HobbyKing.com New Products Flash Sale
Reply
Thread Tools
Old Mar 31, 2012, 01:41 PM
Al Ducharme
photronix's Avatar
Orlando, FL
Joined Jan 2005
734 Posts
Discussion
Potential GPS performance enhancement

We believe we have found a potential reason for the GPS position hold poor performance seen by some users.

We are working to make this process better and easier for you. But for now we believe you will see improved PH performance by repeating your accelerometer calibration. Here is what you should be aware of. The pro/gps boards must be level to each axis when going through the calibration. BUT level must also be parallel to the crafts level. This may be confusing. Level for the craft is when you switch on autolevel, without wind, and the craft does not drift in any direction.

Performing the trim during autolevel will not affect the PH hold performance.

Please try the following:

Start by zeroing out your trims for ail/elev. Now repeat the accelerometer calibration. Fly once again and try autolevel. If the craft drifts then the calibration was not done properly. During the calibration the positions you used must be off from the crafts level during flight. We believe if you repeat this process untill you do not have drift in autolevel then PH performance should improve. Try slightly different angles.

If autolevel drifts with zero trims then the PH performance will be poor.

If autolevel does not drift then we believe you PH performance should be good.

Once again. We are working to make this easier for you. But for now if you can try this and try to report back your results.

I would like to know.... If you have no drift in AL after calibration is your PH performance good? If you do drift is PH performance poor.
photronix is offline Find More Posts by photronix
Reply With Quote
Sign up now
to remove ads between posts
Old Mar 31, 2012, 02:12 PM
A man with too many toys
United States
Joined Feb 2001
17,001 Posts
Looks like I need to get out my big carpenter’s level for this one.
RC Man is offline Find More Posts by RC Man
Reply With Quote
Old Mar 31, 2012, 05:31 PM
Crash & Carry..
Norway
Joined Jan 2012
66 Posts
This is interesting, good news!!
This means that later FW versions will make PH check the AL trim?
Fastlane66 is offline Find More Posts by Fastlane66
Reply With Quote
Old Mar 31, 2012, 05:36 PM
manual flying apologist
Adidas4275's Avatar
United States, CA, Fresno
Joined Mar 2011
6,942 Posts
Yes hopefully they can get this to be a simpler setup process in the future.

Good to see they found something!
Adidas4275 is offline Find More Posts by Adidas4275
Reply With Quote
Old Mar 31, 2012, 06:29 PM
Al Ducharme
photronix's Avatar
Orlando, FL
Joined Jan 2005
734 Posts
Quote:
Originally Posted by Fastlane66 View Post
This is interesting, good news!!
This means that later FW versions will make PH check the AL trim?
Yes. Right now PH depends completely on your accelerometer calibration. We will make it much easier for you.

A small 5 degree error in one axis will give you a lateral PH error in that axis and lead to the circling effect with wind.
photronix is offline Find More Posts by photronix
Reply With Quote
Old Mar 31, 2012, 06:56 PM
Crash & Carry..
Norway
Joined Jan 2012
66 Posts
Quote:
Originally Posted by photronix View Post
Yes. Right now PH depends completely on your accelerometer calibration. We will make it much easier for you.

A small 5 degree error in one axis will give you a lateral PH error in that axis and lead to the circling effect with wind.
Sounds great, this explains the different behaviour of the different aircrafts..
Looking forward to try out the next FW release..!!

Any chance of speeding up the RTH a bit at the same time?
(Or add an "expert mode setup" with possibility to change this)
Fastlane66 is offline Find More Posts by Fastlane66
Last edited by Fastlane66; Mar 31, 2012 at 07:02 PM.
Reply With Quote
Old Mar 31, 2012, 08:05 PM
Flight Instructor
Chico, CA
Joined Jun 2008
683 Posts
Advanced Settings

Quote:
Originally Posted by Fastlane66 View Post
Sounds great, this explains the different behaviour of the different aircrafts..
Looking forward to try out the next FW release..!!

Any chance of speeding up the RTH a bit at the same time?
(Or add an "expert mode setup" with possibility to change this)
No one is ever going to agree on every performance characteristic so advanced settings would be nice!
mspencer is offline Find More Posts by mspencer
Reply With Quote
Old Mar 31, 2012, 08:43 PM
Al Ducharme
photronix's Avatar
Orlando, FL
Joined Jan 2005
734 Posts
We need to make sure everyone is having a good experience with PH first. Then we can add more variables.

We are trying to make sure that basic functions are working as well as possible. There seems to be alot of FC's chasing features but first just normal flight mode should provide good stable, not robitically jerky, video. Just my two cents worth.
photronix is offline Find More Posts by photronix
Reply With Quote
Old Apr 01, 2012, 06:36 AM
SuperX Hex and X8 Flyer
ross_badger's Avatar
Columbus, IN
Joined Jul 2009
632 Posts
I've been using a home-made calibration fixture for more accurate accelerometer and compass calibrations. It cost me about $20 from parts at the local hardware store. While calibrating, I use a bubble level so I am sure that the boards are perpendicular to the ground, in all directions. My PH and AL have been very good after using this setup.

