Flight Data Analysis

[Nippernaper]

I have a G3X touch in my J, and I've been wondering for a while if there might be more to learn from the data recorded than what Savvy analyzes.  Savvy is great, but they tend to focus on engine related parameters, and a lot of the new PFD and MFD equipment record a whole lot more than EGT and CHT.  With altitude, lat and long, GPS speed, etc., I figured there might be a way to review typical take off and landing performance, as opposed to the POH numbers.  Take off data was more straightforward to extract.  Landing data not so much, as the ground roll tends to be related to where the turn-offs are.  Hopefully I figure out how to attach a few graphs below.  The most meaningful I found was distance to a 50' altitude over field altitude versus Density Altitude on take off, and distance from 50' altitude to touch down on landing, for two weight conditions:  just me with typical baggage, and fully loaded (within 50 lbs of gross weight).  Taking mean and standard deviations, I then set a minimum field length based on the mean plus 2.5 times the standard deviation.  This results in a conservative estimate of performance, based on how I fly my plane, at the loadings I typically have.

 

[MikeOH]

Is the 50' altitude data calculated from GPS?  If so, I'd be interested in the mean and std. dev. of that data. I assume NOT WAAS, but either way what is the altitude error?

[N201MKTurbo]

It seems like wind would have a big effect on this data. Does your system record wind?

[Nippernaper]

1 hour ago, MikeOH said:

Is the 50' altitude data calculated from GPS?  If so, I'd be interested in the mean and std. dev. of that data. I assume NOT WAAS, but either way what is the altitude error?

Altitude is indicated, not GPS (from the air data computer).  Mean of take off distance was 2075', standard deviation was 224' at max gross.

[Nippernaper]

17 minutes ago, N201MKTurbo said:

It seems like wind would have a big effect on this data. Does your system record wind?

Good point.  I figured wind based on GPS speed versus TAS at 50' altitude for the landing data.  See graph below.  I did not do that for take off (unfortunately).  If I get around to reprocessing, I'll add that variable to the calculations.  It general, landing distance decreases, but there is still significant scatter.

[N201MKTurbo]

Once corrected for wind, the scatter may be technique. So many things to perfect in aviation. Tightening up these scatter plots is one we don’t routinely concern ourselves with.

[MikeOH]

1 hour ago, Nippernaper said:

Altitude is indicated, not GPS (from the air data computer).  Mean of take off distance was 2075', standard deviation was 224' at max gross.

How did you correlate crossing 50 feet (altimeter indicated) with distance traveled?

[Nippernaper]

15 hours ago, MikeOH said:

How did you correlate crossing 50 feet (altimeter indicated) with distance traveled?

Identified time when altitude reached 50 feet above that during the takeoff run, calculated distance based on latitude/longitude values from the GPS data.  

[laytonl]

In my experience, wind can be a significant factor.  Especially in downwind takeoffs.  Lee

[philiplane]

the data junkies on the Cirrus forum are currently obsessed with this type of data analysis, in an effort to reduce the damage done to nose landing gear and fairings.

They're trying to pull AHRS data that shows the pitch upon landing, and then teach pilots to maintain a higher pitch attitude if theirs is normally flat.

But what they really need, is a few sessions with a decent CFI to fix their landings. They already have a great attitude reference system for landing, it's called the windshield.

Don't get wrapped up in the flight data. Get wrapped up in flight.

[kortopates]

the data junkies on the Cirrus forum are currently obsessed with this type of data analysis, in an effort to reduce the damage done to nose landing gear and fairings.
They're trying to pull AHRS data that shows the pitch upon landing, and then teach pilots to maintain a higher pitch attitude if theirs is normally flat.
But what they really need, is a few sessions with a decent CFI to fix their landings. They already have a great attitude reference system for landing, it's called the windshield.
Don't get wrapped up in the flight data. Get wrapped up in flight.

As an instructor, couldn’t agree more!! It’s as simple as holding the cowling just below the end of the runway while looking out the windshield and hold off the nose wheel as long as possible.


Sent from my iPhone using Tapatalk

[Nippernaper]

On 10/16/2023 at 8:21 PM, kortopates said:

As an instructor, couldn’t agree more!! It’s as simple as holding the cowling just below the end of the runway while looking out the windshield and hold off the nose wheel as long as possible.


Sent from my iPhone using Tapatalk

What I've been trying to do with the data is not use it to adjust my flying technique (although it could definitely use some improvement), but rather to better understand what performance I'm getting, and make sure I stay well within that performance envelope.  For example, regardless of what the POH says, when I'm fully loaded at a reasonable density altitude, based on the data I shouldn't be using a runway any shorter than 2500 feet.  I'll continue to work on my skills and technique with my instructor, but I want to make sure I'm not pushing the envelope.

[kortopates]

What I've been trying to do with the data is not use it to adjust my flying technique (although it could definitely use some improvement), but rather to better understand what performance I'm getting, and make sure I stay well within that performance envelope.  For example, regardless of what the POH says, when I'm fully loaded at a reasonable density altitude, based on the data I shouldn't be using a runway any shorter than 2500 feet.  I'll continue to work on my skills and technique with my instructor, but I want to make sure I'm not pushing the envelope.

