My club just got a very nicely equipped Bonanza with a King KFC-200
A/P system installed (along with a CNX-80 GPS). I think I've got most
of it figured out, but yesterday something happend which I didn't
understand.

The active leg in the GPS was a course of about 230, with a stiff wind
from the south. I was about 2.5 miles right of course, and 30 miles
from the next waypoint. I put the course pointer on 230, the heading
bug on 190, and selected FD, HDG, ALT, and AP-ON, and watched the
cross-track error on the GPS; it was decreasing slowly, so I figured I
had a nice shallow intercept angle set up. I then hit the NAV
button. The NAV and ARM lights came on, as expected.

Eventually, Otto intercepted the desired course (as indicated by the
GPS XTE going to zero and the HSI course deviation bar centering.
What I expected to happen at this point was that the HDG and ARM
lights would go out and the CPLD light would come on, and Otto would
start tracking the course. Instead, it just stayed in heading mode
with ARM lit up, slowly taking us left of course. When I recycled the
NAV button, it instantly went into coupled mode and started tracking.

Is my understanding of how ARM/CPLD works in error? Is it possible
that the very shallow intercept angle I had set up (just a couple of
degrees) had somehow fooled the coupling trigger circuitry?

Have you read the POH supplement?



"Roy Smith" > wrote in message
...
| My club just got a very nicely equipped Bonanza with a
King KFC-200
| A/P system installed (along with a CNX-80 GPS). I think
I've got most
| of it figured out, but yesterday something happend which I
didn't
| understand.
|
| The active leg in the GPS was a course of about 230, with
a stiff wind
| from the south. I was about 2.5 miles right of course,
and 30 miles
| from the next waypoint. I put the course pointer on 230,
the heading
| bug on 190, and selected FD, HDG, ALT, and AP-ON, and
watched the
| cross-track error on the GPS; it was decreasing slowly, so
I figured I
| had a nice shallow intercept angle set up. I then hit the
NAV
| button. The NAV and ARM lights came on, as expected.
|
| Eventually, Otto intercepted the desired course (as
indicated by the
| GPS XTE going to zero and the HSI course deviation bar
centering.
| What I expected to happen at this point was that the HDG
and ARM
| lights would go out and the CPLD light would come on, and
Otto would
| start tracking the course. Instead, it just stayed in
heading mode
| with ARM lit up, slowly taking us left of course. When I
recycled the
| NAV button, it instantly went into coupled mode and
started tracking.
|
| Is my understanding of how ARM/CPLD works in error? Is it
possible
| that the very shallow intercept angle I had set up (just a
couple of
| degrees) had somehow fooled the coupling trigger
circuitry?
|
|

"Jim Macklin" > wrote:

> Have you read the POH supplement?

Yes. Do you think I got this far by just randomly pushing buttons? :-)

Roy Smith wrote:
> Is my understanding of how ARM/CPLD works in error?

If it is, then so is mine - and I've used several different flavors of
KFC autopilots.

> Is it possible
> that the very shallow intercept angle I had set up (just a couple of
> degrees) had somehow fooled the coupling trigger circuitry?

Of course it's possible. Neither one of us knows how that circuitry
works, but I can certainly imagine some designs that would be fooled
that way. However, since King keeps those service manuals expensive,
we have no way to find out.

My question is - how far did you let the XTE get in the other direction
before you recycled the NAV button? You might simply not have waited
long enough. All it would take is one comparator out of whack.

Michael

> > Have you read the POH supplement?

> Yes. Do you think I got this far by just randomly pushing buttons? :-)

That's usually how I do it.

Roy,

You need to always be sure to give the aircraft a heading
that will actually provide an intercept else it will never go
from ARM to CPLD. Sounds like there is a small offset
in that detector given that you got a couple soon as you
cycled NAV. It is very rate sensitive... it will go to
CPLD when the needle is not nearly centered if the needle is moving
fast
toward the center. If the needle is moving slowly, it will have
to be centered or even a bit past center. The reason is that
the crosswind integrator is only sort of shorted out during
this ARM- capture phase of flight.

When you go to CPLD, the yellow arrow
becomes the course datum.

Note that you can even intercept from FD-APR-ARM mode--wing
leveling only. Heading does not have to be selected to effect the
coupling--but the airplane needs to be aimed so as to
acheive an intercept.

