"Interesting" instrument approach

[Shunter]

On my way back to Leeds today, sat at about 3500ft in IMC I asked if I could track in to the beacon for an NDB/DME approach. Usual routine is a simple vectored ILS, but since I haven't done an NBD for ages I thought, why not.

This was accepted, so I commenced tracking in, only to be asked if I could accept vectors. Scratching my head a little, I said ok, thinking he must be wanting to vector me around other traffic. This carried on for quite a while, and looking at the ADF/DME it was fairly apparent I'd been vectored to essentially a downwind position, nowhere near the procedure track.

Anyway, eventually he vectored me to somewhere roughly on the base turn and I was instructed to continue with the procedure. Needless to say this left me a little out of sorts and I made a complete arse of it, to the point where if I hadn't just broken cloud I would have gone around.

Is this common? I've done quite a number of NDB approaches and never had this occur before.

[2close]

Controllers swapped over, possibly, with a bit of confusion over what you'd asked for.

Every NDB I've asked for has always been procedural, by the IAP. I've never been vectored to the base turn for a Non-Precision Approach but I'm sure there are those more experienced than I who have.

[tmmorris]

I've been vectored onto finals for the NDB app at Cambridge - they asked if they could because they could then maintain separation by positive control and hence accommodate more traffic. So I went beacon outbound as normal, was then asked if I would accept vectors, then vectored onto base and then given an intercept to final approach.

Tim

[BEXIL160]

Used to be very common in Europe (Spain in my experience), and I believe Southampton sometimes vector to the NDB...

Last time I personally vectored traffic to the NDB was when the 08R ILS was withdrawn at Gatwick (mid 90s?) All traffic was vectored to a closing heading and told to report established on the inbound track.

Curious to watch the "Iberian Penisula" airlines nail every approach, while the Air UK a/c (and other UK carriers) made S turns to one side and then the other until visual :O We guessed at the time that the "foreigners" had more practice on their day to day routes

BEX

[jamestkirk]

Yes. It is not un-common to be vectored on a NPA.

[Shunter]

I never even got near the beacon. I was vectored north of it by 3.5nm. Just as I was thinking, "WTF?!", my well-qualified accomplice pointed out that we'd blasted straight through the localiser. It happened to be tuned in, but being an NDB/DME approach I hadn't included it in the scan. At this point there was no way I'd be within 30deg for a non-prec approach so since I'd broken cloud I just trundled on in visually still thinking, "WTF?!".

[windy1]

Like TMM, I too have been offered vectors to NDB at Cambridge. I had just started outbound for the 05 NDB Procedural and didn't ask for it, but was happy to accept as a new experience. ATC positioned me at base turn altitude on a 30deg intercept to the final approach track.

[Spitoon]

Usual advice - go and speak to the controllers and ask what they do and whether something went awry today.

It's not an uncommon procedure but there's not a lot of guidance to controllers on how to do it - and probably less for pilots. When I first went yo an airport where it was done I went and spoke to some instructors and pilots to see what worked for them - general consensus seemed to be thay they wanted to be on a closing heading of 30 deg or less (ideally 15-20 deg) at the level assigned for the end of the base turn or whatever, and at a greater range than the base turn ended - final reques was a range check when I turned them towards the FAT. All of that is pretty easy to accommodate - certainly in comarison to trying to vector and sequence other traffic around a procedural approach!

[IO540]

At Corfu LGKR one gets radar vectors to the inbound of the VOR approach.

[moggiee]

There is no reasomn at all why the controller can't vector you to an NDB (or VOR) FAT - just as they can vector you to an ILS.

Unfortunately, there is a training mindset in many FTOs that says "NDB approaches must be procedural".

[Shunter]

Well that's fair enough. I asked simply because I didn't know and I've never experienced it before. ILS is fine, you can easily see what you're doing. NDB, despite being simpler in principle is actually more difficult to fly, requiring more than just glancing at a gauge to execute. When you fly a standard procedure you know where you are; outbound leg, base turn, final. Being vectored I was unaware of the intention; did he mishear and was vectoring me for the ILS? When I finally heard, "further with the procedure", I could have used a few seconds to feed the DI/ADF info into my brain, but by the time I figured out what he was doing it had all gone pear-shaped.

Oh well, chalk it down to esperience I guess.

[moggiee]

You live and learn (hopefully). As you say, good experience.

