I first became aware of these during a visit to the AAF Tank Museum in Danville, VA. They have an awesome RC tank battlefield up there and host major events several times a year. These are 1/16 scale fully articulated radio controlled tanks from Tamiya. They have an IR-based combat system which allows them to basically play "laser-tag" with other tanks. Each tank can take so many hits before shutting down and can only fire once every few seconds (time to reload).  Anyway, some organizations (e.g. RCACN) provide standard rules and sponsor events that include formal, structured battles.

I belong to the tank club H*E*A*T, based in NC.  Our website is entirely new, check it out. An active club of experienced tankers running predominately Allied tanks (my Tiger is the exception). The H*E*A*T club currently has a single four man competition team, the Red Devils.  They have won a couple of RCACN battles held at the AAF Tank Museum, including the last two competitions! Come up and see us run sometime, you will be hooked too!

 This was the first tank I built. A stock build with only a couple of mods.  I changed to a new gearbox from etoarmor with 480 motors. I added the recoil package from Wecohe. Both have worked great for me. I could not recommend this model strongly enough for someone's first build. The parts fit well and there is plenty of space to work inside. Overall, a great model and a blast to build.

After running these things a few times up at Danville, I found that it might be handy to have more of the bookkeeping at my fingertips. For example, I would like to be able to know when "reloading" is complete and I am ready to fire again. I would like to know how many more hits I can take and other handy bits. There was plenty of room in my transmitter for some extra electronics, thus, the Vibrating Battle Tally (VBT) project was born (thanks to Dana Lowell for the name). I considered various hardware approaches to adding these features to my radio transmitter and it led to this project including both analog and digital circuits. In the end, I ended up adding a microcontroller (a Basic Stamp right now) into the space in my radio where the trainer board would be. I added a couple of switches to interact with the microcontroller and a pager vibrator for non-visual, non- (barely) audible feedback.

A complete writeup of this project with a lot more details can be found here. This version is based on the BS2 OEM chip from Parallax. The circuit for this design is pretty much the same as version 1.0, except it uses a different microcontroller that is easily 2-4x faster and has 10x the memory capacity. To keep the cost down and allow for some future options, I opted for th OEM version of the Stamp module. As a result, I really needed to put together a circuit board.  I have a physical circuit board designed (using Eagle) and manufactured by a company in the pacific rim (just email them the .brd file and do a "paypal-like" thing and they FedEx you the boards in 5 days). The design holes out many additional pins so that they can be used in other projects and additional designs. Here is an example board:

I have built one of these out and it looks like this:

The only variation here is a 0.1uF capacitor between +5v and Gnd where the regulator would be. I tap the 5v supply inside my transmitter directly, so no battery is needed (but the board includes holes for a voltage regulator if you choose to use one anyway. This design can be built at a much lower price point that version 1.0 and has a lot of spare capacity. Perhaps most interesting from my perspective is that the Stamp chip can be replaced with other 28pin Microchip PIC microcontrollers. Some of the new features I am considering will required the additional speed I can get from such a switch.

I have this one up and running now in a 27Mhz Futaba radio and it is running well. You need to be careful with the placement of the wires (like version 1.0) to avoid interference, but it does work out just fine. Now I have plenty of spare space to fill up. I now have a software update that uses the hit counter as a pair of reload timers to monitor other tanks firing at you as an option.

Thanks to another of the HEAT tankers, we have a VBT running in a 75Mhz Futaba 6 channel radio.  The radio already had a mod to fire via buttons and this links in parallel with it.  We removed the trainer board and put the VBT in that space (perfect fit). The power switch was wired to one of the unused shoulder switches. One important note, the amount of interference is far less with the 75Mhz radio than the 27Mhz radio.  I will try to get some pics (the current ones are a little blurry).

The next project will be a working control for a fixed gun emplacement and control of a smoke generator with IR detection.  Stay tuned...

VBT Version 1.0 (no longer an active project)

This version is based on the BS1 chip from Parallax. A  complete writeup of the version 1.0 project can be downloaded here. The code for this program nearly completely filled the memory of the BS1 module and I still had some things I wanter to add, so I made one working copy of this version and sold it to another tanker to start work on version 2.0.

On RF transceivers, Marco-Polo protocols and Manchester codes

I've been giving the topic of asynchronous protocols a lot of thought recently.  I picked up a pair of TX/RX chips from Rentron recently and have been looking to upgrade the system to monitor the tank systems directly. The big problem stems from a number of tanks talking at the same time, corrupting the airwaves. I have come up with a scheme that looks like it will work. A PIC chip running in the tank keeps track of the tank hit status and monitors the RF for packets. The chip in the tank and the one in the radio have the same internal serial number. The radio starts by sending a packet of:

SOP
{command byte}
{serial number}
{command payload}
EOP

The number of bytes in the command payload is implied by the command byte. All tanks see the packet and the one that matches the serial number can respond with another similarly formed packet. The entire serial stream must be encoded such that byte/packet boundaries can be detected and autosynchronized.  Many systems use Manchester coding for this (and there are some advantages of that approach), but I planned to use a 9bit scheme with a limit on the number of consecutive set bits. In this scheme, the SOP and EOP each start with 5 set bits and then 0 or 1 for SOP or EOP.  No other 9bit sequence has more than 4 consecutive set bits and no more than 2 bits set at the start and end.  Looking at the 512 possible 9bit values, we have over 300 workable patterns, so mapping  256 values to the 300 can be easily done with a table. The system should work well and has both sync properties and error tolerance. Now, all that having been said, there is a great alternative that is pretty cost effective. Saelig has a module (ER900TRS) so slick, I will have to buy one on first principles (~$40). Not only does it handle all the packet and Manchester and transmission collision issues, it buffers small packets so the PIC only needs to fill a buffer and say go and makes it look like a broadcsat serial port. Very slick chip, especially if you are limited by the speed of your PIC (e.g. using a STAMP).

More updates on this front.  I am leaning toward using an 802.15.4 ZigBee chip now.  This chip has everything you could want and is only $20!  It needs a little TLC to talk cleanly with a PIC, but I just picked up a PIC programmer (Easy PIC3) which I like alot and am working on VBT 3.0 using PIC chips directly.  The ZigBee could end up as part of this project in a couple of ways (remote programming, direct tank queries, etc).