VansAirForceForums  
Home > VansAirForceForums

-POSTING RULES
-Advertise in here!
- Today's Posts | Insert Pics

Keep VAF Going
Donate methods

Point your
camera app here
to donate fast.

  #1  
Old 01-31-2019, 01:37 PM
Dugaru's Avatar
Dugaru Dugaru is offline
 
Join Date: Jan 2016
Location: Richmond VA, USA
Posts: 485
Default Next Garmin autopilot question

So I've got the 307, G5, and two GSA28s in an RV-9A. What kind of "accuracy" should I be seeing in turns if the autopilot is properly configured?

For example, let's say the autopilot is flying this GPS approach. I'm coming from the west and doing maybe 100 knots as I fly from Harcum to JUBJI:

https://aeronav.faa.gov/d-tpp/1902/09364G1.PDF

My 430W will plot, and the G5/307 will follow, a turn (indicated by a dotted line) that smooths out the corner as I get near JUBJI and start heading north toward IFESO.

When executing this turn, the plane will however overdo it to the left by about 10-15 degrees, before straightening out and heading to IFESO.

Meanwhile (if I don't command a new altitude) the altitude change will be maybe - 40 feet before it corrects and settles back at the original altitude.

The same things happen in shallower turns in cruise flight, although there's definitely less overshoot in shallower turns.

This overturning is not as significant as with my S-TEC with GPSS in the old Warrior, and is certainly not a major problem. But is it par for the course, or does it indicate that my autopilot settings still need tweaking? Is the thing actually capable of riding the plotted course like a rail?

Assuming I need to do some more tweaking, are there "starter" settings recommended for an RV-9 and the latest G5 software?

Thanks
DMG
__________________
N929JA, 2007 RV-9A
Based W96: New Kent International Aerodrome
(near Richmond, VA USA)
2021 Dues Paid
Reply With Quote
  #2  
Old 01-31-2019, 03:06 PM
gfb gfb is offline
 
Join Date: Jul 2014
Location: Madison, WI
Posts: 715
Default

I have similar overshoot problems and have been waiting for weather to go back above zero to re-do the autopilot tuning procedure.

Has anyone managed to get it so it doesn't overshoot?
__________________
---
RV-9A - started 01/2015, completed 07/19/2018, 500hrs 03/14/2020
RV-10 - started 04/2020
Reply With Quote
  #3  
Old 02-01-2019, 10:13 AM
g3xpert's Avatar
g3xpert g3xpert is offline
 
Join Date: Sep 2010
Location: Olathe, KS
Posts: 2,924
Post

Quote:
Originally Posted by gfb View Post
I have similar overshoot problems and have been waiting for weather to go back above zero to re-do the autopilot tuning procedure.

Has anyone managed to get it so it doesn't overshoot?
Hello GFB,

Seeing this autopilot over-shoot a turn would be very unusual.

I have been flying this autopilot since 2012 and have never seen it over-shoot a turn, and I fly lots of approaches, all with video and data logs.

I would start by reviewing the data log and making sure that the servo isn't being limited from doing its job by a Maximum Torque setting that is too low.

For example, if you have your maximum roll servo torque set to 25%, and you see in the log that it is pegging out against the 25% limit in a turn, it is being prevented from having as much roll authority as it needs to fly the plane accurately and follow the flight director commands.

Please feel free to contact us if you need assistance in resolving this or in interpreting the data log.

Thanks,
Steve
__________________
Garmin G3X Support
g3xpert@garmin.com
1-866-854-8433 - 7 to 7 Central Time M to F
Please email us for support instead of using Private Messaging due to the limitations of the latter.
Reply With Quote
  #4  
Old 02-01-2019, 10:17 AM
Brantel's Avatar
Brantel Brantel is offline
 
Join Date: Mar 2006
Location: Newport, TN
Posts: 7,503
Default

Agree with Steve, you should be able to fine tune this AP to better than airliner performance. If you are having to go way outside the box with settings or the AP is not performing well after fine tuning, something is wrong.
__________________
Brantel (Brian Chesteen),
Check out my RV-10 builder's BLOG
RV-10, #41942, N?????, Project Sold
---------------------------------------------------------------------
RV-7/TU, #72823, N159SB
Lyc. O-360 carbed, HARTZELL BA CS Prop, Dual P-MAGs, Dual Garmin G3X Touch
Track N159SB (KK4LIF)
Like EAA Chapter 1494 on Facebook
Reply With Quote
  #5  
Old 02-01-2019, 10:40 AM
gfb gfb is offline
 
Join Date: Jul 2014
Location: Madison, WI
Posts: 715
Default

Glad to hear it's likely just a tuning issue. I'll try to reproduce the problem in the next few flights.

