ED Tracker 2.9 (Friday 13th)

Dead Fred

Dead Fred
I must agree. There are a lot of really good people here.
I think I am the biggest numpty.
And even I got it to work with a bit of fiddling and a bit of help!

:)
 

Dead Fred

Dead Fred
n0ahg said:
The InvensenseMotionDriver directory should be inside you libraries directory along side the I2CDev directory.

I am using my custom board as it is a lot easier!
Click to expand...

What is this custom board you speak of? can we have a picture please?
Might help me understand.
:S
 

n0ahg

N
Dead Fred said:
What is this custom board you speak of? can we have a picture please?
Might help me understand.
:S
Click to expand...

Fred, you are the 'sane brain' (same as mine?) around here...

The custom board is a description within the Arduino IDE which includes all the core file customisations and libraries as one which is installed under Documents\Arduino\hardware\
You then simply select 'ED Tracker' to compile with our mods and it doesn't affect the base install!

Edit, I'm drinking Directors ATM!
 
Last edited:

Dead Fred

Dead Fred
Teachers, My old man used to dabble with that. :)

Oh that custom board.
I thought you were talking about a real one.
All this terminology is potentially confusing.

All my ones appear as an Arduino Micro.

If it ain't broke don't fix it unless you really know what you are doing.
That's my methodology.
I am riding on the shoulders of two giants and the view is fabulous!

:)
 
Last edited:

pocketmoon

P
djcslip said:
Good question - i'll get back to you in a bit with the answer.

EDIT:
Can rotate my head left (the device went "hard right" to +180) Also - can re-zero with "3"
Click to expand...

That sounds fine as far as the flip issue goes (it would stick at 180). The drift in your final pic is a on the high side and the yaw has drifted round 180 degrees over the course of 20/30 minutes.

In game (with the UI closed) the sping-back should keep the yaw drift under control but I would go through the calibration/drift comp process again with the sensor as flat as possible and aim for a final yaw drift of less than +/- 1.0

I've put a vid up of the calibration process. It's not yet annotated but if you watch in 720p in a large window you can follow the mouse to see what I do.

http://www.youtube.com/watch?v=4ozsWSO2X-o&list=UUSw0UVl3II08WffAI7ugdfQ&feature=share

Cheers

Rob
 

Voz

Voz
I've just ordered the bundle from Hobby Components and put my hand up for a PCB through EDTracker.org.uk . Really looking forward to seeing if even I can get this working :)
 

Lasse B.

Lasse B.
pocketmoon said:
http://www.youtube.com/watch?v=4ozsWSO2X-o&list=UUSw0UVl3II08WffAI7ugdfQ&feature=share
Click to expand...

Thanks for posting this. Looks like the accelerometer of my tracker is faulty, because it's all over the place even with the unit flat on the table. I let the unit heat up for like two hours, and 7 calibration attempts later it's still drifting between -30 and +30. I can get the raw X accel to single digits when I'm lucky, and it won't stay there for long. Just now I started another recalc, starting raw x accel at -30, and after recalc was done accel was -40.
 

pocketmoon

P
Lasse B. said:
Thanks for posting this. Looks like the accelerometer of my tracker is faulty, because it's all over the place even with the unit flat on the table. I let the unit heat up for like two hours, and 7 calibration attempts later it's still drifting between -30 and +30. I can get the raw X accel to single digits when I'm lucky, and it won't stay there for long. Just now I started another recalc, starting raw x accel at -30, and after recalc was done accel was -40.
Click to expand...

Hi, can you run the UI and post a screenshot before any calibration ?

[edit] Actually fluctuation +/- 30 isn't too bad. Checking the specs for the MPU-6050 i think +/-30 or 40 is still within the noise and bias/calibration limits. +/- 30 works out at +/- 2mg. I'll sample a few more devices when I get home and see what the spread is. The accelerometer values are passed through a Low Pass Filter before being used by the chip for fusion. So as long as you can get sensible DMP behaviour (drif compensation less than +/10) from the main sketch you should be fine.
 
Last edited:

ZZleeZZ

Z
With such a good product you guys are bound to go pro with this but I would just like to offer some advice first.

If you want to appear to be a large professional organisation then all this helping people out with their problems is going to have to stop ✋

You don't catch t-mobile offering solutions to their customers when things go wrong do you? No, to be more like them you will need someone with an accent as thick add frozen treacle (Welsh, Indian, Birmingham are all good ;) ) to read badly scripted answers that are in no way related to the questions asked, they will be required to use the words "escalated" and actioned too often, your helpline could then either pretend to transfer the poor hapless soul to another department but instead just hang up or put them on hold listening to good music amplified to the point that it sounds like a screech.

If you aspire to playing with the big boys then heed my advice.
 
Last edited:

Voz

Voz
ZZleeZZ said:
With such a good product you guys are bound to go pro with this but I would just like to offer some advice first.

If you want to appear to be a large professional organisation then all this helping people out with their problems is going to have to stop ✋
Click to expand...

Love it :D

Set those expectations low and the customer can never be disappointed!
 

ZZleeZZ

Z
brumster said:
Now this I can help with! Awwroooooiiiit?!
Click to expand...

I did have you in mind when I wrote that but I dont think you will fit the bill as I have no trouble understanding what you say in your vids. Im sure you have some friends or relatives you could draft in :)

Going shopping now, try not to change the world too much before I get back, finding it hard to keep up!
 
You must log in or register to reply here.
Back
Top Bottom