Tracking towards a DME station

[767bill]

Following a recent failure of ADF and VOR equipment simultaneously on one flight I realised that with the DME still working I could still track towards that - as it happens I was after a turn routing towards a working DME station. Weather was fine so probably not really essential but comforting all the same. It also occurred to me that once the DME distance started increasing (meaning that I had gone abeam the station) then allowing for height above station I could have tracked 90 in either direction to continue tracking towards it - had I not in fact seen the station - on the basis that if distance had increased I would then have turned 180...probably all quite basic but worth remembering since if visability had been poor this would have been very useful in confirming position...

[dirkdj]

The only DME-only navigation I would do is a DME arc, where you know the distance to the station and fly a (say) 10 NM circle around it, not the best way to do XC fliying but used in approaches.

If you have a good map, you could do quick fixes on three DME stations and plot three circles on your map, your position would be in the center of the intersection of the three circles at the time of observation.

All very useless in VFR, just look out of the window and use common sense.

[FlyingForFun]

Interesting subject. Makes a change to find something I don't recall being discussed before!

Although I can't argue with what you say, I'm not sure I'd like to rely on it, especially with airspace nearby. As you point out, you can't track directly towards the DME, only to a position which is somewhere abeam. Even when abeam, you can turn 90 degrees towards it, as you say, but it's trial and error which way you turn. Hardly the most efficient method of navigation.

Here's my thoughts; I wonder what others think:

VFR, you ought to be able to revert to your map+compass at any point on the flight if you experience a multiple navaid failure.

IFR in IMC, you should immediately tell ATC of the failure and ask for assistance. They will quite happily vector you in these circumstances, I'm sure.

IFR in VMC is a slightly more tricky one, where it might be reasonable to expect the flight to continue VFR by reference to map+compass in some cases, but not all. The type of aircraft, the recency of the crew's experience of VFR flight, whether a VFR chart is being carried and the type of airspace the failure occurs in and the weather conditions on the day are all factors which immediately spring to mind.

If the DME really is the only method you have of navigating, bear in mind that lots of DMEs are located on airfields with ATC, and lots of ATC units have VDF, in which case you can ask ATC for a QDR or a QTE and plot the bearing and range on your chart for position fixes. Another method of getting a position fix is to plot 3 different DME arcs and find where they intersect, although this is probably not for the faint-hearted if in tricky weather conditions, especially without an autopilot.

And finally (because if I don't say it, someone else will), you could just refer to the backup GPS you always carry!

Sounds like you were pretty switched-on when dealing with your multiple failures, though, so well done. I'm pretty sure I wouldn't have been able to come up with any of this under pressure in the air!

FFF
-----------------

[vancouv]

whether a VFR chart is being carried

Slightly off topic, but isn't it compulsory to carry a VFR chart, even if you're flying IFR?

[HR200]

Slightly off topic, but isn't it compulsory to carry a VFR chart, even if you're flying IFR?

I wont say this for definate, but I do not think it is compulsory. I may be wrong however, good question. I would be interested myself if im incorrect.

Andrew (HR200)

[IO540]

VFR, you ought to be able to revert to your map+compass at any point on the flight if you experience a multiple navaid failure.

IFR in IMC, you should immediately tell ATC of the failure and ask for assistance. They will quite happily vector you in these circumstances, I'm sure.

IFR in VMC is a slightly more tricky one, where it might be reasonable to expect the flight to continue VFR by reference to map+compass in some cases, but not all. The type of aircraft, the recency of the crew's experience of VFR flight, whether a VFR chart is being carried and the type of airspace the failure occurs in and the weather conditions on the day are all factors which immediately spring to mind.

Not sure I agree with the above cases, not least because they could all be legally done OCAS and non-radio. And we all know there is no ATS service OCAS in the UK, well not one that is likely to provide you with a radar service ;) And if you are talking to a radar unit already (e.g. LARS) then loss of nav isn't going to be a problem anyway. Unless of course it is a total electrical failure, in which case (take IMC) where is your handheld GPS?? Short of a global nuclear war, nothing is going to take out ground based navaids and GPS.

