Saturday, May 26, 2007

GPS - SNAFU or FUBAR?

In case you haven't noticed, the specification and approval process for aviation RNAV/GPS navigation systems is a mess. Peruse the recent changes to the Aeronautical Information Manual section on GPS and RNP and you'll see what I mean. The main reason that things are a mess has to do with the fact that much of the certification process has been made transparent to the pilot.

By transparent I mean pilots must know some pretty detailed stuff about their GPS before they can use it under IFR. It used to be sufficient to 1) check the aircraft you were going to fly for an Approved Flight Manual supplement that stated the GPS receiver could be used under IFR, 2) verify the GPS database had not expired, and 3) perform a RAIM prediction. Contrast that with using a VOR receiver where the only regulatory requirement is that you perform and log a successful VOR check every 30 days. There is no operational test available to pilots to ensure proper glide slope receiver operation - I guess we're just supposed to accept that on faith. The bar has apparently been set much higher for GPS receivers and the databases they use.

Designing RNAV departure, arrival, and approach procedures that will keep pilots from running into stuff is a complicated business, which helps explain why the certification process for aviation GPS receivers is pretty darn complicated, too. But that complexity should be opaque as far as the pilot is concerned, the way it is for ILS, LOC, LDA, SDF, VOR, and NDB procedures. Instead of hiding the complexity for RNAV/GPS procedures, the FAA has published some of the GPS certification data in the AIM (cf. Table 1-1-5) and apparently they expect pilots to understand the various Technical Specification Orders (TSO) and Letters of Approval (LOA). Just what we pilots needed: More trivia, more acronyms, and more chances to break regulations without realizing it.

If you haven't already guessed why I'm grinding away on this topic, it's because AOPA (among others) recently realized that an update to AC 90-100A made in March of this year contains a bombshell - the only GPS receivers widely available for GA aircraft that are now approved for substitution for DME and ADF are the Garmin 400, 500, and G1000 series. Suddenly all the old Apollo units, the GNS 480, the Chelton Flight Systems units, and many of the King GPS units that thousands of pilots have been using as a substitution for DME and ADF are suddenly no longer approved for such uses. Oh, and these units can't be used to fly T-Routes either.

The affected GPS unit presumably still comply with the original TSO under which they were certificated, but they no longer comply with AC 90-100A. And let's not forget about all the Approved Flight Manual Supplements out there that say these GPS units are approved for these uses. One would assume that they are now incorrect. Talk about The large print giveth and the small print taketh away!

A reasonable person has to wonder "Why did it take until the end of May for this to become common knowledge?"

I've spent considerable time trying to parse AC 90-100A and AC 20-153 and the only thing that is crystal clear to me is that the FAA isn't making it easy for us pilots. Nor are the manufacturers of GPS receivers. It seems that the reason that the majority of GPS receivers certificated under TSO 129 have been rendered non-compliant is that the manufacturers do not have a Type 2 Letter of Approval (LOA) for the aviation databases that they generate for said GPS receivers. And if there is no Type 2 LOA, then the GPS receivers' use is restricted until such time as an LOA is granted by the FAA. For some GPS receivers, the LOA is pending while for others there are no plans to ever obtain an LOA, hence they will forever be limited.

The interesting thing here is that the FAA has recognized the importance of valid data for GPS receivers. I appreciate this as I had to point out to Garmin on at least one occasion that their database did not have all the initial approach fixes defined on a plate for an ILS approach to an airport in Northern California. Equally interesting is that a GPS that complies with the original requirements specified in the TSO years ago can still be rendered less useful after the fact by an advisory circular.

While you're contemplating that cascading chain of dominos, consider this spreadsheet which lists the compliance of virtually all GPS, RNAV, and FMS from Boeing to Rockwell. That spreadsheet was apparently created by someone at the FAA at the end of 2006.

So did the FAA not notify the manufacturers of the non-compliant hardware until a few weeks ago, even though they apparently knew this in December of 2006 and the updates to AC 90-100A were released in March of this year? For their parts, Garmin and Rockwell-Collins, posted their LOAs right around the time AC 90-100A was updated in March. But did the manufacturers notify their potential customers, customers who purchased their products, not to mention the pilots who don't own but use their equipment? I don't know the answers to these questions, but I wish I did.

With the push for ATC user fees, the FAA keeps talking about providing better service to their customers. Based on the state of FSS under Lockheed-Martin, this GPS fiasco, and the fact that virtually all Boeing, Airbus, Embraer, Dassualt, and Gulfstream equipment already seem to be in compliance with AC 90-100A, one has to wonder just who are these customers the FAA keeps talking about?

7 comments:

john said...

you've listening to The Main Bang have you?

I'm only on the fringes and I'm worried. First the GPS controls are obviously designed by an engineer who ran away screaming anytime human factors was mentioned and now the FAA is following the same path.