[Curtis E Carr]

Did a radar-vectored VOR/DME approach only this morning.

Not uncommon.

[mad_jock]

It is very common at large airports to be vectored onto a NPA when the ILS is out. The have been quite a few cock ups in rather large aircraft doing them. The one at Birmingham when an airbus got down to mins with 10 miles to run springs to mind.

The NDB approach at LBA is rather special. You will please to know that many an aspiring commercial pilot has made a right royal arse of that one. It has several features including the fact that it bends which make it memorable for anyone who has used an EXAM call sign there.

The trick with these is to always insist that they decend you to the platform height on the procedure. Don't except lower or higher. The reason for this is because the work load of trying to figure out where the top of decent is when your not at the platform is quite high. And if you remember always ask for them to vector you on with a couple of miles to spare on final to get your tracking right before decent. Its a high work load situation for a multi crew commercial aircraft if one is sprung on you short notice, never mind single crew.

[matspart3]

You can be vectored to the initial, intermediate or final approach fixes at the appropriate platform height/altitude. You should have been told which bit the controller was aiming at and should have been given a closing heading for the relevant leg.

[bookworm]

You can be vectored to the initial, intermediate or final approach fixes at the appropriate platform height/altitude. You should have been told which bit the controller was aiming at and should have been given a closing heading for the relevant leg.

Really? MATS Pt 1 seems to suggest that radar vectoring to final approach track is the only option.

[mad_jock]

Mats that is best practise. But with alot of controllers with limited experence in procedural control if any Just don't realise the problems that occur if you don't. Unfortunatly I don't think its being taught anymore.

They have their cunning plans which work great with vectors for the ILS. But when used for vectors for NPA they can cause problems. Its lack of appreciation more than anything.

This is going slight out side the realms of Private GA flying but illustrates the problems that experenced commercial operations have.

Alot of companys have a different set of SOP's for flying NPA's to ILS's

On an ILS you can configure while decending. NPA's you have to be fully configured before the final decent point back to Vref + 10 which can be as low as 140knts. So vectors to a 10 mile final at a none platform height with top of decent at 6 miles and speed control of not less than 160 to 4 is a bit of an issue.

From a light aircraft point of view it is the same, by far the easiest and safest way of flying these things is by flying them fully configured for landing. Anything which deviates away from the published procedure increases the work load 10 fold.

[FlyingForFun]

Being vectored for a non-precision approach is indeed very common.

When I flew at Blackpool, it was standard practice for all instrument traffic, including airliners, to be vectored onto the NDB for runway 10. (No ILS on that runway.) Bexil160 mentioned Southampton; I don't believe they routinely vector traffic onto the NDB as he suggests, but they certainly vector traffic onto the VOR approach for runway 02 all the time. (No ILS on that runway either.)

Mad Jock suggests that you should always be vectored down to the platform altitude; this frequently doesn't happen at the airports I train at, where it is quite common to be vectored higher than the platform altitude. In this case, once established on the final approach track, you should be "cleared to descend with the procedure", which generally means you can descend to the platform altitude, then further at the FAF (but check all the notes on the plate to make sure this is the case; the VOR approaches at Southampton, for example, have a different descent profile if you are vectored straight in). Having said that, on my last IR revalidation, I had a CAA Staff Examiner in the back seat, and during the debrief he suggested that, if cleared to descend with the procedure from a long way out, it may be better to maintain your height until a little closer in, and descend in order to be at the platform altitude a mile or so before the FAF.

I do agree with Mad Jock that you should not be vectored below the platform altitude. Although I don't have any reference to back this up, I do recall a thread somewhere on PPRuNe from several years ago, where a commercial pilot said that he'd been vectored to an NDB approach below the platform altitude, and he was asking if this was common and how it should be flown. The unanimous reply was that this should not happen.

I would suggest, as well as speaking to the Leeds controllers, having a chat with an IMC or IR instructor if you are still in any doubt.

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

[mad_jock]

FFF this one of these ones where local interpretation, Nats and ICAO differ on.

ICAO if you are cleared for the procedure or to decend with the procedure you can go down to procedure alt from the point that they clear you for the procedure.

Some UK airports (none NATS) this will bring you into conflict with other aircraft. Thier local training training team has decided that thats the rules and they are right and everyone else in the world is wrong. I have seen Captains on the phone to the SATCO livid after a discussion about an RA due to decending as you suggest and it ended up with the usual opposing MOR's being filed. I did notice the next time we operated into there they had changed the clearance slightly telling you to maintain level until beacon inbound.

