Autopilot question

[jrwilson]

I have a kfc 150 autopilot/garmin430/Aspen PFD.  When I’m in nav mode,  with and without gpss steering mode active on the Aspen, the autopilot always keeps the plane about 1.5 dots left of course on the cdi.  Is that a feature?  Is it a setting I can change?  It centers in approach mode.0
 

any ideas?  Thanks!

[Mooney Dog]

I noticed this on our club's 172. It also has a aspen pfd/kap 150 but has a 750/650 combo instead. I think its something to do with the aspen units personally, but i cant say for sure. 

[jlunseth]

Well, I would check the Aspen manual and see what it says about using the NAV mode of the autopilot with the Aspen. Perhaps the Aspen operates differently from the gpss + analog AP systems I am familiar with. So there are as I understand it, two ways these systems operate. With the autopilot in approach mode, the system is wired so that the AP uses the RWY "X" waypoint in the same way it would use an analog ILS signal. It uses the waypoint for guidance straight in to the waypoint. You set the course needle for the inbound course and it flies that course just as it would an ILS or a VOR approach - by flying straight along the course line to the endpoint. The AP itself is computing the course using the waypoint.

The roll steering works differently. What I am familiar with is to put the AP in HDG mode and enable the roll steering. The gpss then gives constant guidance to the analog AP through the HDG bug, to essentially fool the AP to fly the course guidance provided by the Garmin and the gpss. That guidance does not need to be straight line guidance, and that is the beauty of roll steering. The roll steering + Garmin can, for example, anticipate interception of the final approach course and begin a turn to intercept rather than flying to the final approach course and then making the turn, which results in going through the course and correcting back. The roll steering + Garmin can also fly approach courses that are curved, and there are a few, particularly up in AK.

The NAV mode, as I understand it, works in the same way as putting the AP in approach mode except with greater tolerance for course deviation, that is, it accepts a waypoint from the Garmin and then flies straight to that waypoint along the course you have dialed in with the course needle, as though flying to a VOR that you have tuned in and set a course to. The AP is computing the course and not using course information from the roll steering + Garmin (except for the endpoint).

That is how my system works, and a couple of others I have flown with, but I have not flown with an Aspen, maybe it does something different.

Definitely not an avionics guy, but my understanding is that the wiring is such that the signal bypasses roll steering and goes directly to the AP from the Garmin, if the AP is in APCH or NAV mode, and goes through the roll steering function only if the AP is in HDG mode and roll steering is enabled.

I could be completely wrong, I have not flown with an Aspen, but I would look it up.

Another explanation which I have run into is that when the gyro in the 256 begins to fail, it does that. When mine went it kept me persistently right of course at cruise but worked fine on approach. Eventually it began to fail on approach also, which is when I got it replaced and thereafter, no problem.

 

[PT20J]

It's not right. My similar setup doesn't do that. Perhaps @Jake@BevanAviation knows the answer.

[Jake@BevanAviation]

With most legacy A/P system GPSS should be used in heading mode.  If you are doing VOR, ILS or GPS navigation it should be in a navigation or approach mode, depends on what you are trying to do.  The KFC 150 system was built around intercept angles before the thought of GPS.  There is a circuit called LBT (lateral beam tracking) that uses a intercept angle to trigger the mode.  Without LBT mode the system will not properly track a navigation signal.  This is most noticeable on a GPS approach with no intercept to final.  The KFC 150 POH has this depicted as to how to utilize the ARM mode of NAV and APPR modes.  If you are doing the proper sequencing the LBT mode will always trigger.

There are some other things that come into play like how does the heading track (hdg bug 180 and tracking 180)?  Can it hold wings level with perfectly level wings? If it can't fly wings level or stay true to the head error it will have problems in a navigation mode.  The exact same circuit is used for nav or approach modes of operation just different gains.  So if it flies the approach mode fine it should do the same for nav pending the LBT mode is active.  The computer is pretty dumb in a sense, all it knows is what signal you are feeding it.  VOR, ILS and GPS are all treated as a lateral deviation signal,  HDG and GPSS are a heading error.  Of course if you have a issue with the AI source that will throw everything off as this is the back bone of the system.

Edited August 11, 2021 by Jake@BevanAviation

[N201MKTurbo]

An analog autopilot sums the heading error with the nav error. If your nav is tracking 2 dots left you have a heading error. set the heading bug a little right of course and it should center.