As far as I know Mitre is the brains of the FAA and I have seen no work/studies from them on GPS human interface, design, functionality. Probably they haven't done anything on GPS databases either so of course no timely information was given to manufacturers of user community.

Until the Bush administration goes away and the FAA is replaced by another agency the system is just going to get worse.

I'm gonna go into the Light Sport hole and hide until its over.

Ron said...

Re: "There is no operational test available to pilots to ensure proper glide slope receiver operation" -- I'd have to disagree with that.

Approach plates always show the glideslope altitude at the final approach fix. For example, the ILS 19R approach at SNA shows an altitude of 2202' MSL over LEMON, the FAF for that approach.

The whole reason that's on there is so the pilot can cross check their glideslope receiver before descending too close to the ground.

Not as precise as a RAIM check, perhaps, but it does the job.

Regarding the Advisory Circular, I, too, perused the revised document and couldn't make heads or tails of it. It's beyond ridiculous that the CNX80 -- which in many ways is a generation ahead of the GNS400/500 series -- has been relegated to the bleachers, even if only temporarily.

John is right, the situation with the FAA and the Administration is only only going to get worse. I shudder to think what our once vibrant GA system will look like once they're done poisoning it.

John said...

Ron,

When I said "operational test" for glide slope operation, I meant prior to taking flight. The only way I know that a GS receiver can be tested prior to becoming airborne is by an avionics shop with the necessary equipment.

I agree that pilots should always verify their altitude on the GS when they can and the SNA ILS 19R is a good example. You can verify your altitude at LEMON because there is a DME distance and an outer marker. Either or both will tell you when you are over LEMON.

However there are ILS approaches where there is no way to tell your distance to the runway unless you have DME. The OAK ILS 27R is a good example. There is a DME distance published for CUVSA, but DME is not required for the full ILS approach.

So depending on how your aircraft is equipped and the approach you are flying, you may or may not know that your GS receiver is working properly. I had a GS receiver failure on the OAK ILS 27R approach in actual conditions, so I have some firsthand experience, but that's a story for another day.

When flying a plane with dual glide slope receivers, I recommend that both NAV radios be tuned to the localizer and that the pilot verify that the GS indications on the #1 and the #2 indicators agree.

Ron said...

Fair enough. I figure virtually nobody is flying IFR these days without either DME or some sort of GPS unit. Even if it's just a cheap handheld not legal for IFR use, it can be used to confirm position over the final approach fix and cross check the glideslope.

A pre-flight glideslope check might prevent you from taking off with an inoperative glideslope, but it wouldn't necessarily point out a receiver or CDI that failed mid-flight any more than a RAIM check means you'll never have a RAIM alert on the approach.

The dual glideslope idea is good. Unlike DME or GPS, I'm not sure most aircraft are equipped with dual glideslope capability. Even the new glass airplanes only have a single HSI display. I always wondered why you couldn't display the glideslope information from both radios on the HSI in different colors, and have the system annunciate a warning if they deviated by more than a prescribed amount.

Peter said...

I agree that there's no glideslope test you can do prior to flying an approach. But what about the localizer? There's no test for that either. At least if you find a bum glideslope at many locations you can execute a localizer approach. But how do you find out if the localizer works, or which of the two (if so equipped) is indicating correctly?

John said...

AC 90-100A dated 3-1-2007 deals with requirements to perform RNAV 1 and RNAV 2 type RNAV routes and does not deal with whether or not a GPS can be used in lieu of a DME or ADF. I do not believe this AC is relevant to the "In Lieu of ADF and/or DME" discussion.

However, the AIM in chapter 1: Navigational Aids has had a section 1-1-19 (f), that has in the past described in detail the requirements for "Use of GPS in Lieu of ADF and DME". In the 3-15-2007 update to the AIM, the section was deleted. However, in Table 1-1-6 that is titled "GPS Approval Required/ Authorized Use", it still lists a column for "In Lieu of ADF and/or DME" and if the equipment is approved for: "IFR En Route and Terminal"; or "IFR Enroute, Terminal, and Approach"; or "IFR Oceanic / Remote"; then it lists it as authorized to substitute GPS in Lieu of ADF or DME with a note requiring that the database be current. This covers all IFR approved GPS navigators. What is missing from the AIM is the detail
guidance for how to use GPS in Lieu of ADF or DME. That section should be reinstated into the AIM IMHO.

John said...

Peter,

A LOC is very similar to a VOR except that it only broadcasts one course, so when you perform the VOR check you are also verifying (to some extent) that localizer reception will also function correctly.

The glideslope is a different animal altogether.

John,

The section in the AIM about RNAV (formerly GPS) substition is now contained in 1-2-3(c) "Allowable Operations".

Section 1-2-3(b) "Allowable RNAV Equipment" says:

"Standalone GPS systems, compliant with AC 90-100, are included in this set of equipment. A list of compliant systems is available under "Policies & Guidance" at the following website: ..."

So either their documentation is messed up and confusing or several people (including folks at AOPA) have mis-interpreted what has been written.