Vectors to Final

October 23, 2008 by Bruce Landsberg

A friend called me the other day to say that several times in last few months he was vectored to an ILS in such a way that his autopilot was unable to capture the glideslope. For those who shun autopilots this may not be an issue, but for many professionals, especially those flying single pilot in high performance aircraft, routine autopilot use for an IMC approach is standard.

To my knowledge all autopilots, when armed for an approach, must intercept below the glideslope for capture to occur. Diving to capture a glideslope makes it tough to stabilize the approach.

My friend called the tower after landing and in the ensuing discussion found that there was some confusion about why this was important. The controller’s guidance says, “For a precision approach, (when vectoring an aircraft to an approach) at an altitude not above the glideslope/glidepath or below the minimum glideslope intercept altitude specified on the approach procedure chart.” That is, they should put you in a position to intercept from below the glideslope.

In the example above, by my interpretation, outside of FREST you should be at 2,400 and between FREST and MEANS it would be 2,100.

One way to make sure that happens is to ask for a “coupled approach,” which will remind ATC that the autopilot will do the honors.

Has anybody else had any difficulty in this area recently?

Bruce Landsberg
Senior Safety Advisor, Air Safety Institute

ASI Online Safety Courses  |  ASI Safety Quiz  |  Support the AOPA Foundation

  • David A. Strahle, MD CFI-AI

    Since I first brought this to your attention, our tower chief wrote back to thank me for raising the issue. The tower personnal were simply trying to make things easier by rounding off the altitudes to the nearest 1,000 feet…not realizing the impact this change would have on our position in relation to the glideslope after localizer intercept. Diving to capture the glideslope is not a good thing and the autopilot refuses to dive as well !

    I have also encountered this problem at other locations this past year and await other pilots response.

  • George Horn

    I have flown Honeywell autopilots (ver.2000) in Hawkers which will capture a GS from above if already captured on LOC and in a descent using either VS, Pitch, or FLC modes.

  • Jeff Johnson

    The 2400 and 2100 are minimun altitudes for the procedure and do not necessarily indicate the minumum vectoring altitude that the controller must comply with.

  • Rudy Holesek

    As a Angel Flight pilot I frequently fly patients into Boston Logan (BOS) and often it is an IFR day. Because my Turbo Saratoge is much slower than most other traffic going into Logan they often give me a vector to the ILS that is not only high but many times a 90 degree intercept. To make matters worse, they ask you to “give me your best speed on final”. Under those conditions, it is impossible in my airplane to do a coupled approach. Once I’m established I couple. The Logan controllers do a great job of fitting us slow pokes into their high-speed traffic, however, on a low IFR day this makes for a very high stress situation, especially when I’m a single pilit.

  • Frank Gorham

    My CFII guru, Joe Romack ( taught the Wright bros etc) and now about to retire, tells me the reason we must approach the GS from below is to avoid capturing a false glideslope. Seems that the ILS glideslope transmitters send out 8 glide slopes, the upper 7 of which are false, not to mention increasingly steep. Just watch your glide slope needle as you execute a missed approach and see how many times it bounces as you go by the transmitter building. 1505538CFII

  • Klaus Kreis

    The Collins autopilot on the CJ2 does also capture from above. The story is right insofar as only a very few a/p capture from above.

  • Ken Trampe

    I fly a lot of 414 , 421 Cessnas single pilot. These are some of the most work intensive airplanes flying especially if you get behind the aircraft. It starts with planning for the approach well in advance. If the controller hasn’t started me down by the time I see close to a 1000 FPM to destination then I ask for it. We also know that at that rate of descent for these aiplanes the airspeed is too high to lower flaps and gear unless you get the gear first to keep your speed down or decend with high speed and level off to slow down and lower the gear. Either will not be the most comfortable for the passengers. If all this is going on while being cleared to intercept the localizer at or above the glide slope intercept altitude at some point inside the APPROACH GATE, you are in trouble. A prudent pilot should be PLANNING AHEAD and situationally aware of the distance to the approach gate and glide slope intercept and if its not acceptable don’t be afraid to ask the controller for a vector to a point farther out. I would rather have a controller a little upset with me than do something unsafe. If you simple drop the APPROACH GATE hint with his coworker or supervisor listening, he won’t be upset for long.

  • Andre Skonieczny

    Cirrus SR22
    Besides being vectored above the glide slope, I also have problems with my autopilot (S-tec 55X) intercepting the localizer when vectored at more then 45 Deg
    It will blow by the localizer and zigzag to establish. In the meantime ATC is issuing
    corrections. It would be nice if ATC took autopilot operation into an account while issuing vectors for approach.
    Andre Skonieczny

  • Peter Cruz

    Being vectored too close to the FAF, and too high to couple the glide slope has happened to me several times. My Bonanza is equiped wiith a Century III A/P which requires at least a minute or so on altitude hold prior to intercept, and will not capture from above. Perhaps controllers think they are doing you a favor by shortening the time required for the approach, when in fact all it does is increase the pilot’s work load.