[jrwilson]

39 minutes ago, N201MKTurbo said:

An analog autopilot sums the heading error with the nav error. If your nav is tracking 2 dots left you have a heading error. set the heading bug a little right of course and it should center.

Ok I’m confused . In nav mode I’m not using the heading bug at all.  The bug could be 180deg off and it doesn’t seem to effect the autopilot at all, in nav mode.  Unless I’m just doing something totally weird, which is possible…

[jrwilson]

52 minutes ago, Jake@BevanAviation said:

With most legacy A/P system GPSS should be used in heading mode.  If you are doing VOR, ILS or GPS navigation it should be in a navigation or approach mode, depends on what you are trying to do.  The KFC 200 system was built around intercept angles before the thought of GPS.  There is a circuit called LBT (lateral beam tracking) that uses a intercept angle to trigger the mode.  Without LBT mode the system will not properly track a navigation signal.  This is most noticeable on a GPS approach with no intercept to final.  The KFC 200 POH has this depicted as to how to utilize the ARM mode of NAV and APPR modes.  If you are doing the proper sequencing the LBT mode will always trigger.

There are some other things that come into play like how does the heading track (hdg bug 180 and tracking 180)?  Can it hold wings level with perfectly level wings? If it can't fly wings level or stay true to the head error it will have problems in a navigation mode.  The exact same circuit is used for nav or approach modes of operation just different gains.  So if it flies the approach mode fine it should do the same for nav pending the LBT mode is active.  The computer is pretty dumb in a sense, all it knows is what signal you are feeding it.  VOR, ILS and GPS are all treated as a lateral deviation signal,  HDG and GPSS are a heading error.  Of course if you have a issue with the AI source that will throw everything off as this is the back bone of the system.

It flies a heading fine in heading mode and flies level.  Can I set or adjust the LBT mode?

[Jake@BevanAviation]

In nav mode or appr mode it is a combination of the lateral deviation and the course pointer error.  You can move the heading bug anywhere in a coupled nav or appr mode and the system should not respond.  If you trigger LBT in nav or appr there is a course washout circuit that will start, after about 2 minutes it will wash out the course error and primarily look at lateral deviation error.  This allows for cross wind compensation.  There is a 30 deg dead band under LBT that if the course error is less than 30 deg it will get washed out over time.  Moving the course point will provide a correction that direction but if the error is less than 30 deg with LBT active it will get washed out.

There is a adjustment pot inside the computer for nav null but if it flies the approach fine I would not suggest adjusting it.  I do this on the bench with a nulled AI, nulled lateral and nulled course error.  Allow LBT to activate and wait 2 minutes if the computer is not centered I will adjust the nav null pot.  I will only do this if I know the computer has a problem with lateral deviation track for nav and appr modes.

Edited August 11, 2021 by Jake@BevanAviation

[N201MKTurbo]

When you say it washes out the error, you are referring to an integrator. It accumulates error over time, trying to reduce the error. It is the proportional error times the integral gain over time. This converges on the set point, which in this situation is the nav error signal. This lends some credence to adjusting nav error offset.

[Bob Weber]

21 hours ago, jrwilson said:

I have a kfc 150 autopilot/garmin430/Aspen PFD.  When I’m in nav mode,  with and without gpss steering mode active on the Aspen, the autopilot always keeps the plane about 1.5 dots left of course on the cdi.  Is that a feature?  Is it a setting I can change?  It centers in approach mode.0
 

any ideas?  Thanks!

JR I can help you with this with a quick call. 616 822 1999

Jake is correct that a centering adjust will likely correct your issue.

Has if ever flown a centered NAV course with the Aspen installed?

 

[Jake@BevanAviation]

I apologize, I referenced the wrong system in my initial reply to the OP.  The KFC 150 still uses LBT but the circuit is vastly different from the KFC 200 system.  The KFC 150 uses microprocessors, digital converters and multiplexers to generate a composite roll command signal.  This composite signal is what is used to drive the servo amplifiers.  There is a nav deviation null pot inside the computer used to adjust a offsite nav signal before it goes to the multiplexer and summed with the course error.  There are two additional pots located on the front face of the unit labeled RN and RC.  RN for roll null, used to null the AI roll error, RC is roll centering that can be adjusted for track discrepancies .  A word of caution, be very careful when adjusting the RN or RC pots as you can damage the pot.  Use a long small flat blade jewelers screwdriver to adjust the pot.  Make small adjustments and wait for the computer to settle before adjusting again.

[Rick7576]

Its all Greek to me   ;-}