Maybe you could explain better by what you mean for “fully loaded” and reasonable DA” for 2500’ and how you arrive at the 2500 min.
I am trying to follow how your analysis supports or contradicts the POH.


Sent from my iPhone using Tapatalk

[Nippernaper]

On 10/19/2023 at 10:59 PM, kortopates said:

Maybe you could explain better by what you mean for “fully loaded” and reasonable DA” for 2500’ and how you arrive at the 2500 min.
I am trying to follow how your analysis supports or contradicts the POH.


Sent from my iPhone using Tapatalk

I will do my best.  For all of the flights in the past year where I had all 4 seats loaded and some baggage, and with tanks at 50 gallons (this was not a small number of flights, maybe 50 or so), from the recorded data I extracted the distance from start of take off roll to 50'.  For my home airport I calculated the mean and standard deviation of the distances.  Taking the mean and adding 2.5 standard deviations results in a distance in terms of take off over a 50' obstacle that should be exceeded only about 6 times every 1000 takeoffs.  This yields the 2500' number.  So, based on the weather conditions typical at my home field, the field elevation, the weights I usually fly the aircraft, and how I fly the aircraft, 2500' would seem to be a relatively conservative number for the distance to a 50' obstacle.  I of course also check the POH for the take off performance numbers, but they are always less than that.  I found it very interesting to see how I was actually performing in terms of take offs, versus what the POH was predicting for a given DA.  We're often advised to allow a healthy margin over the numbers in the POH since we're not trained test pilots flying brand new aircraft, and now I have a basis for determining what a safe margin is, based on how I fly my airplane.

[kortopates]

Okay, that proves your getting pretty repeatable performance based on exactly as you say, how you fly your airplane. And 2500' to 50' is also very reasonable agreement with the POH depending on the specific. Earlier you said you needed a 2500' runway.

But if your getting this performance without following the POH numbers this doesn't mean we can't get better performance when we need to it to operate out of a shorter runway.

For example, when I don't have obstacles to be concerned with I am all about climbing out at Vy+10-20 kts to keep CHTs cool and give a good view out of the cockpit. But when I need to climb at steeper rate, such as departing from an airport with really no nearby off-field leanding spots or even more rarely, short with obstacles off the end. Then I'll follow the POH climb profile to get max performance. Doing so should get a J off the ground in under 1200', but if I was doing far worse, I would draw a different conclusion.

First though, to get POH performace, you have to fly the profile in the POH. You didn't provide specific conditions but assuming a near fully loaded 2740 lb J at a lower density altitude of around 2000' it should lift off by 1200' and climb to 50' AGL in under 2500'. That's no wind at max gross weight. and rotating at about 58 kias and climbung out steeply at under 70 kias which will require about 12 degrees pitch. Per the POH this should get us to 50' by 2500'  But this doesn't call for a minimum runway length of 2500', that's 2500 to get to 50' and and a standard IFR departure only requires we clear the departure end of the runway by 35' climbing at 200 FPNM to ensure standard obstacle clearance.

But if I am doing much worse than POH performance, while using the POH numbers, I would be checking health of the engine since something is off such as possibly worn down cam preventing the cylinders from making full power. 

[Nippernaper]

On 10/23/2023 at 5:17 PM, kortopates said:

Okay, that proves your getting pretty repeatable performance based on exactly as you say, how you fly your airplane. And 2500' to 50' is also very reasonable agreement with the POH depending on the specific. Earlier you said you needed a 2500' runway.

But if your getting this performance without following the POH numbers this doesn't mean we can't get better performance when we need to it to operate out of a shorter runway.

For example, when I don't have obstacles to be concerned with I am all about climbing out at Vy+10-20 kts to keep CHTs cool and give a good view out of the cockpit. But when I need to climb at steeper rate, such as departing from an airport with really no nearby off-field leanding spots or even more rarely, short with obstacles off the end. Then I'll follow the POH climb profile to get max performance. Doing so should get a J off the ground in under 1200', but if I was doing far worse, I would draw a different conclusion.

First though, to get POH performace, you have to fly the profile in the POH. You didn't provide specific conditions but assuming a near fully loaded 2740 lb J at a lower density altitude of around 2000' it should lift off by 1200' and climb to 50' AGL in under 2500'. That's no wind at max gross weight. and rotating at about 58 kias and climbung out steeply at under 70 kias which will require about 12 degrees pitch. Per the POH this should get us to 50' by 2500'  But this doesn't call for a minimum runway length of 2500', that's 2500 to get to 50' and and a standard IFR departure only requires we clear the departure end of the runway by 35' climbing at 200 FPNM to ensure standard obstacle clearance.

But if I am doing much worse than POH performance, while using the POH numbers, I would be checking health of the engine since something is off such as possibly worn down cam preventing the cylinders from making full power. 

Ah, I understand where you're coming from now.  Agreed that what I'm looking at is a normal, "lazy" take-off without any attempt at max performance.  Also, I see your point that 2500' is not necessarily the same as runway length (shorter may be OK, since lift off will be much sooner).  I was just interested to see what kind of performance I'm actually getting, and set my expectations, and operating procedures, accordingly.  I'll still be practicing short and soft fields to keep my skills up, but I try and avoid the need to use them.