I'm new here, so if I put this in the wrong place, somebody please correct me.

My school's robotics team is building a flying robot and we're considering using RC heli gyros. The big problem is after calling a couple of companies and searching the web constantly, I haven't been able to find anything related to signal I/O. So far, all I've learned is that many companies do not publish information regarding signals because it doesn't fit within RC applications (or something like that), and that gyroscopes do have an input called gain (my guess is this is sensitivity).

What I would like to know is how most RC gyros handle output. It is my understanding that we're going to let a microcontroller on the robot handle all of the stability/movement related issues, such as correctly adjusting control surfaces. Since we're on a limited budget, ordering one to play around with may not be feasible. If anybody happens to have experience in this area, I would greatly appreciate your help.

Thanks in advance,
Phillip K.

You must understand that the gyro amplifier takes pulses for inputs. The pulse length normally encodes servo position information. The two inputs from the receiver: one channel is desired yaw, and the other is gain/mode (depending on the gyro brand and model). The output is a pulse train that combines the input information from the gyro sensor and the two receiver channels as a command stream to the yaw servo. I do believe that such information as pulse frequency and width vs. signal are published somewhere.

Gyro drives a servo directly.

Servo connector 3 pins are Vcc, Gnd and data.

Data is pulse width, where only "uptime" matters critically. 1.5ms pulse is center, approximately 1 ms and 2ms are the endpoints. (leave marginal). Send the pulses at 18ms sequense or so. Thats not too critical thougt. And if you measure, just measure the "uptime"

Jyrki

The 18ms is actually important. Overdriving a servo (lower than about 10ms between pulses) has a chance of burning out the servo motors, because the deadzone on analog servo's isn't very good and it can cause feedback inside the servo if the pulses are too close together because they're not internally voltage regulated very well, this makes the motor tweak at high frequency (you won't be able to physically notice this but it's equivilant to a dead short electrically). This is brand and actually batch dependant (no two servo's will react quiet the same way overdriven at different levels) but it's definitly something you don't want to push without testing.

ravingidiot, what is this robot supposed to do? Fly steady? Fly in a straight line. Please let me know what exactly it is you're trying to do because there are a few methods (both physically speaking and design speaking) that you can utilize to counteract undesired results. For example, with a hellicopter the tail is designed to couteract the torque created by the blades. A disc of solid steel/lead/whatever geared to rotate in the opposite direction of the blade of a helicopter made to weigh the same (centrifically weighted) amount as the blades at a given speed will eliminate the need for a gyro, at least from a human standpoint. If it's supposed to be an autonomous vehicle you'll still need a gyro it just won't have to be as high precision to handle the demands of counteracting the torque created because it's reduced from the counter rotating mass.

Hi ravingidiot, I have made some projects that involve connect an RC gyro to a PIC and I think that the best solution is to use directly the analog output of the sensor. You must amplify the signal but this is easy with an OP amp.

If you like more information look at this project :
http://www.rcgroups.com/forums/showthread.php?t=297067

This project uses 3 gyros connected to an 16F876 PIC (via analog inputs).

I hope this helps...

Jose.

Thank you for all your responses.This is more than I could've ever hoped to find on Google.

In response to Lynx's question:

We're basically building a robot that can take off, and within a slight range, can tilt in any direction and move around (at least that was my understanding of it). As it stands the propellor is mounted at the top and there are four control surfaces underneath that give control over all three axes of motion by working in certain combinations. From what I can understand all of the payload is going to be mounted on the outside to help with stability. As far as I know, the robot is piloted via a transmitter on the ground.

JoseJ:

I'm glad to hear something like that has been done before.The person over the robotics team was strongly considering such an idea, but wanted to be sure about specs before we take any action. Nifty heli, btw!