Instrument approach and autopilot weirdness

[Joe Larussa]

Ok so Stec 60-2 autopilot, 530 non Wass , King Hsi. I notice while being vectored to final on the ILS with autopilot engaged to intercept the ILS it blows through final then overcompensates to get back on course and weaves back and forth before settling down. I'm thinking a few things. I've been flying at 120 mph before it turns final. Too fast? On some approaches its about a 70 degree turn to intercept final. Being it's an ILS there is no Garmin style anticipate turn to final. I'm thinking the autopilot can't keep up with the closure rate. At a 70 degree heading change that should be roughly 24 seconds. Thoughts? Does this even make sense?

[Marauder]

On my YouTube channel one of my videos shows an LPV approach with close to a 90° intercept angle with my STEC 60-2. It flies through the course about 1 dot before recovering. What you are seeing is normal. Most intercept angles that I have been given are 30° to 45° and there are no issues with those angles.

Sent from my iPad using Tapatalk

[Joe Larussa]

8 minutes ago, Marauder said:

On my YouTube channel one of my videos shows an LPV approach with close to a 90° intercept angle with my STEC 60-2. It flies through the course about 1 dot before recovering. What you are seeing is normal. Most intercept angles that I have been given are 30° to 45° and there are no issues with those angles.

Sent from my iPad using Tapatalk

Just thinking though on a LPV approach the GPS anticipates the turn to final or am I missing something? What's your YouTube channel?

[Marauder]

8 minutes ago, Marauder said: On my YouTube channel one of my videos shows an LPV approach with close to a 90° intercept angle with my STEC 60-2. It flies through the course about 1 dot before recovering. What you are seeing is normal. Most intercept angles that I have been given are 30° to 45° and there are no issues with those angles.

Sent from my iPad using Tapatalk

Just thinking though on a LPV approach the GPS anticipates the turn to final or am I missing something? What's your YouTube channel?

It can anticipate a turn but factor in strong, variable winds and it is just like us, guessing at what heading. Look at the last couple of minutes of this video:

[Yooper Rocketman]

8 hours ago, Joe Larussa said:

Ok so Stec 60-2 autopilot, 530 non Wass , King Hsi. I notice while being vectored to final on the ILS with autopilot engaged to intercept the ILS it blows through final then overcompensates to get back on course and weaves back and forth before settling down. I'm thinking a few things. I've been flying at 120 mph before it turns final. Too fast? On some approaches its about a 70 degree turn to intercept final. Being it's an ILS there is no Garmin style anticipate turn to final. I'm thinking the autopilot can't keep up with the closure rate. At a 70 degree heading change that should be roughly 24 seconds. Thoughts? Does this even make sense?

I assume you are talking a full published report, not vectors?  ATC is not supposed to give you more than a 30 degree turn on your last vector to final.

Tom

[Marauder]

8 hours ago, Joe Larussa said: Ok so Stec 60-2 autopilot, 530 non Wass , King Hsi. I notice while being vectored to final on the ILS with autopilot engaged to intercept the ILS it blows through final then overcompensates to get back on course and weaves back and forth before settling down. I'm thinking a few things. I've been flying at 120 mph before it turns final. Too fast? On some approaches its about a 70 degree turn to intercept final. Being it's an ILS there is no Garmin style anticipate turn to final. I'm thinking the autopilot can't keep up with the closure rate. At a 70 degree heading change that should be roughly 24 seconds. Thoughts? Does this even make sense?

I assume you are talking a full published report, not vectors?  ATC is not supposed to give you more than a 30 degree turn on your last vector to final.

Tom

I'm thinking, like my video, he is playing around with the intercept angles. Even on an LPV approach I saw the autopilot fly through the final course before correcting on a ~90° intercept with the winds playing a role. I'm just amazed how precise these autopilots can be.

[Joe Larussa]

58 minutes ago, Yooper Rocketman said:

I assume you are talking a full published report, not vectors?  ATC is not supposed to give you more than a 30 degree turn on your last vector to final.

Tom

Well yes and no. Your right the published approach gives me a 77 degree intercept. With practice approaches I'm getting 90 degree intercepts all the time. I'm really trying hard getting used to this new to me system. I'm beating it to death wondering if it's the equipment , me or both. Today I'm going to try slowing my intercept speed to maybe 100 mph and see if that helps.