If it helps, I have all my flight telemetry uploaded to the fly.garmin.com logbook. Can you guys access that and search for the issue you mentioned? This would be a perfect use for machine learning
__________________
---
RV-9A - started 01/2015, completed 07/19/2018, 500hrs 03/14/2020
RV-10 - started 04/2020
Reply With Quote
  #6  
Old 02-01-2019, 11:02 AM
lr172 lr172 is offline
 
Join Date: Oct 2013
Location: Schaumburg, IL
Posts: 5,659
Default

Also, I am not sure you should expect it to follow the pink line. I thought the GPS steering turns were done by calculating an appropriate roll rate, based upon the degrees of the turn and the current airspeed. It also picks the start and end points of the roll command. I am sure the 430 models this on the screen to draw the pink line, but each plane will behave somewhat differently to the commanded roll. You do need to insure that you are not limiting the AP's roll rate to less than what the GPS steering commands are calling for. I don't believe that this steering approach accounts for wind either (though it might - I am not sure), so that will move your actual track relative to the plotted course.

I don't believe that the 430 plots an arc that AP precisely follows, like it does a straight line.

Larry
__________________
N64LR - RV-6A / IO-320, Flying as of 8/2015
N11LR - RV-10, Flying as of 12/2019

Last edited by lr172 : 02-01-2019 at 11:07 AM.
Reply With Quote
  #7  
Old 02-01-2019, 12:45 PM
g3xpert's Avatar
g3xpert g3xpert is offline
 
Join Date: Sep 2010
Location: Olathe, KS
Posts: 2,924
Post

Larry,

Pretty rare for it not to be on top of the magenta line in my plane. Here is a GDU 470 screen shot I took yesterday while in a holding pattern at the end of the missed approach. You do have to keep the ball centered since the GTN calculates needed bank angle assuming the plane is in coordinated flight.



Good example of target trend. The traffic will be high and off my right wing in 60 secs.

Here is a screen shot from an Aera 660 (different flight) which is listening to the same MapMX out of the GTN as the G3X Touch system. Don't try this with a tablet!



Thanks,
Steve
__________________
Garmin G3X Support
g3xpert@garmin.com
1-866-854-8433 - 7 to 7 Central Time M to F
Please email us for support instead of using Private Messaging due to the limitations of the latter.
Reply With Quote
  #8  
Old 02-01-2019, 12:45 PM
Dugaru's Avatar
Dugaru Dugaru is offline
 
Join Date: Jan 2016
Location: Richmond VA, USA
Posts: 485
Default Interesting point

Quote:
Originally Posted by lr172 View Post
You do need to insure that you are not limiting the AP's roll rate to less than what the GPS steering commands are calling for.
Starting to wonder if this is my issue. Or perhaps I'm not allowing the AP to fly a steep enough bank angle.
__________________
N929JA, 2007 RV-9A
Based W96: New Kent International Aerodrome
(near Richmond, VA USA)
2021 Dues Paid
Reply With Quote
  #9  
Old 02-01-2019, 12:48 PM
Dugaru's Avatar
Dugaru Dugaru is offline
 
Join Date: Jan 2016
Location: Richmond VA, USA
Posts: 485
Default We should compare notes

GFB, what are your current settings for your autopilot? Since we fly the same aircraft, I'm thinking we should probably compare notes.

I'll try to get out to the hangar this weekend and write mine down.

Quote:
Originally Posted by gfb View Post
Glad to hear it's likely just a tuning issue. I'll try to reproduce the problem in the next few flights.

If it helps, I have all my flight telemetry uploaded to the fly.garmin.com logbook. Can you guys access that and search for the issue you mentioned? This would be a perfect use for machine learning
__________________
N929JA, 2007 RV-9A
Based W96: New Kent International Aerodrome
(near Richmond, VA USA)
2021 Dues Paid
Reply With Quote
  #10  
Old 02-01-2019, 01:46 PM
Dugaru's Avatar
Dugaru Dugaru is offline
 
Join Date: Jan 2016
Location: Richmond VA, USA
Posts: 485
Default Forgot I had this

From about a year ago. This was a fairly gentle turn at cruise speed, and the overshoot was about 6 degrees, with maybe +/- 30 feet on the altitude. Both would be more pronounced on a 90-degree turn.

Again, hardly scandalous performance, but apparently improvable:

https://share.icloud.com/photos/0hE9...UQvqUJvwpU9E_Q
__________________
N929JA, 2007 RV-9A
Based W96: New Kent International Aerodrome
(near Richmond, VA USA)
2021 Dues Paid
Reply With Quote
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


All times are GMT -6. The time now is 10:33 PM.


The VAFForums come to you courtesy Delta Romeo, LLC. By viewing and participating in them you agree to build your plane using standardized methods and practices and to fly it safely and in accordance with the laws governing the country you are located in.