CAA Approve GPS approaches [Archive]

IO540

2nd May 2007, 09:05

There is little chance of DME going away. It is firmly established on conventional approaches (e.g. ILS) and they will "never" go away. It's mandatory in many situations around the world.

Even in the USA, where an IFR GPS can susbstitute for a DME, you get problems with the choice of alternate if you don't have conventional navaid landing capability. And DME is mandatory above, can't remember, FL-something, as is Mode C.

I would never get rid of VOR/DME. It forms an essential backup for the GPS. Remember most pilots have only one GPS. I've had equipment failures and had to fall back on other stuff.

The real key to improving GA utility is doing away with mandatory ATC for instrument approaches. There is a lot of politics involved in this, including some ATC trade union issues (the unions are sure to object), and the "user pays" NATS charging practices (which would mean the airfield getting a bill from a nearby ATC unit for the "coordination") would need to be rethought. The simple solution (self announcement on a common approach frequency) is not likely to be accepted in the European anally retentive "safety" climate, even though it would clearly work perfectly well, and would be considerably safer than the currently perfectly legal practice of multiple arrivals carrying out DIY approaches non-radio in Class G.

Roffa

2nd May 2007, 11:27

IO540, the NATS atco trade union (Prospect) would probably take no interest whatsoever in whether or not instrument approaches were allowed at non ATC airfields. They negotiate our terms and conditions of our employment. There are also other trade unions that represent non NATS ATC staff and I'm sure they wouldn't be that interested either.

The organisation that might take an interest is GATCO (http://www.gatco.org/), the Guild of Air Traffic Control Officers (most definitely not a trade union), who like to represent all UK atcos (civil and mil) on professional matters. I am not a member so won't comment on what their position might be any further.

Stop banging on about the union on this subject, I doubt they'll take any notice.

Genghis the Engineer

2nd May 2007, 12:28

but if the DME and GPS agree then you are almost certainly where you think,

If the DME and GPS agree, then you are almost certainly at exactly the same height as the DME station - and where you think.

But, since you are then almost certainly within a few feet of the ground, you should probably be looking out of the window, not at the instruments!

:}

G

(Please excuse the flippancy - of-course this is a good thing, and no doubt as people start to use them, they'll become more widely available and may not mandate other instrument approaches also being available, which can only be good for GA, since it'll cut costs enormously, whilst expanding the usability of many airfields.)

Fuji Abound

2nd May 2007, 12:33

If the DME and GPS agree, then you are almost certainly at exactly the same height as the DME station

I rather had in mind there not being a gross difference!

Genghis the Engineer

2nd May 2007, 12:41

Of-course, but it's important to remember whether you're fine tuning or doing a coarse check.

IO540

2nd May 2007, 14:00

Come on you lot; the difference between GPS and DME is under 0.1nm in almost any flying context.

Work out the cos of 3 degrees! It's 0.9986. The difference is negligible, on a 3 degree ILS.

denhamflyer

2nd May 2007, 15:25

My point was to remove the mandatory requirement for carriage. The French and USA have done this. However it would still be required for any approach that actually stated its use - same as an NDB. I can see that around Europe (in the not too distant future) that some airfields would be GPS only..:cool:

and some day maybe even a FISO could be allowed to handle it.

Fuji Abound

2nd May 2007, 15:36

DF

Interesting point.

Leaving aside the possibility of GPS failure (and I encompass in that either the aircraft born unit(s) or the external system (due to interference with the signal for example), and given in these circumstances you could (and would) fly an alternative approach using different instruments, would you ever be happy with completly relying on a GPS approach without any other positional cross check in circumstances where a gross error would place you in terrain before becoming visual?

If so why?

(and for the avoidance of doubt it is a genuine discussion point and not because I am expressing a particular view one way or the other or trying to say it would be irresponsible to do so).

bookworm

2nd May 2007, 16:02

If so why?

