fully functional. To get an idea of how to wire all the components we put it on a breadboard first. We then tested for proper communication with a modified Meghan Desai’s code (Figure 1).
After debugging the transmitter and receiver circuits, we transferred over the design to two of Bruce Land’s custom PCB boards for the ATMega32. This required a fair amount of soldering and debugging.
After both boards and the LCD were able to function without the aid of the STK500, we moved on to building our digital thermometer of a cheap thermistor probe (here). In order to accomplish this task, we had to research the dynamic range of our thermistor, as well as the application of a Wheatstone bridge to get a linear voltage response from its measurements (here). We also performed some empirical experiments on our thermistor probe to ensure that it was calibrated properly.
We attempted to incorporate a large LCD display, the Hyundai HGS2504, into our design for the receiver. Although supposedly easy to write for, we were not successful in displaying anything on the device (see Figure 2). This was after we studied numerouscodeexamples and consulted multiplesources for help. After struggling with the LCD for four days, we decided that we had to move on if we wanted to complete the project. Luckily, we sampled a 4x20 LCD display from Bruce Land, and quickly incorporated it into our design. This design mishap turned out to be fortuitous in several ways, as we will discuss later.
Next, Jeff split off to work on implementing buttons for our receiver device, as well as an audio amplifier for an 8 ohm speaker. Much breadboarding, soldering, and flights of fancy ensued (Figure 3). Matt concurrently worked on the software design of the receiver. This included several user menus, incorporation of USDA recommended cooking temperatures into LCD feedback, and a speech alert system. Much debugging and yelling at Codevision ensued. (Figure 4).
Figure 1: Testing reception and transmission of data
Figure 2: The Hyundai HGS2504 – scrapped from our design
Figure 3: One of the many, many, many times that Jeff soldered
Figure 4: Matt at the end of a long day of coding