These days I refuse to do it. I don't care if I get spun out of the sequence. Either they clear me down to the platform alt before the top of decent or I don't play. I will always ask for it, if not given by base. And I have been asked about it by Flight ops inspectors. My reply of I am not :mad:ing about doing a NPA on the fly away from the plates in crap wx was received by a smile and a quite right mate thats why they are called air traffic service. We have more than enough work to do on a ****ty day dealing with a high work load approach than compromise safety by having to make stuff up as we go along.

Edited to add I think it was myself that started that thread. And I had a very productive discussion with one of the training team at MAN, who decided that it was good practise and would deal with the issue.

[flybymike]

MJ puts the whole "procedure" into context in his usual inimitable way...:)

[FlyingForFun]

Can't disagree with anything you're saying, MJ, about not wanting to get caught out in an unfamiliar situation in crappy weather.

However, with regards to your example, where:I did notice the next time we operated into there they had changed the clearance slightly telling you to maintain level until beacon inboundI don't really see how that's different to what I said, which was:In this case, once established on the final approach track, you should be "cleared to descend with the procedure"I've added some emphasis to show that what I'd expect (descent once established on the final approach track) is exactly what you'd expect (to maintain level until beacon inbound), except that you'd like a more explicit descent clearance to the platform altitude.

Certainly, though, asking for clarification from a controller is always the only sensible course of action if you're in any doubt about exactly what it is you're cleared to do, I'm sure we agree on that!

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

[stillin1]

Shunter,

Like the man said (Spitoon) - good question - wrong person.
Unless I have missed something! Why did you not ask the ATCO after landing? Since it was so recent an event, you may still be able too.

[mad_jock]

The original RA occured with a conflict between transiting traffic 1000ft below at 10miles with a top of drop at 5 miles when we were cleared to decend with the procedure we were at 12 miles on the final approach track.

For my sins I have something lilke 2000 hours now flying a manual turboprop, about a quarter of that in some pretty horrible wx. If by sharing my experence I can save even one person from having to use some of the bag of luck, I will be a happy man.

Saying that though you have to use some PIC judgment. If the cloud base is 3K and the platform height is 1.5k and your more than likely to go visual you don't make a fuss about it. You know the controller is just using the approach to allow him reduced seperation minima. If the cloud base is below the platform alt I would recommend you do it by the plate.

[IO540]

MJ, can you please clarify the way you would handle this.

Let's say you are on your way in, or perhaps already in the hold.

The ICAO phrase is 'cleared for the XXX approach' and you can then descend to the published platform altitude as you wish.

In the UK, they don't do that. They give you explicit descent levels, and then all of a sudden you might get a vector which takes you to the localiser (or the VOR/NDB inbound, etc) but the only way you know that they are not going to give you the approach at the published platform is when you get no additional descent instructions before you intercept the inbound.

By which time it is too late to do anything about it, other than go missed.

I don't know if perhaps ATC have another rule in their big book which prevents them vectoring you onto the 'localiser' until they see your altitude has reached the altitude which they regard as the platform.

[mad_jock]

This is the sort of situation which I am on my guard with.

It happens in the UK as well in the none NATs fields.

Teeside is an example where you will be cleared for the procedure and won't hear anything until swapped to tower.

It all boils down to situational awareness, you should know where you are at all times through the vectoring procedure because you are still responcable for missing the ground. Which is why thank goodness there are still so many NDB's on the airfield. A combination of RMI and DME from the field combined with the heading the controller has got you on should tell you when to start asking. ie if they put you on an intercept heading and your not at platform you start asking for it. They may at that point been intending to leave you there. But after you ask they then know what you want and generally get it.

Problems do occur when either they try decend you to below the platform alt or their radar vector minimas don't go as low as the platform height.

The desent below platform is a bit dodgy because in the event of a coms failure you are on a potentially none terrian safe vector potentially with not alot of track miles to sort it out. You just have to climb to MSA and then head towards the standard miss approach and do all the other coms failure stuff. It wouldn' be very nice.

