Radio control car telemetry system

Ductapemaster

I've been looking for something else to work on in my spare project time after I just recently got my SX lite kit. I think I have finally found a good project that will actually benefit me in some way and not just be some blinking lights or something like that. I have been thinking about telemetry systems more and more because recently, many companies in the Radio Control world are now manufacturing pre-built systems, but they're relatively expensive, about $100 to $150. I would like to design my own system that will be a 'real time' telemetry system. Eventually I might break down and buy a pre-built one once the prices start to drop, but this would be a great start and from what I have though up so far, a much cheaper alternative.

So, more with the project itself...I'm going to start work on a telemetry system for my R/C car. The car itself is powered by a nitromethane burning engine and is pretty large, allowing for less size/weight constraints. Also, the car has a rechargeable battery pack (its a little over 6 volts fully charged) so that solves the power issue. I plan on using an array of sensors, including a thermometer for measuring engine temperature, an accelerometer (maybe), and a couple of infrared sensors for measuring engine RPM's and RPM's after the transmission (for calculating speed). The system will most likely be a real time system, with the SX updating an LCD display that I will have along with the controller for the car.

Now, for the questions...As far as thermometers go, whats the best way for me to measure the engine temp? I need something small enough to (hopefully) fit through the cooling fins and it must measure temps with a max of about 300+ degrees centigrade. Also, is an infrared sensors the best way to go for measuring RPM's? I cant think of any other way to do it at the moment...And one last question, what is the best way to send all this data to another SX? I would prefer the data to be updated pretty fast (5-10 times a second if possible) so what kind of wireless link would I need to use? I know I need to control the LCD by parallel, as a serial interpreter chip in the middle of all this wouldn't keep up with the updating.

Let me know if you guys have any other suggestions.

Thanks for tuning in,

Dan

Capt. Quirk

You should ask these questions in the SX forums, I think you will get more help there.

I have been working on a similar project with an SX 28 & 52, also a BS2p-40. I would recomend using the melexis hall sensor and·tiny magnets instead of a led optical sensor for rpm.·The optical sensor is sensitve to dirt and the Hall is not. Eagle Tree uses something like an LM34 and type K thermocouples for some of their temp sensors. Also the Parallax App mod protocol is an efficient way to communicate between SX's with only a single wire.

Before you start though, I would recomend buying the Full SeaGull system that includes a LED screen for your radio. It is CHEAP compared to most Real Time or even normal Data Logger. Works decent too. (it uses a PIC)

The only real problem with the SX I see, is the rpm sensor vrs the other sensors. Depending on how accurate you want to be. Possibly the use of Interrupts may help, but the Propeller would do the job easier.

Bill

Ductapemaster

Yeah, hall effect sensors would work muck better there. I completely forgot about them when I wrote the post...Thanks. Also, thanks for the tips on the thermal sensors and the App Mod protocol. I will need to look into that.

Oh, and for the RPS sensors...I would just use an interrupt/prescaler technique to get a timebase and count the pulses from the sensor. Put the two together and you have revolutions and time (=RPM).

Thanks for the tip on the SeaGull system. I had looked at that some time ago when it first came out and they did a review in my R/C magazine. It's a bit over my budget right now (I'm 16 and on a student budget), but maybe eventually I will give it a try. The one I have been looking at most intently is the one manufactured by Spektrum. It's basically the same thing, just without all the options for add-ons.

Thanks again,

Dan

Capt. Quirk

Could you explain a bit more about how you would use the Interrupt and Prescaler?, for RPM.·I haven't tried much with interrupts with my SX's. Also I am hung up on accuracy and in my opinion even RPS * 60 is too inaccurate.·(and my models top out at only 11,500)

Whats the link to the magazine? TY

Also: a traction control wouldn't be that hard either.

Ductapemaster

You have an external oscillator that pulses a pin and you set an interrupt routine to count the pulses on the pin. The prescaler just divides the count internally by a certain number. This count would just be stored in a variable and would give me a time to divide the counts from the hall effect sensor by. And here I'm not to worried about accuracy, somewhere in the ballpark of 2.5% to 5% accuracy is pretty good for me. My car tops out at somewhere above 30k RPM so with error I could get within 1000 RPM. Thats pretty good for me, but if you can find another, more accurate, way to count pulses on a timebase, that would be great.

Also, the magazine I subscribe to is RC Car Action (www.rccaraction.com). I will look through my old copies for the review I mentioned.

Thanks,

Dan

Capt. Quirk

Why don't you build an inertia dyno too. (Not for HP, but for how many seconds it takes to accelerate to peak HP, That is more important) Or just write your code to use the car as·the load and Stamp Daq to figure and plot torque, hp and acceleration of your current setup of your car (F = M*A). That way you can have·ideal data from the street and compare it to·the track.

The reason I don't care for normal methods of measuring rpm is that many data loggers give you the choice to log rpm every .25 to 1 seconds (smaller intervals = less accuracy). When you look at all the data packets, you can see·the trend, but the data is still all over the place. In my next version I am using a dedicated Cog of a Propeller, just to measure RPM

Post Edited (Capt. Quirk) : 5/31/2007 4:07:36 PM GMT