The tip of the day is to always use APR mode when coupling to
your GPS, even enroute. NAV mode is dumbed down
considerably to make crossing over VOR stations less
traumatic. APR doesn't suffer that, and since the GPS
course needle is rock solid APR works great with sharp,
precise coupling. You can fly a T shaped GPS approach
without ever decoupling it-- just feed the new headings in as
cued by the cnx-80.

There's a lot to learn about the KFC-200. Maybe the POH smart-
asses know what it does when the HSI is failed:
answer is that it will either cancel everything or just ignore
the heading bug, depending on how it's wired. Most cancel
everything if heading reference is lost and any mode requiring
heading is selected. But you can turn
FD and ALT modes back on and it will wing-level at a constant
altitude which can be very useful if you are trying to figure
out what to do after a gyro failure. Explore this by pulling
out the HSI breaker.

Have you completely figured out the pitch modes?

Bill Hale BPPP instructor.




Roy Smith wrote:
> My club just got a very nicely equipped Bonanza with a King KFC-200
> A/P system installed (along with a CNX-80 GPS). I think I've got most
> of it figured out, but yesterday something happend which I didn't
> understand.
>
> The active leg in the GPS was a course of about 230, with a stiff wind
> from the south. I was about 2.5 miles right of course, and 30 miles
> from the next waypoint. I put the course pointer on 230, the heading
> bug on 190, and selected FD, HDG, ALT, and AP-ON, and watched the
> cross-track error on the GPS; it was decreasing slowly, so I figured I
> had a nice shallow intercept angle set up. I then hit the NAV
> button. The NAV and ARM lights came on, as expected.
>
> Eventually, Otto intercepted the desired course (as indicated by the
> GPS XTE going to zero and the HSI course deviation bar centering.
> What I expected to happen at this point was that the HDG and ARM
> lights would go out and the CPLD light would come on, and Otto would
> start tracking the course. Instead, it just stayed in heading mode
> with ARM lit up, slowly taking us left of course. When I recycled the
> NAV button, it instantly went into coupled mode and started tracking.
>
> Is my understanding of how ARM/CPLD works in error? Is it possible
> that the very shallow intercept angle I had set up (just a couple of
> degrees) had somehow fooled the coupling trigger circuitry?

You never know. But in order to capture, the system must be
armed with the CDI a full scale and the intercept angle/rate
must be such that a clear intercept happens. This generally
requires a 30 degree angle. This is from memory, I don't
have the manuals at this time for review.



"Roy Smith" > wrote in message
...
| "Jim Macklin" >
wrote:
|
| > Have you read the POH supplement?
|
| Yes. Do you think I got this far by just randomly pushing
buttons? :-)

"Michael" > wrote:
> My question is - how far did you let the XTE get in the other direction
> before you recycled the NAV button? You might simply not have waited
> long enough. All it would take is one comparator out of whack.

That's certainly possible. I didn't let it get very far. Enough that it
was obvious to me that it had gone past center, but less than one dot out
on the HSI. I don't recall what the GPS XTE was, but I'd don't think the
A/P is aware of the XTE directly, just the HSI deviation.

I love playing with this stuff. It makes me feel like I'm back in the
1980's playing with analog computers in the control systems lab :-)

"Bill" > wrote:
> If the needle is moving slowly, it will have
> to be centered or even a bit past center.

Yeah, it sounds like that's exactly what happened. I'll have to set up the
same scenario again the next time I'm in the plane and let it go a bit
further to see what happens.

> Note that you can even intercept from FD-APR-ARM mode--wing
> leveling only. Heading does not have to be selected to effect the
> coupling--but the airplane needs to be aimed so as to
> acheive an intercept.

That makes sense. The arm/couple circuitry is looking at the deviation
bar. It doesn't know if the bar is moving because you pointed the plane in
the right direction or because the A/P driven by the heading bug did the
same thing. Is suspect you could even get it to go from ARM to CPLD by
moving the course pointer so as to make the deviation bar center.

> The tip of the day is to always use APR mode when coupling to
> your GPS, even enroute. NAV mode is dumbed down
> considerably to make crossing over VOR stations less
> traumatic. APR doesn't suffer that, and since the GPS
> course needle is rock solid APR works great with sharp,
> precise coupling. You can fly a T shaped GPS approach
> without ever decoupling it-- just feed the new headings in as
> cued by the cnx-80.

Cool, I'll have to try that. I assume you mean "feed it new courses",
though, right?

