Software 5.4 for EFIS D10 A on an MCR01 Sporster

[francfi]

I have a EFIS D 10A + AP 74 + 2 SV 32 with upgraded 5.4 software
Everything is Ok besides;
- gap of 30 to 50 ft between prest targeted altitude and rechad one when aircraft levels
- preset VSI is 400 ft/mn; actual one during climb or deescent is about 200 to 300 ft/mn
Please advice as I did not find any relevant assistance/explanation in the relevant manuals.
Thanks for your help.
François ( from paris France)

 

[DBRV10]

Re: Software 5.4 for EFIS D10 A on an MCR01 Sporst

I have the same issue.

I have not been able to fly the plane in the last 4 weeks since installing 5.4 in the Beta format, but it did the same thing for me.

Some of the gains....and from memory I can not recall which one it was helped the rate of VS to improve, but we always had about a 30-40 foot above the selected altitude. If you capture any altitude its fine, its just a preselected one.

I am sure it will appear elsewhere and 5.4 will get a minor update.

Can't wait for Thirsday so I can get back flying again! Will have 4-5 hours to test on the way home!

 

[dynonsupport]

Re: Software 5.4 for EFIS D10 A on an MCR01 Sporst

I have a EFIS D 10A + AP 74 + 2 SV 32 with  upgraded 5.4 software
Everything is Ok besides;
- gap of 30 to 50 ft between prest targeted altitude and rechad one when aircraft levels
- preset VSI is 400 ft/mn; actual one during climb or deescent is about 200 to 300 ft/mn
Please advice as I did not find any relevant assistance/explanation in the relevant manuals.
Thanks for your help.
François ( from paris France)

Hi François,

Here are a couple of questions and suggestions:

• What are your min/max speeds set to?   When you climb or descend, do you reach those speeds?  If those speeds are reached the AP will no longer fly vertical speed but will instead climb or descend holding airspeed.
• There is an altitude bug-bug which can result in an error of 10 feet on preselected altitudes, which we're working on.
• Altitude error of more than that suggests that increasing ALT GAIN may help.

Would you mind sharing your settings?

Thanks!

 

[flybuddy]

Re: Software 5.4 for EFIS D10 A on an MCR01 Sporst

Just fyi--after 5.4 on an RV9 (D10A), sv32s and AP74, we had identical problem--altitude approx 40 over any set vale and was only getting 300 fpm with 500 set. Yes, the min max speeds are set right out of the Vans manual as are all the v-speeds. We did adjust baro but have yet to fly after the baro adjust. I reset Vmax to 800 fpm and it is doing that vspeed well. Tried 500 fpm again and only getting 300 again so have left it on 800, will report back if the baro change helped on next flight.

 

[francfi]

Re: Software 5.4 for EFIS D10 A on an MCR01 Sporst

My settings are:
Max speed; 304 km per hour
Min speed: 140 km per hour
Torque 100 per cent
Sensivity 15
VSI 400 ft per min
All other settings on default value
With previous software , targeted values ( alt and VSI) were reached
I would rather get former software until 5.4 is finalized
How can I get it on yr website?
So far , 5.4 is not an improvememt .......
Thanks for year help
Francois

 

[dynonsupport]

Re: Software 5.4 for EFIS D10 A on an MCR01 Sporst

HI Francois,

The only way to go back to a previous version is to restore a backup.

I think you will find that 5.4 is much better than 5.3, but you will need to adjust the new settings for your airplane.

If the airplane is flying smoothly and holding an altitude, the thing to do is to increase ALT GAIN slowly until the correct altitude is held.  After that you may need to increase VSI gain if it overshoots the target alt after a climb or descent.

Increasing ALT GAIN will also cause the AP to hold the desired vertical speed until nearer to the target altitude, and not round out so soon.

Here are the complete tuning instructions for your convenience:

TORQUE:
Adjust until servo does not slip during normal AP flight, and be comfortably overridden by the pilot.

SENSITIVITY:
ADJUST TORQUE FIRST
Increase gradually until airplane "twitches" when on altitude, then reduce by one
or two. If servo slips, increase torque.

PITCH GAIN:
ADJUST SENSITIVITY FIRST
After sensitivity is set, increase gradually if airplane does not settle on altitude.

Altitude Gain:
ADJUST SENSITIVITY FIRST
Increase gradually if airplane levels off too soon.
Reduce gradually if airplane overshoots altitudes after climbs or descents.

Pull Rate:
ADJUST SENSITIVITY FIRST
Controls the rate the AP will pull when changing vertical speed.