In reality, backing up radio nav with dead reckoning means flying every leg with a map and a stopwatch, which few people (navigating using radio nav) will be doing.

Back to the subject, I don't see a problem with using DME alone as an aid to navigation. Often, on long distance flights OCAS, there is no navaid to hand to backup one's GPS. One may have to resort to using a TDME from somebody's ILS to get a position check. It's easy to use a single DME in this way; take several readings. In fact I was told by a Concorde pilot (in the cockpit, after one of the last flights in 2003) that this is how the Concorde INS does its correction (a single DME, not triangulating two of them).

Printed charts are not mandatory (UK or USA) and it is a good question whether VFR charts are mandatory on IFR flights. They can hardly be so, since there are no VFR charts for most of the world (start with Greece and work further along :) ). How many VFR charts would a long haul 747 flight need to carry? Approx 100.

[Tinstaafl]

Tracking (homing, more accurately) using a DME alone was a required part of an Oz IR with a 'DME Descent' endorsement (ie 'approval'. Australia requires a endorsement for each tracking aid you wish to use as part of your instrument rating). A DME Descent is an now obsolete (except, I think, in PNG where OzEx keeps some around for the fun of it... ;) ) type of circling instrument approach that completes a 'DME Homing & Descent' procedure.

It typically was used to meet alternate approach procedure requirements in more remote areas where it would be prohibitive or impossible to carry enough fuel to fly to another airstrip with a useable instrument approach. Distances can be rather large between approaches in Oz... Lots of places would have a single azimuth aid (NDB mostly, occasionally VOR) with or without a DME. Those with a DME would usually have a DME Descent available. OCTA & no radar available either.

Generally the expectation was that it was a backup, to be used if all the azimuth aids fail. The scenario then involves orienting to the DME, homing to it & completing the approach, all using DME alone.

Orienting was usually quite easy. You probably were already tracking to the station so approximate direction was already known. It could also be an off-track diversion in which case an initial orientation would need to be done. I liked to do a rate one turn & watch the rate of closure & distance. Note the distance when ROC is zero, then note the distance when ROC is zero heading in the other direction. Turn 90 deg towards the station.

Homing involved finding the required HDG by assuming a number of 30 deg heading changes in the same sense & noting the ROC (or, for the older DME equipment that didn't have ROC functions, timing over a distance to find ROC). As HDG became more & more towards the station the ROC would increase. When a 30 deg turn eventually resulted in a reduced ROC then HDG was turned 15 deg in the opposite direction. This is 'bracketing'. Periodically you'd have to rebracket. A log of the required HDG & ROC would give enough info to have a reasonable stab at estimating W/V & drift allowance to better track to the aid instead of homing.

The requirement was to arrive within 2 DME of the aid. HINT: Descend to LSALT/MSA some miles prior to the aid. It gives a greater lateral distance to stay within tolerance compared to staying high.

If you couldn't get within 2 nm on first arrival but were within ~ 10DME and saw the distance start to increase a series of turns based on distance would get you within tolerance. If you turned the wrong way then continue the turn into a 270 deg.

The Descent part involved a timed HDG outbound and a procedure turn to head inbound. Can't remember, but it might have always been an 80/260 procedure turn rather than a base turn. Descent would similar to NDB or VOR approaches, either commencing outbound with a limit until inbound or commencing inbound. You had to arrive back within 2 DME after the inbound leg.

If you were smart you'd pay attention to the wind effect at LSALT while still homing inbound & use that information to lay off drift on the Descent's out- & inbound legs. Made getting back into the 2nm circle a bit easier. No fancy turns to compensate as could be done with impunity above LSALT/MSA.

I never had to do one in anger, although often used the DME to meet alternate approach requirements.

[tmmorris]

Wow.

Thank you for that. It makes VDF approaches look child's play!