> There's a lot to learn about the KFC-200. Maybe the POH smart-
> asses know what it does when the HSI is failed:
> answer is that it will either cancel everything or just ignore
> the heading bug, depending on how it's wired. Most cancel
> everything if heading reference is lost and any mode requiring
> heading is selected. But you can turn
> FD and ALT modes back on and it will wing-level at a constant
> altitude which can be very useful if you are trying to figure
> out what to do after a gyro failure. Explore this by pulling
> out the HSI breaker.

I will have to do that too.

> Have you completely figured out the pitch modes?

You mean the one that suddenly puts the airplane into a 3000 FPM dive?
Yes, I've got that one figured out :-)

But, seriously, I understand altitude hold, and the UP/DOWN rocker switch
doing the 600 FPM climb/descent thing as long as you hold it. I'm still
working on getting the knack the of using CWS to change pitch, and getting
it to fly a coupled ILS, but I think one more flight and I'll have that
nailed.

Roy Smith wrote:
> "Bill" > wrote:
> > If the needle is moving slowly, it will have
> > to be centered or even a bit past center.
>
> Yeah, it sounds like that's exactly what happened. I'll have to set up the
> same scenario again the next time I'm in the plane and let it go a bit
> further to see what happens.
>
> > Note that you can even intercept from FD-APR-ARM mode--wing
> > leveling only. Heading does not have to be selected to effect the
> > coupling--but the airplane needs to be aimed so as to
> > acheive an intercept.
>
> That makes sense. The arm/couple circuitry is looking at the deviation
> bar. It doesn't know if the bar is moving because you pointed the plane in
> the right direction or because the A/P driven by the heading bug did the
> same thing. Is suspect you could even get it to go from ARM to CPLD by
> moving the course pointer so as to make the deviation bar center.
>
> > The tip of the day is to always use APR mode when coupling to
> > your GPS, even enroute. NAV mode is dumbed down
> > considerably to make crossing over VOR stations less
> > traumatic. APR doesn't suffer that, and since the GPS
> > course needle is rock solid APR works great with sharp,
> > precise coupling. You can fly a T shaped GPS approach
> > without ever decoupling it-- just feed the new headings in as
> > cued by the cnx-80.
>
> Cool, I'll have to try that. I assume you mean "feed it new courses",
> though, right?

>>> Yes that's what I meant. With the yellow d-bar.
>
> > There's a lot to learn about the KFC-200. Maybe the POH smart-
> > asses know what it does when the HSI is failed:
> > answer is that it will either cancel everything or just ignore
> > the heading bug, depending on how it's wired. Most cancel
> > everything if heading reference is lost and any mode requiring
> > heading is selected. But you can turn
> > FD and ALT modes back on and it will wing-level at a constant
> > altitude which can be very useful if you are trying to figure
> > out what to do after a gyro failure. Explore this by pulling
> > out the HSI breaker.
>
> I will have to do that too.
>
> > Have you completely figured out the pitch modes?
>
> You mean the one that suddenly puts the airplane into a 3000 FPM dive?
> Yes, I've got that one figured out :-)
>
> But, seriously, I understand altitude hold, and the UP/DOWN rocker switch
> doing the 600 FPM climb/descent thing as long as you hold it. I'm still
> working on getting the knack the of using CWS to change pitch, and getting
> it to fly a coupled ILS, but I think one more flight and I'll have that
> nailed.

>> The easy way to remember this: If the ALT light is OFF, both the
rocker and the CWS button control Pitch ATTitude. Not
specifically
annunciated. When U let up the rocker or cws, the airplane will
hold
the new pitch ATTitude at that moment.

If ALT is ON, both the button and cws hold the new ALTitude
when released. The rocker slews the airplane to a new ALT.
Mostly useful when receiving new altimeter settings

The high taste way of changing altitude is to use the controls in
the ATT
mode, leveling the pitch before engaging ALT. That way there will
be no
altitude dipping or overshooting.

I wrote several articles about this for the ABS magazine as there is
big confusion on these points.

Bill Hale BPPP instructor

In article . com>,
"Michael" > wrote:

> Correct. The only time there is a datalink between the GPS and AP is
> when you have GPSS, and even then XTE is not what drives it.

We've got GPSS in a couple of our planes (not this one). It's an add-on
box that sits between the HSI the GPS, and the A/P (Century-2000 in that
case). You engage the GPSS and put the A/P in HEADING mode. The GPSS
widget then takes XTE (or maybe angular error?) from the GPS and
synthesizes a heading error signal that makes the A/P do the right thing.