The full post can be found here:

http://www.rcgroups.com/forums/showthread.php?t=1570188
ross_badger is offline Find More Posts by ross_badger
Reply With Quote
Old Apr 02, 2012, 10:45 PM
Registered User
Joined May 2011
364 Posts
Quote:
Originally Posted by photronix View Post
We believe we have found a potential reason for the GPS position hold poor performance seen by some users.

We are working to make this process better and easier for you. But for now we believe you will see improved PH performance by repeating your accelerometer calibration. Here is what you should be aware of. The pro/gps boards must be level to each axis when going through the calibration. BUT level must also be parallel to the crafts level. This may be confusing. Level for the craft is when you switch on autolevel, without wind, and the craft does not drift in any direction.

Performing the trim during autolevel will not affect the PH hold performance.

Please try the following:

Start by zeroing out your trims for ail/elev. Now repeat the accelerometer calibration. Fly once again and try autolevel. If the craft drifts then the calibration was not done properly. During the calibration the positions you used must be off from the crafts level during flight. We believe if you repeat this process untill you do not have drift in autolevel then PH performance should improve. Try slightly different angles.

If autolevel drifts with zero trims then the PH performance will be poor.

If autolevel does not drift then we believe you PH performance should be good.

Once again. We are working to make this easier for you. But for now if you can try this and try to report back your results.

I would like to know.... If you have no drift in AL after calibration is your PH performance good? If you do drift is PH performance poor.
Hi Al, I believe this actually may be it...looking at the logs and comparing them to the flying experience it seems like after trimming autolevel and switching on PH, the copter just starts drifting in the direction it was drifting in before PH and trimming. So you're basically saying that PH ignores the trims at the moment and essentially the TBE effect is because the PH algorithm is fighting against the information it gets from the ACCs?
IconFlyer is offline Find More Posts by IconFlyer
Reply With Quote
Old Apr 03, 2012, 05:39 AM
Registered User
Joined May 2011
53 Posts
I have a question. If we perform ACC calibration correctly, why do we need to trim? Is it necessary at all?
Krzysiek is online now Find More Posts by Krzysiek
Reply With Quote
Old Apr 03, 2012, 07:23 AM
Al Ducharme
photronix's Avatar
Orlando, FL
Joined Jan 2005
734 Posts
Quote:
Originally Posted by Krzysiek View Post
I have a question. If we perform ACC calibration correctly, why do we need to trim? Is it necessary at all?
Because board level may not be craft level in the air. Unless everything is mechanically spot on.

The opposite is true (I believe...still testing). If you trim in AL, why do we need accel calibration. If this is correct then we will removing the accel calibration from the setup client.
photronix is offline Find More Posts by photronix
Reply With Quote
Old Apr 03, 2012, 08:39 AM
Registered User
United States, FL, Melbourne
Joined Mar 2011
220 Posts
Been following this thread and I always wondered about how to confirm your board was perfectly "level" with your air frame. I realized that the trim function solved the issue in non-gps auto level mode and stopped thinking about it.

Does it make any sense to zero trim on a windless day, look for any drift, and then "shim" the board until the drift goes away? If I understand correctly that would make the board and the craft functionally level with each other and make all other calibrations more accurate.

Am I thinking correctly?
Dbasileccim is offline Find More Posts by Dbasileccim
Reply With Quote
Old Apr 03, 2012, 10:11 AM
Dirty Dozen FPV
Wayno Complaino's Avatar
United Kingdom, Diss
Joined Aug 2011
2,374 Posts
Quote:
Originally Posted by photronix View Post
If you trim in AL, why do we need accel calibration.
I did wonder
Wayno Complaino is offline Find More Posts by Wayno Complaino
Reply With Quote
Old Apr 03, 2012, 07:23 PM
Al Ducharme
photronix's Avatar
Orlando, FL
Joined Jan 2005
734 Posts
Made a lot of progress today with George and we will be rolling the changes into the next firmware release which I know is going to make us all very happy.

There are a whole "slew" of changes and improvements in the next release (wink wink).
photronix is offline Find More Posts by photronix
Reply With Quote
Reply


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion Enhancing transferable skills Anatomist Simulators 3 Feb 29, 2012 11:01 AM
Mini-HowTo skyangel F86 performance enhancing surgery madingle Foamy EDFs 0 Dec 26, 2011 10:52 AM
Discussion GPS Performance mspencer Hoverfly Technologies 61 Dec 23, 2011 09:04 PM
Discussion Excellent GPS Performance! mspencer Hoverfly Technologies 17 Dec 11, 2011 12:12 PM
Discussion GPS performance nickwad Hoverfly Technologies 9 Oct 23, 2011 07:27 PM