VSI GAIN
ADJUST SENSITIVTY FIRST
ADJUST ALTITUDE GAIN FIRST
Increase gradually if overshooting climbs, decrease if rounding out too early.

G_error_gain
ADJUST SENSITIVITY FIRST
Increase if bumps are not smoothed enough; Decrease if ride is too harsh.

G_err_limit
ADJUST SENSITIVITY FIRST
Controls the max G the autopilot will push or pull to correct a bump (up to the G limiter).

 

[francfi]

Re: Software 5.4 for EFIS D10 A on an MCR01 Sporst

Thanks for your last message.
However I have a question.
When setting the Advanced Pitch Gain and Rate Adjustments on default values (for all 6 of them) , as per your AP Installation and Configuration Manual -page 6-20, does the 5.4 firmware operate as the former 5.3 with its only two parameters ( Torque and Sensitivity)?
If not which values the Advanced Pitch Gain and Rate Adjustments have to be on to run the AP bypassing/neutralizing these incremental 6 parameters.
I remind you that my aircraft has been flying pretty well ( meaning reaching targeted alt and holding it ) much better with 5.3 that with 5.4

The Pull Rate definition looks to me quite ambiguous .What happens physically when increasing or decreasing it ?

I will appreciate your assistance.

KR

Francfi ( from Paris-France)

 

[dynonsupport]

Re: Software 5.4 for EFIS D10 A on an MCR01 Sporst

The 5.4 pitch algorithm is completely new, and there isn't a set of new settings that equal any of the old settings. So unfortunately there isn't a way to go back to the exact way it was in 5.3. But, we'd expect you to be able to get better performance from 5.4 than 5.3, or at least "the same" if 5.3 was working for you.

When you increase pull rate it pulls on the stick quicker to change altitude.

 

[DBRV10]

Re: Software 5.4 for EFIS D10 A on an MCR01 Sporst

5.4 is way better........one small bug to fix

Here are some suggestions from my experience so far.


TORQUE: 100%
Adjust until servo does not slip during normal AP flight, and be comfortably overridden by the pilot.

SENSITIVITY: 22 or maybe more
ADJUST TORQUE FIRST
Increase gradually until airplane "twitches" when on altitude, then reduce by one
or two. If servo slips, increase torque.

PITCH GAIN:18-22
ADJUST SENSITIVITY FIRST
After sensitivity is set, increase gradually if airplane does not settle on altitude.

Altitude Gain: 18-22
ADJUST SENSITIVITY FIRST
Increase gradually if airplane levels off too soon.
Reduce gradually if airplane overshoots altitudes after climbs or descents.

Pull Rate:16 or so I think??
ADJUST SENSITIVITY FIRST
Controls the rate the AP will pull when changing vertical speed.

VSI GAIN 16-20 I think?
ADJUST SENSITIVTY FIRST
ADJUST ALTITUDE GAIN FIRST
Increase gradually if overshooting climbs, decrease if rounding out too early.

G_error_gain
ADJUST SENSITIVITY FIRST
Increase if bumps are not smoothed enough; Decrease if ride is too harsh.

G_err_limit
ADJUST SENSITIVITY FIRST
Controls the max G the autopilot will push or pull to correct a bump (up to the G limiter).


We set our VSI rate at 700FPM

The G limits I think are pretty much default, may have increased the G err limit a fraction, but take it easy with these.

You will have a better result in the end

 

[francfi]

Re: Software 5.4 for EFIS D10 A on an MCR01 Sporst

Att Dynon Support,

I am still uneasy with the 5.4 firmware which is not delivering the same good and simple control of pitch as the 5.3 did.

Having got back a 5.3 firmware from an other Dynon client ( by email) , I have tried so far unsuccessfully to upload my EFIS/AP with the 5.3 firmware ; the PC message indicated that the former 5.3 firmawre cannot paste the new 5.4 firmware ... Is that fair ?
I confirm again my intention to install the former 5.3 .Please help me to do it.
Thanks for your assistance .

KR
Francfi (from Paris france )

 

[dynonsupport]

Re: Software 5.4 for EFIS D10 A on an MCR01 Sporst

Unfortunately, unless you have a full backup (which the support program always recommends you make) of your 5.3 load, you can't revert (there are technical limitations here). Just having a copy of the 5.3 support program isn't enough, and you can't load a backup from another unit since it contains that unit's calibrations.

We do want to help you get your AP flying well, though. Can you send a data log of the AP flying to us at support@dynonavionics dot com, along with a list of your pitch settings (such as David provides above). For best results with the datalog, you want to clear it before you fly, have it set for 1 update a second, and pull the log after the flight that you're using to demo the behavior for us.