Tim

[MikeJ]

Nobody has seemed to mention that most DMEs, if not all, also show speed to beacon. Why not alter course 10deg and see if speed goes up or down.
If up, try another 10degs the same way and repeat.
If down, go 20 degs the other way and see what happens.
When you get two near enough the same, with distance decreasing, head half way between. Max speed = going straight to beacon.
Should work properly even with significant cross wind.

I thought the obvious answers of getting vectors from a radar unit or using GPS were avoiding the question - even if undoubtedly the best!

MikeJ

[Tinstaafl]

In the days of the DME Descent procedure Australia had its own DME system operating in VHF frequencies (~200MHz), developed before the current internationally used system was in service and incompatible with it. Some of the earlier airborne equipment (eg VAN5) wasn't capable of displaying GS. Distance was displayed on an analogue scale so timing over a distance was necessary to manually derive closure rate. Timing a change in distance over 36 secs was easiest because closure rate would then be 100 x change-in-distance. Later equipment such as VAN X used an additional needle to show closure rate using a dual scale.

[IO540]

MikeJ

I wonder how well this works, because the GS value is more or less constant for a sizeable range of deviations from the track to the beacon (due to the mathematics of small angles).

10 degrees off track will hardly show up on the GS change, and don't even think about this in turbulence :)

[MikeJ]

IO540,

I too wonder how it would work. Because in a real situation of you would go instantly to GPS (which you would have on anyway) and/or a radar unit (I am rarely not with a LARS), I took this as an academic puzzle. Turbulance is not allowed to the theorists.

Perhaps we should both try it next opportunity!
Nevertheless, the 'small angles' don't seem to be as limiting as you suggest.
At 10deg, the speed difference is 1.5%, which if cruising at 140kts, is over 2Kts down, and my DME is generally within 1Kt of my GPS ground speed, allowing for slant distance when close to the beacon.
Even better if the tests are done at 15deg course change. Just fly half way between those headings where you can observe a reduction. It should get you somewhere close.

Only theorising,
MikeJ

[IO540]

Unfortunately, even a +/- 100fpm vertical air velocity (basically, a very small amount of turbulence or altitude variation) will affect your airspeed (and thus your GS) by a lot more than 1.5%.

I can see it working in totally calm air, on autopilot in altitude hold.

[chevvron]

In evaluations of different data for use in FMS systems many years ago, it was found that mulltiple dme's gave the most accurate guidance, hence the fact that most present day FMS' search for and rely on dme's and only start using VOR info if they can't continually track 6 dme's.

[IO540]

That's very different, because there you have gyro guidance and you are just using DME/DME fixes to instantly fix-up positional errors. It's a bit like me skirting some CAS boundary using the GPS, and dialling in their ILS TDME to make sure I am at the right distance from their runway.

I think DME-only nav with no other position reference is going to be very hard in practice.

[buzzc152]

If I remember correctly, most modern nav systems in airliners navigate using DME/DME fixes as the primary input.

[IO540]

No, their primary nav is inertial (gyros), and the FMS uses DME fixes (or GPS, on the more modern ones) to correct the errors that gradually creep in.

Obviously this works only in range of DMEs. Over the ocean it's 100% inertial, possibly with GPS corrections.

[MikeJ]

IO540,
I find that I generally agree with your comments on the various threads, but not on this one.
Whilst I initially answered to what I thought was a totally academic question, I am sure that the use of the 'speed to beacon' would be easily the best method to get there.
Whilst for the academic addressing of the question I suggested 10deg and 2 knots difference, in reality, use 30deg and 50kts!
As long as the speed to the beacon is kept positive, ie with distance reducing, you will eventually get there, even if its as low as 10kts, and you go spiralling round it. The only time you will not get there is if you let the speed to beacon drop to zero, ie you are passing abeam.

In practice, keep the DME speed display high, say within 50kts of max, keep on top of whether turning left or right increases/reduces the displayed speed, don't worry about 20kt variations, and you will see the ground distance to beacon (allowing for slant range) drop to less than a mile.

This is the mathematics of large angles!
Please try it, I have, hence the delay in my response.

MikeJ
p.s. I do have a degree in maths.