If not why not? We do that with other navaids, some of which are several orders of magnitude less reliable than a RAIM-equipped GPS system.

fireflybob

2nd May 2007, 16:53

The real key to improving GA utility is doing away with mandatory ATC for instrument approaches. There is a lot of politics involved in this, including some ATC trade union issues (the unions are sure to object), and the "user pays" NATS charging practices (which would mean the airfield getting a bill from a nearby ATC unit for the "coordination") would need to be rethought. The simple solution (self announcement on a common approach frequency) is not likely to be accepted in the European anally retentive "safety" climate, even though it would clearly work perfectly well, and would be considerably safer than the currently perfectly legal practice of multiple arrivals carrying out DIY approaches non-radio in Class G.

Inevitably there will be opposing views on this but I for one would not like to be conducting a "proper" IFR let down in IMC without a licensed air traffic controller separating aircraft in accordance with standard criteria.

The fact that some fly DIY approaches without ATC is not a reason for changing the current rules and the CAA approval of GPS (which I endorse) will no doubt mean that we will see more of this activity.

This has been covered on other threads before I know!

IO540

2nd May 2007, 17:18

Why a licensed ATCO, FFB?

I also wouldn't worry about too many pilots trying to do this in the same location at the same time. Get in the air on any BKN008 or worse day and the sky is like a cemetery - almost nobody up there. You can even get an RIS from Gatwick Director. 99% of UK GA is doing burger runs on nice days...

The # of UK pilots carrying TSOd GPS units is tiny. Most of them are N-reg; this was obvious from the poor participation in the CAA GPS trial. Very few UK PPLs flying a G-reg have an IR, and IMC Rated pilots don't need an IFR GPS with a current database costing a couple of hundred quid a year to keep updated. These people won't be able to fly the GPS approaches - well, no more than they can fly any DIY approach right now with their handheld unit. With the sort of decision heights we will be getting on GPS approaches, working out a DIY letdown (GPS, or VOR/DME) from the 1:50k O/S chart is very easy.

When non-ATC airfields get GPS approaches, the utility value in GA will improve significantly but I predict it will take a little while to filter down into more activity all around. For that, we will need a lot of avionics upgrades, and a lot more new aircraft. Still, it's very good news.

High Wing Drifter

2nd May 2007, 17:42

but if the DME and GPS agree then you are almost certainly where you think.
Come on you lot; the difference between GPS and DME is under 0.1nm in almost any flying context.
Is GPS distance relative to the threshold of the runway in use or the ARP? Depending on the answer, the airfield and the procedure the DME and GPS distance difference could be several miles. Even the slant distance at the start of a typical base turn you see probably about a 0.5nm difference (I think it should be sin(3), not cos).

And how do you check that the DME is working when you start an approach that needs it?
You will know because you have ID it been using it before you start the procedure and its reading corresponds to the other nav instruments which agree with each other. Many airfields have timed approach procedures, some don't. If not and DME is wonky then you cannot do an instrument approach. Where permitted, it is standard practice to start a timer at the FAF/FAP for all approaches so that it is possible to revert to a timed approach should the DME fail.
would you ever be happy with completly relying on a GPS approach without any other positional cross check in circumstances where a gross error would place you in terrain before becoming visual?
I don't think anybody would be happy with just one instrument. Most approaches involve two or three with the non approach receivers tuned to other navaids (if possible).

Fuji Abound

2nd May 2007, 17:50

If not why not?

http://youtube.com/watch?v=GelRBhJ4gmI

One possible reason where relying on something as "reliable" as a localiser without a cross check proved fatal.

Islander2

2nd May 2007, 18:49

Even the slant distance at the start of a typical base turn you see probably about a 0.5nm difference
Wow, 350nm final, commenced at 110,000ft! That's some approach.

IO540

2nd May 2007, 21:17

HWD

From school: SOH-CAH-TOA

To get the ratio of the adjacent side of a triangle, and its hypothenuse, you use the cosine. This gives you the radio of the GPS distance, and the DME distance, respectively.

For any realistic approach angle, the two are virtually identical.