[Marauder]

58 minutes ago, Yooper Rocketman said: I assume you are talking a full published report, not vectors?  ATC is not supposed to give you more than a 30 degree turn on your last vector to final.

Tom

Well yes and no. Your right the published approach gives me a 77 degree intercept. With practice approaches I'm getting 90 degree intercepts all the time. I'm really trying hard getting used to this new to me system. I'm beating it to death wondering if it's the equipment , me or both. Today I'm going to try slowing my intercept speed to maybe 100 mph and see if that helps.

I don't think it is the equipment. I've been flying approaches with my STEC 60-2 since 1998. It is a great autopilot and considering it is an ILS, the angle and the speed you are trying to enter at, it may fly through the final approach course. It is a rate based autopilot that will do a max turn at standard rate.

Also, if you are in soft capture mode, it will not be aggressive in capturing the final approach course. To verify things, you can hand fly the approach being exact on the assigned heading and have you try to manually intercept without going past standard rate. That is what the autopilot is trying to do. A GPS approach will take into consideration other factors available to it and pump those to the autopilot. Even then, like in my video, changes in wind direction or speed can influence the accuracy of the intercept.

[Danb]

2 hours ago, Yooper Rocketman said:

I assume you are talking a full published report, not vectors?  ATC is not supposed to give you more than a 30 degree turn on your last vector to final.

Tom

Tom unfortunately that always doesn't always happen, esp. at Kilg.  Ive been given a sharp turn a few times, I guess its actually Philly approach.

[Yooper Rocketman]

28 minutes ago, Danb said:

Tom unfortunately that always doesn't always happen, esp. at Kilg.  Ive been given a sharp turn a few times, I guess its actually Philly approach.

I have a young man I mentored to getting his PPL and IFR before graduating from high school who went on to ATC school and graduated.  He said that is a requirement, no more than 30 degree intercept to the final approach course.  We may have an actual controller on the board who can clarify this?  I've never been given more without a query from ATC if I could manage it.  I agree with Chris, these autopilots typically will not take an aggressive turn to intercept.  Add wind and other factors and I've had to take over or shift to "Heading Mode" until the plane would intercept.  Most times once I stolen the plane from the autopilot, I am not confident in it enough to turn it back over during the approach and I hand fly it to the runway.

I've had my experiences with Philly Approach.  I won't get into that now though.

Tom

[Danb]

53 minutes ago, Yooper Rocketman said:

 

I've had my experiences with Philly Approach.  I won't get into that now though.

Tom

Tom

Couldn't agree any more.

[carusoam]

Part of the system documents, like user manuals, may give some insight to how well your system can work.

The computer's calculation speed vs. intercept angle, air speed and response characteristics are interesting with the various APs.

One thing that drives the response rate is the limitation the AP has for maximum bank angle.  

For comparison...

1) Cross a river turn 90° using a steep bank angle.  Note how close to the river you are when done.

2) Cross the river turn 90° using the AP in heading mode, adjust the heading.  What bank angle does it use?  Probably turns at a standard rate using nearly a 15° angle chewing up the whole East and Hudson Rivers...

The same comparison can be done using different air speeds.

Being too close to the ILS at the airport can give real analog challenges.  If following the approach as depicted, the analog issues shouldn't be much of a challenge.

The things that improve the machine's response are not necessarily safety wise.  Slowing down and increasing bank angle to capture the ILS has it's limitations.

I am only a PP, my IFR skills are currently rusty....

Best regards,

-a-

[Steve Dietrich]

With the KING HSI   flying intercepts like you describe on occasion I would just switch the mode to heading, arm the approach and then get the airplane turning towards the final with a twist of the heading bug as the course came alive

I think part of the "sluggishness" is that the system wants to know that this is a real ,  staibilzed localizer before it turns the airplane.  .  It's something the pilot should be monitoring and a quick shift to CWS mode on a close in 90 degree intercept  and then reengagement seems to make sense. 

Edited April 4, 2016 by Steve Dietrich

[Joe Larussa]

I shot a series of Rnav approaches yesterday. With the autopilot that is the business! More accurate and a lot less button pushing. So much so I had my 530 sent out today for the Wass upgrade. LPV here I come!!