Or, at least that's the theory. It's basically just a huge ball of duct
tape interconnecting several systems that were never meant to talk to each
other in this fashion. The one I've flown is hideously under-damped and
just flies big S-turns if you're dumb enough to turn it on. I'm told
they're working on getting it adjusted. A big ball of duct tape with rows
of trim pots and an STC does not a useful tool make.

Can't comment on all this... but the 225 has an option
that does GPSS in the NAV mode. You can tell if it's
in that mode by moving your d-bar or heading bug... if
it's tracking the GPSS, these heading references will have
no effect. I think GPSS in the 225-- if you have it-- overides
the apr coupling functions.

The GPSS that is s-ing all over the place is busted. All
the ones I've seen are flawless. Their output to the
autopilot is only a turn rate command--up to standard
rate. All the other roll components in the AP are shut off.

OK- I meant FD+ALT+AP. Key concept remains... you must
aim the airplane by whatever means to get an intercept or
it will never go to clpd. If you have GPSS, this is done
automatically.

Peter sure has had worse than average luck with his 225.

Bill Hale


Peter wrote:
> >> The tip of the day is to always use APR mode when coupling to
> >> your GPS, even enroute. NAV mode is dumbed down
> >> considerably to make crossing over VOR stations less
> >> traumatic. APR doesn't suffer that, and since the GPS
> >> course needle is rock solid APR works great with sharp,
> >> precise coupling. You can fly a T shaped GPS approach
> >> without ever decoupling it-- just feed the new headings in as
> >> cued by the cnx-80.
>
> I have missed most of this thread, but it's very interesting because I
> fly a KFC-225 (Socata TB20).
>
> I've had to look into this in more depth than some because, in 4
> years, I have had the computer replaced 4 times (various faults,
> notably the internal pitch sensing accelerometer), the roll servo
> replaced 6 times (keeps burning out), and the pitch and pitch trim
> servos replaced once :)
>
> Interesting about the APR mode. I thought it worked only with a
> localiser input, i.e. somehow knowing if the KX-165 radio/nav was set
> to a LOC frequency (I believe the radios have an output which
> indicates if the nav freq is within the LOC/ILS range).
>
> The NAV mode is seriously defective in that if you engage it when the
> HSI deflection is less than 3 divisions, the AP doesn't continue to
> intercept as it should but immediately turns onto the HSI course
> pointer track, and then (over a very long time) reduces the error.
>
> If the GPS (KLN94 in my case) is set to the default enroute
> sensitivity of 5nm full-scale then you have to be at least 3nm off
> track to engage the NAV mode.
>
> It works better if the GPS is set to 1nm full-scale.
>
> It might work even better if the GPS was set to 0.3nm full-scale but
> the loop gain is then so high it doesn't track the GPS; it oscillates
> too much. I wonder what this is telling me about the system's ability
> to fly a proper GPS approach, past the FAF...
>
> However, the AP does fly an ILS perfectly, and the final phase of that
> has to be at least as sensitive as 0.3nm FS.
>
> >> heading is selected. But you can turn
> >> FD and ALT modes back on and it will wing-level at a constant
> >> altitude which can be very useful if you are trying to figure
> >> out what to do after a gyro failure. Explore this by pulling
> >> out the HSI breaker.
>
> Can one turn on AP+ALT? If you turn on FD+ALT you don't get a
> wing-levelling autopilot; you just get a wing-levelling flight
> director.
>
> >I will have to do that too.
> >
> >> Have you completely figured out the pitch modes?
> >
> >You mean the one that suddenly puts the airplane into a 3000 FPM dive?
> >Yes, I've got that one figured out :-)
>
> What is this? One of my failed computer units would randomly select
> +2000fpm or -2000fpm, and fly that vertical speed!! Quite exciting.
> Honeywell blamed a static leak for this :)
>
> I have had extensive comms with Honeywell (Olathe) about this stuff
> but nobody there seems to know anything about it.
>
> After a lot of work I managed to get my hands on the
> maintenance/installation manuals for the AP and its three different
> types of servos, and (I am an electronics hw/sw designer) found some
> highly suspect stuff in the servos, but (apart from not having a
> watchdog) the main unit seems well enough designed.
>
> The performance, when it's working fine, is excellent - probably the
> best general aviation AP I have ever used. Excellent ride in
> turbulence too.