Obviously they can be well apart if flying directly over the navaid at some height, but then it's obvious what one is doing. Every time you fly over a waypoint that is a VOR/DME, using the GPS, with the DME tuned in (the normal practice enroute) you see this.

It's a good point about a GPS distance to the MAP being potentially different to the DME distance to the MAP, even at the same airport, but that's OK because many approaches around the world read non-zero DME at the MAP, and the pilot is supposed to read the approach plate.

What does differ, and has caused crashes, is the fact that the DME distance is to the same point as you sequence down the approach, whereas the GPS distance is to the next WP on the approach. I don't see a solution to this except being extra careful. It works fine in the USA, in heavy usage, and far more people have got killed flying NDB approaches, etc. AFAIK there is simply no way to code a GPS database so that the GPS reads the distance to the MAP throughout the approach.

I should also mention that identing a DME means only that the right frequency has been set, and it isn't undergoing official maintenance. It's just like a VOR really - no way to see the reading is real. With an IFR GPS it's possible to get a duff reading which looks "right" but it's very unlikely. Only by checking two independent systems can one be sure.

bookworm

2nd May 2007, 21:40

One possible reason where relying on something as "reliable" as a localiser without a cross check proved fatal.

The problem in that case was with the glideslope and it didn't prove fatal, though the point that it could have done is well made. Nevertheless, aircraft can and do trust their safety to a single navaid, an ILS, virtually every minute of every day.

With GPS, there are more in-built intergity checks. Failure modes still do exist. No system can ever be 100% safe. The issue is the cost/benefit of a cross-check. Neither DME transceivers, nor transponders, are cheap.

IO540

3rd May 2007, 07:35

To me, the most suprising thing was that the crew flew the ILS despite knowing it had been notamed as "unmonitored".

I know nothing about approved company procedures but would have expected them to be a bit tighter. But then CAA approved procedures do allow a G-reg 747 to fly from LA to the UK on 3 engines, so who am I to question flying a suspect ILS?

bookworm

3rd May 2007, 12:05

To me, the most suprising thing was that the crew flew the ILS despite knowing it had been notamed as "unmonitored".

Until I read the report, I had absolutely no grasp of the implications of "unmonitored". I suspect the crew were the same. ANZ has done us all a favour by highlighting this.

soay

3rd May 2007, 13:13

Silly Old Harry - Caught A Herring - Trawling Off America
or Some Officers Have - Curly Auburn Hair - To Offer Attraction. :ugh:

ThePirateKing

3rd May 2007, 13:35

Some Old Hangers Can Almost Hold Two Old Aeroplanes.

Honestly - that's what I was taught! Coincidental, eh?

High Wing Drifter

3rd May 2007, 15:24

Hmm :\, yes. Well yes I guess the sine is a rather long winded way to arrive at the answer, but a rather weak defence would be that it sort of integrates with the gradient that some plates show on the final segment. Ok, I'll get me coat...but first :-
and the pilot is supposed to read the approach plate.My assertion remains is that the conclusion is that cross-checking a GPS and DME is not a trivial activity.
AFAIK there is simply no way to code a GPS database so that the GPS reads the distance to the MAP throughout the approach.I presume you mean with a current product. Why couldn't or wouldn't this be added as a feature by the likes of Garmin? I guess they must have considered it.
Nevertheless, aircraft can and do trust their safety to a single navaid, an ILS, virtually every minute of every day.I would count the DME as the second instrument that should tie in with the crosscheck as shown on the briefing strip on the plate.

IO540

3rd May 2007, 15:42

My assertion remains is that the conclusion is that cross-checking a GPS and DME is not a trivial activity.

Well, they will read practically the same in the case of co-located DME and WP but if the two are not co-located then they won't, but the pilot would know that...

I presume you mean with a current product. Why couldn't or wouldn't this be added as a feature by the likes of Garmin? I guess they must have considered it.