If the radar vector minima don't allow you to decend you have to know what the protected area is of the approach you are doing. I use the out bound limit of the procedure ie if the plate has you beacon out bound to 8 D once your etablished on your inbound track and passed 8D you can then decend to your platform alt. In some cases the platform alt for a A,B class aircraft is lower than the C,D approach. In that case just use the C,D platform alt and its top of decent. There is some debate that if you do that you then need to use the C,D approach minima. I don't, I use the B for my aircraft, I can't see why your minima should change when there is no difference to the approach below 1000ft.

But as all these things its not a fight with ATC. The sooner they know what your requirments are the less stress for both partys. So request it before it becomes and issue ie before they put you on the downwind with 200knts ground speed. Intially they might sound suprised because they have been vectoring big tin all morning which can work it all out with a couple of buttons. But I can't imagine many who would get the hump.

The only hassels I have had is when its been a change in runway causing a major workload in the cockpit and on the ground. By the time the plate have been swapped and the brief started you are already late down wind.

Normally I just put the request after the intial booking in when they tell you the approach to expect. eg "Radar Vectors NDB/DME RW 08 request platform alt before decent with the procedure"

[IO540]

Interesting, thank you MJ.

A GPS is of course wonderful for SA :)

I've been vectored above the platform plenty of times, and occassionally below the ILS glideslope too...

A couple of times I got a vector at 90 degrees to the localiser, and the autopilot refused to intercept it. Had to do a very quick manual turn - this was at LJLJ.

[DFC]

At this point there was no way I'd be within 30deg for a non-prec approach so since I'd broken cloud I just trundled on in visually still thinking, "WTF?!".


Can you clarify what you mean by the "within 30deg for a non-precision approach?

Regards,

DFC

[Shunter]

Yes. That was just an illustrative point. ie., "A non-precision approach is considered acceptable if the inbound track places with aircraft within a 30deg intercept of the runway". As a Cat A aircraft yes I could have corrected it, but it would have been a right dogs dinner. I'd broken cloud and was essentially positioned at the beginning of right base at a distance of 1.5D at <1000ft AGL (the base turn was left hand, btw) so proceeded visually. If I had still been in IMC it would have been a climbing turn direct back to the beacon!

Basically I was put in an unexpected situation I was unfamiliar with and it didn't work out so well. It's the first time I've been vectored on an NPA, in fact the first time I've ever heard of it being done at all. So there you go, you live and learn.

[DFC]

"A non-precision approach is considered acceptable if the inbound track places with aircraft within a 30deg intercept of the runway".


Shunter,

You misunderstand that requirement.

What you are talking about is a design feature.

For an approach to be considered a straight-in approach then the final approach track - the published one - must be within 30 degrees of the centerline. If it is outside that then it is a circling approach and circling minima apply.

When you are flying an approach the figure you need to apply is +/- 5 degrees for NDB and VOR.

You are not established on the required track unless you are +/- 5 degrees of that published track. You can not descend on final approach until you have established +/- 5 degrees and if you exceed that then an immediate missed approach is appropriate.

Regards,

DFC

[mad_jock]

Ahh I think your about to get abused shunter ;)

The tolerances in the UK for a NPA are +- 5deg and as such you can't decend with the procedure until you are within this tolerance.

In the nicest possible way can I suggest you go and spend some money with a gruff old Instrument instructor and top up your ground school.

Some on here will enjoy having a go because of your lack of knowledge. In my opnion it is a bit on the weak side even for a ppl. But your better than most, at least you now know its weak and can do something about it.

[Shunter]

FFS, it was an illustrative point. I'll not bother next time. I know what the rules are, I have several thousand pages of them in the next room. I humbly apologise if my phrasing and context were less than perfect and open to interpretation by the resident keyboard warriors. The chap I was flying with was very well qualified and he was also scratching his head a little. At the end of the day we knew full well that the cloudbase was well above approach minima and there was no danger of anything other than me looking a pratt. It was simply a bit of practice. The reason you practice is to revise and improve. If I had needed to fly an IAP, as opposed to chose to fly one as practice, I would have selected one with which I was comfortable and current.

It was a simple post: I attempted to fly an approach I had not practiced recently, something unfamiliar was thrown into the equation, ATC instructions were unclear, X and Y occurred and things went wrong, why?
Answer received. Thank you. As with everything, it's not difficult to understand once you know about it.

[frontlefthamster]

Shunter,

Let's be practical:

Your problem seems to stem from a lack of understanding of day-to-day instrument flying procedures. If you want to fly a procedural approach, you'll need to be specific about this with ATC, and you'll need to keep your fingers crossed that traffic will permit it. If you want to fly an NDB approach, then you need to know that this can begin with either an instrument procedure, self-positioning, or radar vectoring towards the final approach track.