 

[DBRV10]

Re: Software 5.4 for EFIS D10 A on an MCR01 Sporst

OK, just to help or confuse you.

I have found the VS function to be performing well, but I never see it nail the VS exactly, but having said that its pretty good overall.

I have had some time to play with the AP in some bumpy conditions and found some combinations that seemed to work well. These are different to my post above so as you can see with the more sophisticated programming comes some extra tuning to get a really good result. I think this is likely to be my last adjustments......but hey I have to tinker with something on long sectors  

Torque 90%
Sens. 22
IAS min 83
IAS max 190
VSpD 700
Pitch Gain 30
Alt Gain 10
VSI Gain 25
Pull Rate 15
G Err Gain 11
G Err Lim 25


PS please note I have a torque enhancer on the pitch servo to increase torque but more significantly increase resolution of the encoder feedback so some of my numbers like the pitch and VSI gain may be a little higher to compensate....I am not sure, but please do not take this all as gospel!

 

[francfi]

Re: Software 5.4 for EFIS D10 A on an MCR01 Sporst

First thanks to D.Brown for his smart comments and suggestions

Attention Dynon Support

I have now the following settings with 5.4
Torque 85%
Sensitivity 25
Pitch Gain 30
Alt gain 13
VSI Gain 25
Pull Rate 16
G Ga 13
G lim 25
and the following results:
-sensitivy slightly twitchy, may need to be reduced by one or two
- overshoots after climb approx 10 to 20 ft
-levels off soon after descent approx 10 to 20 ft
-VSI Max set at 500ft/mn is overshooted by 200ft/mn when changing alltitude for climb and is followed approximately for descent
- however during an 180° turn with AP, aircraft will wallow up and down, before coming back to preset altitude after overshoot when banking is back to 0; 5.3 has kept maintaning the same altitude without any wallow

It is a small improvement from previous tests , but not yet the same behavior and comfort in tuning as former 5.3 which I regret so far

I am still uneasy with your definition of the Pull Rate;
Quote;
"When you increase pull rate it pulls on the stick quicker to change altitude"
Unquote
Does that mean than the course of the stick is increased at the same speed or that only the speed of stick is increased to run the same
course? Please advise

Thanks for your help.

Francfi (from Paris France)

 

[DBRV10]

Re: Software 5.4 for EFIS D10 A on an MCR01 Sporst

My thoughts.....and Brantel has more time on this than me, and a closer setup, butI would drop the sensitivity down a bit, say 20-22.

Maybe the wallowing is from slip, if the Yellow indication comes up its slip, so increase your torque to 100%.

you may find your altitude gain needs reducing, also the pull rate too.

the settling on +10-20 feet is a known software bug which I assume will get ironed out in the future.

DB

 

[dynonsupport]

Re: Software 5.4 for EFIS D10 A on an MCR01 Sporst

Francfi.

Pull rate is how aggressively the AP responds to a request for a change in vertical speed.  It is equivalent to how fast the stick is moved.  How far the stick is moved will depend on conditions.

If your servo is not slipping - it's very important that it not be or all these settings tests will be in error -  you should try increasing ALT gain, and perhaps G Gain a little bit.

 

[francfi]

Re: Software 5.4 for EFIS D10 A on an MCR01 Sporst

Att Dynon Support.

My Package
EFIS D10 S/N 5829
EDC D 10 A S/N 7643
AP 74 S/N 1280
2 servos SV 32 S/N 1564 & 1562

My latter in flight settings
-Torque 100%
Sensitivity 21
Pitch Gain 30
Alt Gain 18
VSI Gain 21
Pull Rate 13
G Error Ga 12
G Error Limit 25

Descents and Climbs :slight wandering before reaching preset alt
(with a remaining 10 to 20 ft overshoot)

180° Turn; significant up and down wandering ( 2 to 3 oscillations) and final overshoot of 10 to 2 ft, with VSI max overshooted also .

So far, I must conclude that the 5.4 is not as easy, good and reliable as the former 5.3 .

I am unhappy to say that as I still rely upon DA fine engineering capabilties , which are , from my end ,to be fully demonstrated ...
I regret the 5.3 A LOT..

Thanks for your assistance.

Francfi ( from Paris France)

 

[dynonsupport]

Re: Software 5.4 for EFIS D10 A on an MCR01 Sporst

Since you've emailed us, we'll take the support conversation to that channel to help get you sorted out.

 

[francfi]

Re: Software 5.4 for EFIS D10 A on an MCR01 Sporst

OK, thanks for your reply, I am waiting for the follow up of my demand by direct email.

KR
Francfi