I am no expert on this, and the format of the databases is in any case a secret owned by Jepp and licensed to the GPS mfgs, but I believe that you can't have it both ways i.e. have an intermediate WP which can somehow be referenced (i.e. you can see when you have passed it by) and at the same time have the GPS show the distance to a different WP (the MAP, usually). Technically one could show two distances concurrently but I don't think it's done. It would be a major source of confusion.

This debate has been done to death over far more years than I've been flying, but one can make a case for both systems equally. Fundamentally, neither is better. One can argue the "always see the distance to MAP" is better, and one can argue (especially on a multi-stepdown approach) that the "distance to the next stepdown" is better.

One is unlikely to get killed flying an ILS (so long as the needles are in the middle, or roughly so) but with a stepdown approach one has to do the stepdowns right, otherwise one will possibly get killed. With a DME, you have to notice the reading passing through a certain value and then take an action (step down). With the GPS, you merely have to notice it reaching zero (which is a lot more obvious) and then take the same action. I don't think - old habits excepted - that either is better and very possibly the GPS system is better.

Having said all that, flying a full GPS approach, plus the missed approach, is a very technical procedure which is highly GPS type dependent. I deliberately avoided GPS equipped schools when doing my IR in the USA because I would not be flying with the same GPS here in the UK, and there are no GPS approaches here anyway - except at airports with other approaches, usually (in Europe, outside the UK) at places with an ILS. But I learnt anough about it to know it's very nontrivial. It's easy enough to fly to an IAF and then fly the approach to the runway, manually or autopilot-coupled, but the way the GPS sequences (or doesn't) onto the missed segment varies according to unit type and I guess most UK pilots would just transition to conventional navaids... which is just as well since that bit is probably going back to some NDB :)

n5296s

3rd May 2007, 16:03

I don't think anybody would be happy with just one instrument. Most approaches involve two or three with the non approach receivers tuned to other navaids (if possible).
That's exactly what people do all the time with a GPS approach. My local airport has a GPS approach which is what most people use. Here (US) I'd say that practically all serious IR planes have IFR GPS - I never hear anyone using the VOR approach which is available although not for practice approaches. My plane has two IFR GPS, but that's not the usual case. GPS has a lot of self-testing built in (RAIM).
wrt checking the DME - sure you can listen to the ID. That doesn't tell you that the DME is giving the right answer though, only that it is transmitting. If your DME receiver was for some reason misinterpreting the signal, you wouldn't know until you stepped down into some granite. Just because it's been around since the vacuum tube (or thermionic valve if you prefer) doesn't mean it's intrinsically more reliable.
n5296s

411A

3rd May 2007, 16:18

Well now, I see 20th century technology has finally arrived on the eastern side of the great divide, well into the 21st century...GPS.

In the land where Oboe, Gee, Decca, Loran, Radar and a host of other quite unique systems were developed, GPS has finally made its appearance for approaches....BRAVO I say.

I have one in my private airplane (IFR approved) for over nine years and its accuracy is unsurpassed, for approaches, never mind enroute and terminal ops.

And yes, I've used Decca too (in the USA no less, as it was on trial in the southern California area in the late sixties) and the old Loran A (not to be confused with the present day Loran C)...very good for their time.
After all, the Brits were quite good at accurate electronic navigation.

Why, I wonder, did approval for GPS approaches take so ....l o n g ?

High Wing Drifter

3rd May 2007, 16:38

Why, I wonder, did approval for GPS approaches take so ....l o n g ?
being somewhat distrustful, I would suggest that Gallileo had a hand in the politics of delay :*

IO540

3rd May 2007, 17:11

Galileo is likely to be abandoned

here (http://www.theage.com.au/news/TECHNOLOGY/EU-warns-against-further-delays-to-Galileo-satellite-system/2007/03/15/1173722628905.html)
here (http://www.gpsdaily.com/reports/Galileo_Development_Stalled_Over_Profitability_Questions_999 .html)
here (http://space.newscientist.com/article.ns?id=dn11392)

I really like the quote in the New Scientist article, where the man wonders why people don't want to pay for it when they can get it free :ugh:

bookworm

3rd May 2007, 17:20

Nevertheless, aircraft can and do trust their safety to a single navaid, an ILS, virtually every minute of every day.