Unfortunately, it sounds to me as if your training for the IMC rating was, as many are, focussed on procedures, at the cost of experience of genuine instrument flying from A to B.

[mad_jock]

training for the IMC rating was, as many are, focussed on procedures, at the cost of experience of genuine instrument flying from A to B.

To be fair this isn't just a fault of IMC ratings the IR is the same.

Which is why I suggested a gruff old instrument instructor for ground school not which you might have missed any flight training I am sure you can point it were you want like the rest of us.

Its knowing were to point it thats the interesting bit ;)

Please don't get the hump on that your getting lectured to in a we are better than you, and you should know better way. Tiz a problem with the internet I am sure a couple of pints in a pub with everyone on the thread and you wouldn't have felt got at. We all went through these shall we say cock ups at some point and learnt from them. Which is why I am quiet anal about the platform height. Had my fingers burn't big time.

PS FO's (and I did when I was a FO) do exactly the same and get caught in the same way if you don't help them out. When asked afterwards where in the books it tells you to operate like you do, you can't tell them. Its just not published. Which is why some feel its unfair to begin with all these airlines asking for xxxx amount of hours. After you have xxxx amount of hours on a steep learning curve you realise why they want them.

Single pilot IFR is some of the most challanging flying you can do. Alot of big tin drivers couldn't do a raw data single crew NDB approach with any amount finesse. They would do it if there life depended on it, but it certainly wouldn't be fun or a walk in the park.

I think sometimes people forget how generally the least experenced pilots in the country are actually doing the most challanging flying.

[frontlefthamster]

Slightly over a decade ago, my colleagues and I routinely included hand-flown, raw data, one engine inoperative NDB approaches in simulator training on a medium turboprop with which we were involved.

Now, on my present 'big aircraft', we do one hand-flown one engine inoperative ILS approach with flight director every six months, and a raw data ILS on two engines every three years. All the rest of the procedural work in the simulator is with autopilot, flight directors, and FMS.

So, the least experienced pilots also used to receive challenging training, and had to meet rigourous standards.

[IO540]

To add to the soup ;) if flying an NDB approach one is pretty likely to have to make a 30 degree dogleg to land, because of the NDB error caused by coastal or terrain assymetry around the final approach track.

I have an RMI which makes it easy to fly the local NDB approach with the autopilot coupled to the GPS ;) and if somebody watches the RMI swing wildly between about 5D and 3D and write down the figures, the result is not amusing. If doing this down to published minima while actually following the ADF, one has a high workload with big heading changes and then one is going to have some major sorting out to do when visual.

But hey this is an approved procedure, established since Marconi invented radio waves in 1890, can't be wrong and doesn't need a risk assessment like the GPS ones :yuk:

Which is why most smart pilots, flying an NDB procedure, use the moving map GPS as primary (especially to quickly work out the heading required to offset the wind drift) and monitor the ADF here and there.

The "overlays" of NDB etc approaches have been in the Jepp IFR GPS database for years. And, for private flight, no regulation mandates which instrument you should be looking at.

Airlines, I gather, do the same. Last I heard, Ryanair fly NDB approaches on the FMS, checking the ADF at the FAF only.

I agree regarding being within 5 degrees of the track before descending to the next stepdown fix (or to the MDA/MDH) but recently there was a long debate in Usenet on this in the USA. Apparently, there are nonprecision procedures around where you are expected to commence descent immediately after crossing the navaid. If the navaid is an NDB that is sort of OK but if it's a VOR, and you are at say 3000ft, you won't be out of the cone of uncertainty for a good little while, and by the time you get a decent signal you have used up a fair bit of the final approach track (or the track to the next stepdown fix) and this encourages diving down at some rate, making it hard to be in a position to land once visual. In these situations, pilots fly a (rough guess at) wind corrected heading until the VOR sorts itself out. Obviously this won't be done with serious terrain around.

[shaun ryder]

Shunter. Might I suggest that if you are not comfortable or current as you pointed out, that you stay away from the NDB let downs in IMC. Try them when VMC under a hat or something, better off for the nerves and life expectancy.

[mad_jock]

So, the least experienced pilots also used to receive challenging training, and had to meet rigourous standards

They also have the most fun and job statisfaction as well. Pity the pay is so low.

