Menu Sign In Contact FAQ
Banner
Welcome to our forums

Autopilots which use GPS to fly an ILS

My previous 30 year analog KAP 150, and KFC 200 are able to track the ILS without GPS aiding,

Yes, of course, this is old stuff.

However, for an autopilot to work properly, it needs to be flight tested all around the loading and speed envelope, and some stability margins established so adequate performance is maintained in production. Some models probably were “not quite” so well tested and oscillate in some airframes (STEC for example). King ones tend to be very good. This is a fair bit of work. I reckon it will take at least a week of flying, per airframe type, to collect this data, and that’s assuming everything works first time.

My guess would be that Garmin found some way to use GPS data to avoid doing – on the GFC500 – all the testing which normally used to be done.

I think, regardless of what the answer actually is, Garmin are not going to say, because it would sound like getting some drug certified by rigging the test group to be only females of a certain age range and then extrapolating from that to males using some data from another trial

Administrator
Shoreham EGKA, United Kingdom

Peter wrote:

I think, regardless of what the answer actually is, Garmin are not going to say, because it would sound like getting some drug certified by rigging the test group to be only females of a certain age range and then extrapolating from that to males using some data from another trial

I guess FAA and EASA would know, since the GFC500 has an STC.

Last Edited by Airborne_Again at 06 Dec 21:15
ESKC (Uppsala/Sundbro), Sweden

Airborne_Again wrote:

I guess FAA and EASA would know, since the GFC500 has an STC.

Not sure. I think that the company that has performed the external software audit should…

EGTR

Peter wrote:

My guess would be that Garmin found some way to use GPS data to avoid doing – on the GFC500 – all the testing which normally used to be done.

I doubt it. Getting autopilot STC for each and every aircraft is very time consuming. Often pilots are incented to lease their aircraft to the manufacturer, typically for a period of 6 months or more to be a test bed for the type STC work. The most expensive part of the support for a new type is the testing that must be demonstrated to FAA for the autopilot. It includes the installation aspects that are unique to the type. The avionics shop that I use has a technician that owns a Mooney. It is not included on the current STC and it has been in the Q for several years. He went ahead and installed the GFC500 and has it tagged as inop for now. I think he has a Mooney M20G or M20F. Currently the support only goes down to the M20J. See https://buy.garmin.com/en-US/US/p/604257#additional Read thru their notes at the bottom if one is interested in having their aircraft added to the list. Not much seems to be going on now due to the Covid.

KUZA, United States

From here

Many GA autopilots are poor at flying a fully coupled approach as they rely on a Middle Marker signal to reduce system sensitivity as you get closer to the runway. With most Marker beacons being removed, this means the autopilot stays in a high gain mode and consequently will tend to over-correct to any deviations. More sophisticated autopilots such as found in Kingairs and above can use radio altitude or DME distance to alter the gain thresholds. The GFC500 using GPS aiding is incredibly accurate. Considering that WAAS-GPS is approved for CAT 1 ILS levels of approach and has to demonstrate very high levels of reliability, the chance of losing GPS when conducting an ILS approach with a GFC500 is extremely remote.

Avionics geek.
Somewhere remote in Devon, UK.

Strange… I watched few webinars on GFC500 and they were very clear on GFC500 ability to couple with ILS, LOC and VOR.

LDZA LDVA, Croatia

It seems not without a valid GPS signal.

Anyway, chances of losing the GPS exactly when flying an ILS are extremely low, I may take my chances on this.

Only if I could find a way to get some value out of my existing Stec 60-2. As it has an altitude preselector as well, do you think I can sell it to somebody that might need it?

Or is there some company that would accept a buy back and install the GFC?

LRIA, Romania

wigglyamp wrote:

Many GA autopilots are poor at flying a fully coupled approach as they rely on a Middle Marker signal to reduce system sensitivity as you get closer to the runway.

Thank you for this, did not know it. Explains a lot.

Smooth
LOXX, Austria

The GFC500 using GPS aiding is incredibly accurate.

Is there something published which states that this is what Gamin use GPS data for, when flying an ILS?

It would be most surprising because it would need an awareness from the GPS end that there is an airport coming up (and to, de facto, use GPS as a TDME) but to fly an ILS with the GFC500 you don’t need to set up an IAP on the GPS. Or do you?. If you do have to set up an IAP on the GPS then that would confirm this hypothesis, but nobody else has reported this. Everybody else posting on this topic (myself included) here and elsewhere has been speculating as to what Garmin use GPS for.

BTW I think the gain is increased as one gets nearer, not decreased, because that is what GPS navigators do. At 30nm they set the gain to 1nm FSD and after the FAF they gradually wind it up from 1nm to 0.3nm FSD. I would think the reason why an analog autopilot flies an ILS badly is because the gain is fixed and is too low. But then how does the KFC225 do it? It flies an ILS precisely. There is a middle marker pin (may be wired up; I haven’t looked) but almost no airport in Europe has a MM anyway. It is “digital” and since it knows when it did the LOC intercept, and the GS intercept, it probably makes some guesses as to where it is from those points.

The gains (proportional and integral; I don’t think autopilots use the derivative term also) need to be in the right band for the aircraft behaviour because if the gain is too high (say, the wire has come off the tacho inside the servo) you get strong oscillations around that axis.

Garmin clearly use the GPS for something, however… That old article I linked above may have clues, even though it is mostly marketing BS about why the GFC500 is not TSOd but the 600 is and costs way more $$$.

Considering that WAAS-GPS is approved for CAT 1 ILS levels of approach and has to demonstrate very high levels of reliability, the chance of losing GPS when conducting an ILS approach with a GFC500 is extremely remote.

Yes the probability of a GPS space segment failure is extremely low, but not the probability of having to fly an ILS (instead of a GPS IAP) because your GPS box has packed up. That is just the probability of the GPS box packing up and if it happened while flying a GPS IAP, the accepted procedure under IFR is to go around and fly a non-GPS approach, there or elsewhere.

One could be on a 5hr flight, with the GPS navigator blowing up after 1hr, so you fly the other 4hrs with a tablet or whatever and with the AP in the HDG mode, and then you want to land with an ILS. The point is that the GPS didn’t have to pack up during the ILS; it had the whole 4hrs+ to pack up, which is a much higher probability.

Administrator
Shoreham EGKA, United Kingdom

Mooney_Driver wrote:

Yea, that would be a total no go for me as well.

Not for me. I don’t see that the likelihood of the GPS box (or the GPS system) failing is greater than the likelihood of an A/P failure. In fact, my experience of both GPS boxes and A/Ps is much the opposite and clearly lots of people on EuroGA have the same experience.

Even with a GPS failure, the GFC500 will work in HDG and V/S modes. The likelihood of a GPS failure happening at the same time that you have to fly an IAP to near minima is even more remote.

ESKC (Uppsala/Sundbro), Sweden
Sign in to add your message

Back to Top