I would count the DME as the second instrument that should tie in with the crosscheck as shown on the briefing strip on the plate.

What DME? You're spolied by this British habit of sticking a DME at every airport. Plenty of ILSs are flown with nothing more than markers.

Of course the ANZ accident shows why relying on markers is a spectacularly bad idea: with an erroneous glideslope you'll probably be close to the correct altitude at intercept around the OM. But you may never reach the MM.

High Wing Drifter

3rd May 2007, 18:50

IO,
Galileo is likely to be abandoned
I suspect this is good news. But if this happens, who will be accountable for the cost to EU citizens thus far? Accountable, Europe...another debate I suspect :=

Bookworm,
What DME? You're spolied by this British habit of sticking a DME at every airport. Plenty of ILSs are flown with nothing more than markers.
I'm wary that I might be over-laboring the point, but marker/lctr is a just as much a navaid and used to gross check the glideslope in the same way the DME is used. I still see that as an instrument crosscheck.

Sir George Cayley

3rd May 2007, 22:24

There-in lies the rub.

As I understand it, in the States pilots prefer to let down at each step down, which is why the equipment suppliers add these fixes as WPs.

I asked a chap from Jepps at Farnboro' and he said "Vee respond to our kustomers" (Cheap German joke):oh:

In this country, all the instructors and examiners I've spoken to prefer pilots to construct a mental picture of a GP based on 1500' QFE at 5 miles or modified as necessary from the FAF.

In this case when attempting to replicate continuous descent, seeing the distance to go constantly reseting each time a step down is passed after FAF can lead to disorientation and a mouthful of cockpit coaming!

For those of used to ILS/DME anything more complex is to be avoided especially with advancing years.:{

Interestingly, a number of US accidents exhibit similar crash sites short of the RWY on the c/l with high rate of descent. Be interesting to see how many of them had step downs.

I vote for simple procedures that replicate what we do now.:ok:
Sir George Cayley

PPRuNe Radar

4th May 2007, 00:24

We should embrace all technology available in the UK and encourage the CAA to issue the appropriate guidelines. But lets try not to crash in to terrain like some pilots in other parts of the world because we blindly follow GPS. NTSB reports refer ........

n5296s

4th May 2007, 00:27

Not sure there have been any accidents as a result of people blindly following GPS... certainly there have been some as a result of people blindly FAILING to follow GPS, as there have with other navaids too. Can you cite actual NTSB reports where people followed their GPS and ran into terrain?

John

mm_flynn

4th May 2007, 05:57

I too would be interested in some facts on GPS approach accidents.

A quick search of the NTSB database using GPS IFR fatal didn't show anything related to diving into the ground at a way point.

I am aware of instances in the US where pilots have stepped down early and hit the mountain the step down fix protects them from. But this probably argues more for the benefit of the distance to go being to the next fix rather than to the MAP,threshold, or the DME station.

IO540

4th May 2007, 06:10

SGC

The "pseudo GS" v. "dive and drive" approach is a separate matter from what we have been talking about which is whether the "DME" should indicate the dist to MAP or whether it should indicate the dist to the next WP.

Obviously a pseudo ILS is safer than doing separate stepdowns, but the approaches were originally designed (using TERPS, in the U.S. case) with the latter method.

Most (not all) stepdown approaches can be redrafted as a pseudo ILS, which is great because, with the right GPS kit, you can fly them as a virtual ILS. They are doing that in the USA but over here this is probably many years away and would require EGNOS (the European version of WAAS).

I am aware of instances in the US where pilots have stepped down early and hit the mountain the step down fix protects them from. But this probably argues more for the benefit of the distance to go being to the next fix rather than to the MAP,threshold, or the DME station.