But anyway back to the GA side of things.

Yes I agree with the GPS pervert. There are ways and means of operating which aid the work load which arn't taught. Maybe they should saving each individual from reinventing the wheel.

It also doesn't help by airports being thieving git's charging extra for IFR approaches.

[frontlefthamster]

They're not thieving gits; they're commercial concerns there to make a profit and nothing else. Other States believe there is value in the public provision of facilities for air navigation, and thus subsidise the cost of the facilities. I won't stray into a discussion of the 'values' which led the UK to this situation, but rather, state with absolute satisfaction, how good it felt to leave those sorry shores for my present abode...

[DFC]

I agree that it is not the fault of the pilot in this case. They obviously did not receive the appropriate training or practice to be able to operate fully and confidently in the IFR system. That is simply a fact of life with the 10 hours or less IMC rating training.

Shunter,


I know what the rules are,



At the end of the day we knew full well that the cloudbase was well above approach minima and there was no danger of anything other than me looking a pratt.


Unfortunately you most certainly do not know the rules.

Anyone could end up in the position you were in. With more experience, some would question ATC and be repositioned etc. No problem with that and you are trying to learn from your mistake.

However, you can not descend below the MSA unless you are established on an appropriate procedure or are being radar vectored by ATC and have crosschecked that they are descending you to a correct level. Established does not mean +/- 30 degrees.

I sincerely hope that you wanted to say that the cloud base was above MSA and you were visual (under the hood?) from the moment you were unsure of position or you were above MSA while this was going on.

Otherwise, you were flying round below MSA and unsure of your position relative to obstacles, airspace and other aircraft.........That can happen but requires prompt action to fix it.

I would receomend that you invest in RANT or similar and spend time orientating yourself and then get some instruction from an experienced instructor (not the friend who accompanied you while lost).

---------

IO540,

You need to get your ADF system checked. You also need to understand that


pilots fly a (rough guess at) wind corrected heading until the VOR sorts itself out. Obviously this won't be done with serious terrain around.


Is how all instrment procedures are flown - you fly a wind corrected heading, see what happens to the pointer and if necessary adjust the heading. To do otherwise would be to chase the pointer and in the case of the NDB would require lots of heading changes.

Of course everyone knows that with dip, quadrantal error, deviation and so on, chasing the pointer while doing an NDB approach causes it to swing round quite a lot..........just like you describe.


Airlines, I gather, do the same. Last I heard, Ryanair fly NDB approaches on the FMS, checking the ADF at the FAF only.



No. many airlines use the FMS to provide the notional track and the notional descent path. However, the primary reference for the approach remains the ADF and this will be overlaid on the display. So if the indications exceed the +/- 5 requirement even if the FMS says the track is being maintained then a missed approach will be flown unless the crew are visual or can use another approved aid.


Which is why most smart pilots, flying an NDB procedure, use the moving map GPS as primary (especially to quickly work out the heading required to offset the wind drift) and monitor the ADF here and there.



Smart pilots can work out the wind correction (and the timing correction) using their brain. When flying an XXX approach the XXX is the primary indicator. Everything is simple back-up.

If the back-up says that the primary aid is taking you in the wrong direction then it is not a case of following the back-up it is a case of missed approach and when at a safe altitude with low workload sort out which of the two is having a problem.

If you find that the GPS is correct and thus the ADF is at fault, you can not try the approach again you have to use another approved aid or divert.

I get the impression that when faced with an NDB approach you simply fly an unapproved GPS approach and even if the ADF stopped transmitting you would not notice or care. That is not something that should be recomended to others.

Regards,

DFC

[bookworm]

I'm still trying to work out how "vectors to somewhere on the base turn" and "continue the procedure" is a permitted procedure. Procedural approach or vectors to final. There is no middle ground.

Here's the PANS-ATM version

8.9.4 Vectoring to pilot-interpreted final approach aid
8.9.4.1 An aircraft vectored to intercept a pilotinterpreted
final approach aid shall be instructed to report
when established on the final approach track. Clearance for the
approach should be issued prior to when the aircraft reports
established, unless circumstances preclude the issuance of the
clearance at such time. Radar vectoring will normally terminate
at the time the aircraft leaves the last assigned heading to
intercept the final approach track.

I can't find any MATS reference that permits vectoring to terminate other than on the intercept to the final approach track.