Originally Posted by capn

Hi guys, I recently changed out the nano board on my tracker as well. I was having an issue with my tracker constantly causing the servos to twitch/jitter when the tracker was motionless. I thought that perhaps it was because of the fact that I had mounted the sensor on top of the nano and was getting some sort of interference. That board was a knock-off nano, built by another company. I had no issues with the pan returning to zero on that tracker, but I thought maybe I would try swapping the nano with an actual arduino nano and since I have, the servos are silent when the tracker is motionless.

Here's the strange part of this though...after swapping out the generic nano for the arduino nano and uploading the code and doing all of the calibrations, I was having the same pan returning to center problem as everyone else. This was also after I remotely located the sensor board with about 8" of wire. The tracker was not causing the twitching in the servos anymore, but was basically useless as a result of the panning problem. So, I tried doing a different method of calibration for the magnetometer. I mounted the sensorboard on my goggles with the sensors up, and the round mounting hole located to the right side of the goggles. Instead of situating the goggles and thus, the sensorboard the exact same way as they are shown in the pictures in Dennis' mag calibration gui, I simple started the first one with the goggles flat, upright and facing away...then clicked set. I then rotated the goggles 180 degrees (now facing me) and clicked set in the second part... then I rotated them back 180 degrees and flipped them upside down for the final setting. For whatever reason, this completely fixed the pan returning to zero and actually the tracker is so smooth and functional now that it is amazing. Perhaps you guys could try the same procedure and see if it works for you. Again, with my first build and calibration I was making sure that the sensorboard was in the exact orientation shown in Dennis' pictures (e.g. in the third picture it shows the sensorboard up, but the text says "now position the unit upside down".)

Hopefully that didn't just confuse everyone and might actually work for someone.

capn

Thanks for sharing. A wrong magnetometer-calibration would explain the return-to-center problem. But I can't explain the difference between the two calibrations you have done.

All the problems with faulty Arduino boards and sensorboards are quite annoying. Buying from Sparkfun etc. should ensure some better quality. With some minor changes the code can also be used with other sensorboards.

A few have asked for the GUI-code. It was never the plan to release it (mainly as it's a bit messy and not coded very well), but as I'm too busy to keep things updated.. oh well

https://dl.dropbox.com/u/3947315/HeadTracker_GUI.rar

Dennis,

How much is involved in making the code work for the sparkfun 9DOF stick ( https://www.sparkfun.com/products/10724 )? If I were to buy one and send it to you could you update the code for it?

capn

Originally Posted by capn

Dennis,

How much is involved in making the code work for the sparkfun 9DOF stick ( https://www.sparkfun.com/products/10724 )? If I were to buy one and send it to you could you update the code for it?

capn

It seems to be pretty much the same sensors used, as on the cheap sensorboard. With some luck you might be able to make it work with just very few or no changes. I don't have much time at the moment, and would feel bad making you pay for hardware that I don't have time to look at, but if you are comfortable with a bit of coding yourself, I don't mind to give you a little guidance on the necessary changes.

Hi Guys

Someone suggested that the Arduino Nano needs to separated from the GY-85 IMU Sensor Module because of noise. Still no luck.

Cheers

Jason

DFHT Test 3 (1 min 36 sec)

Originally Posted by Dennis Frie

A few have asked for the GUI-code. It was never the plan to release it (mainly as it's a bit messy and not coded very well), but as I'm too busy to keep things updated.. oh well

https://dl.dropbox.com/u/3947315/HeadTracker_GUI.rar

Thanks!

Do you plan on implementing a Read button in the GUI? When adjusting the settings in the GUI it's easy to forget what you entered last time. Or is it already there? I might be missing something.

If it's not there, it should "just" be a matter of sending a $Debug to the serial port and parsing the answer.

Originally Posted by XDreamer

Thanks!

Do you plan on implementing a Read button in the GUI? When adjusting the settings in the GUI it's easy to forget what you entered last time. Or is it already there? I might be missing something.

If it's not there, it should "just" be a matter of sending a $Debug to the serial port and parsing the answer.

As always I'm quite busy, and at the moment I use my spare-time on other projects. I don't plan to add the button atm, but as you mentioned, it just requires the GUI to read the serial-data. Shouldn't be very difficult.

Originally Posted by WhiteNite1971

Hi Guys

Someone suggested that the Arduino Nano needs to separated from the GY-85 IMU Sensor Module because of noise. Still no luck.

Cheers

Jason

http://www.youtube.com/watch?v=wL7YKL3DW2Q&feature=plcp

I haven't seen that behaviour before. Do you know what kinda PPM-signal your transmitter accepts?

Maybe a suggestion for a new project:

A small 433MHz RF beacon on arduino based:
which has an input for various GPS (i mean it does understand different baud, update,.. settings, so we can use the GPS we use for the OSD)
and a small arduino with LCD for the ground station to see the coordinates.

I know that similar projects do exists, but no ones based on the simple arduino boards. ;-)

Originally Posted by Dennis Frie

As always I'm quite busy, and at the moment I use my spare-time on other projects. I don't plan to add the button atm, but as you mentioned, it just requires the GUI to read the serial-data. Shouldn't be very difficult.

I'm on it! Will post the result here... I just didn't want to start coding if you were already on the way of implementing it.

Originally Posted by XDreamer

I'm on it! Will post the result here... I just didn't want to start coding if you were already on the way of implementing it.

Very nice, thanks
I can give you access to google-code if you want to upload the improved version

Hi!

It seams that my Arduino nano came without boot loader.

Is it possible to upload boot loader without AVR?
I think i know the answer, but hope...

THX,
Gábor

Originally Posted by Gabek

Hi!

It seams that my Arduino nano came without boot loader.

Is it possible to upload boot loader without AVR?
I think i know the answer, but hope...

THX,
Gábor

You can do it with the cheap USBasp from HK:

http://www.hobbyking.com/hobbyking/s...occessors.html

And it's in stock right now.

Originally Posted by HOL

You can do it with the cheap USBasp from HK:

http://www.hobbyking.com/hobbyking/s...occessors.html

And it's in stock right now.

Thank You HOL,

My friend has one, and the boot loader is uploaded now.

But!

I seam to have strong interference from 72 mhz!
So the tracker stops working. No reaction at all.
When i remove 72 mhz modul than it works as it should.

I will try to use shielded cables.
Will see what happens....

Anyone had similar problem?

Gábor

Just an FYI,

I was able to get a second tracker working well. Testing it without heatshrink on the sensorboard (the gy-85 is mounted remote from the arduino with about 1meter of cable) it worked perfectly. The graph in the GUI responded as it should, so I decided to shrinkwrap the sensorboard to protect it. After I got done, I tested it again and the pan was returning to zero and it was behaving poorly. I removed the heatshrink and situated the wires away from the top of the sensor (I heatshrinked the wires on top of the sensorboard in order to make the whole thing neater and smaller). After clearing the wires away from the sensorboard, the tracker began working well again. I could make the graph show bad behavior simply by moving the wires back and forth over the sensor as it sat motionless. I don't know if it's the fault of the wire I used (an old wired mouse's cable) or if the board is just sensitive. I reapplied the heatshrink, this time making sure the wires ran at the edge of the board, away from the components on its top and all is well again.

capn