Agree totally. Getting the stepdowns one step out of sync (possibly due to poor drafting by Jepp) is nothing to do with GPS.

slim_slag

4th May 2007, 08:18

Obviously a pseudo ILS is safer than doing separate stepdownsWhy is that 'obvious' ?

mm_flynn

4th May 2007, 14:54

Obviously a pseudo ILS is safer than doing separate stepdowns

Why is that 'obvious' ?
1 - You can't get yourself mentally out of sequence and step down early if you are following a precalculated terrain safe glidepath (assuming you worked it out correctly!)

2 - You have a stabilised approach from a long way out, minimizing work

3 - You won't dive through your altitude (re Sir GC's comment)

4 - When you break out you should be able all set up on final on a good glidepath

vs.

1 - You don't have to calculate/check your pseudo glide path is safe if you do a step down approach

2 - You will probably breakout earlier when stepping down

411A

4th May 2007, 15:41

Certainly many here are totally unfamiliar with the way most GPS approaches are actually flown in the United States.
Now, in my airplane I have a Bendix/King KLN89B, and this unit is an earlier generation IFR approved GPS, with a moving map display, which even today, is just as useful as the newer generation color displays found on Garmin 430/530 series units.
It will be found that, during a normal GPS approach, that the distance to go to the next waypoint (or stepdown fix, if you prefer) is continuously displayed, and from this, you the pilot, can clearly see your position in the approach...WHEREAS, during a normal VOR/DME approach, the total distance from the station is displayed.
So, we can see from the above, that once you have passed one particular waypoint (again, stepdown fix, if you prefer) the display changes to the next waypoint, and the distance to go to that waypoint is then displayed.
IF a pilot is unable to follow this type of approach with GPS equipment as used in the USA, realistically speaking, he/she does NOT deserve to have an instrument rating in the first place.
It is, quite frankly folks, easier than falling off a log.
No mental gymnastics required, just follow the procedure as displayed.

As for pilots of GPS equipped aircraft crashing due to misrepresentation of fixes displayed, I can only think of one, and in this particular case, the pilots were following the VOR/DME procedure, ignoring completely the GPS presentation.

Not too bright.:ugh:

slim_slag

4th May 2007, 15:48

Hard to disagree with 411a.

mm_flynn, nice to see somebody look at this in a structured way.

IO540

4th May 2007, 15:52

I agree with 411a and MMF, of course. Flying the approach is dead easy.

The bit where different units differ is the way the missed approach part is handled.

You don't (necessarily) get automatic waypoint sequencing after the MAP. I believe the KLN94 offers the next waypoint after the MAP (which will usually be somewhere straight ahead) and you have to press DCT ENTER to select that.

There is also plenty of older stuff around that doesn't display arcs and thus doesn't display SIDs/STARs in a meaningful manner. I went on a manufacturer course on this once.

411A

4th May 2007, 16:09

slim_slag brings up an interesting point about 'early' descents.
I have been flying for over forty years professionaly, in a variety of airplanes, and from the earliest days, have used the step down method of flying non- precision approaches.
This does NOT require that you absolutley dive toward the next fix, and thereafter fly level for a protracted period of time.
It is so very easy to establish a (roughly) 700-800 ft/min rate of descent, and by doing this, arrive at the next step down fix at right about the correct altitude.
However, in order to actually be able to land the airplane, you positively need the MAXIMUM time at the MDA, to see the airport, and this is especially important for straight in approaches, but it is also important for circling as well.
Don't dive at the ground, use that 700-800 ft/min rate of descent, and 9 times out of ten, it works out as advertised.
In big airliners or small GA airplanes.
It ain't especially hard, folks, it just takes discipline.

I0540 also brings up an interesting point.
Clearly, not all GPS units are the same.
Each requires a certain sequence of 'button pushing' to achieve the desired results...and, many times that operating manual is not the clearest form of information.
What is required is...practise, with operating the unit prior to actually leaving the ground.
And then, fly in VFR weather for awhile with a safety pilot, to get the particular units sequence down pat, prior to sticking your nose in any cloud.
Properly used, these IFR GPS units are a HUGE increase in overall safety.