  • Mike Powell

    Sometimes it is a controller’s lack of ability or understanding of procedures which causes a vector above the glideslope, but sometimes it is procedure based on “letters of agreement” between ATC sectors. One example is KLZU (Lawrenceville, GA) where approach control airspace ended only 2 miles outside the FAF for ILS 25, then center airspace took over. Vectors for the ILS typically brought you to intercept the localizer over the outer marker and well above the glide slope. You could either treat the approach as a localizer approach until GS intercept, or ask to be vectored for another approach, which was lengthy. I no longer fly out of KLZU, so I am not sure if this has changed.

  • David Dodson, MCFI

    Some autopilots and systems (G-1000 and KAP-140, for example) are programmed not to capture from above to avoid tracking a false glideslope. As others have stated, the risk is excessive descent rates for the approach. Pilots should be aware of this issue and disengage the Approach function during vectors to final if above the profile view altitude for the final path to the FAF. If a descent below glide path is made prior to FAF or if the needle is captured at the correct altitude, descent slightly below and reengage the approach function on the AP. Every system has “traps” of which pilots should be aware.

  • walker 777 Capt

    Most apilots on Boeings won’t intercept from above and being vectored in from above happens frequently. A great way to solve the problem is to do a FLCH or flight level change maneuver – simply descending directly to cross the marker at final approach altitude. If you’re alreeady above GS then descending in this manner will A) keep you well above any terrain between you and the FAF, and B) will give you a perfect intercept of the glide slope at the marker (around 1500 AFE). You can dirty up on the way to the FAF if you need to slow the sequence down a bit, but gear down, flaps to a mid position and then with 3 green going to flaps full should still give you a stabilized approach before 1000 AFE which is the normal criteria for landing IFR.

  • Turbine Pilot

    While they are supposed to put you on below the glideslope, in the real world, many times this does not happen. A professional pilot needs to be prepared for all eventualities. If you never hand fly approaches, and don’t thoroughly understand your AP, you can be left in a lurch.

    One tip – if your autopilot system has a CWS button, Control Wheel Steering, you can hold that button in while manually descending the aircraft. Once on or below the glideslope, simply release the CWS and the AP should capture.

  • Robert Zylstra

    The glide slope has echoes or mirrored false glide slopes. The first echo is inverted meaning the higher you go the more it commands up, the next echo is right side up but is about 9 degrees steep! About 3 month ago around 4a.m. with no traffic around and after 4 plus hours of flying, we were vectored so high that the autopilot in the Kingair actually captured the 9 degree glide slope. About 30 seconds later we realized that things weren’t right; fortunately we were still 15 miles or so out and were able to capture the correct glide slope and land uneventfully. I’m glad it was VFR with 50miles visibility.

  • Dave Zittin, CFII

    All the autopilots I use must capture glide below the GS. However, some will allow for above glide GS capture if the pilot recognizes the problem and pushes a button (e.g. the STEC-55. I suspect false slopes can be captured if the pilot pushes the button too late, but I am not sure – always check altitude at the FAF!).

    I like the idea of having an agreement with a controller that informs him I plan to do a coupled approach. However, the pilot-controller glossary definition for Coupled Approaches contains nothing about being vectored under a glide slope, so I am not sure I have the below-slope agreement I am looking for from the controller. Perhaps it is implied? If so, it should be explicit.

    As an instructor, I encounter too many IFR rated pilots who have not read the ship’s autopilot supplement (AFM sect 8) which contains a wealth of information on how to operate the ship’s A/P as well as its limitations (including those that deal with glide slope intercepts).

  • James Reed

    It has been my experience that MOST autopilots WILL couple from above or below the glideslope. It is preferable to intercept from below as that will result in an initialy lower decent rate. There have been some problems with autopilot glideslope coupling when the lateral guidende (LOC) has not yet coupled. I would styrongly recommend that anyone using coupled approaches read closly the autopilot manual for their aircraft.

    Jim Reed
    Chief Pilot
    Berry Construction
    CE-500, CL60, CE650, DA50

  • David Williams

    Recently, while undergoing a BFR and Cirrus annual refresher, after an ILS into KCPK, we climbed to 5500 ft and went west about 15 miles for some airwork. KCPK is near sea level.

    We had not switched our Garmin 430 back from VLOC mode to GPS. Suddenly, the localizer needle went live as did the Glide Slope needle, both of which were centered for a short period of time before we switched to the GPS mode.

    It was a good demonstration of the false localizer and glide slope indications. The pilot must always matain situational and spacial awareness

  • Jeff Burl

    This is a little of a tangent to your question, but may be of interest to some of your readers. The reason that we intercept the glide slope from below is that there are side lobes above the glide slope. Intercepting from below guarantees that the correct glide slope is intercepted. I find it instructive to have advanced instrument students track the first side lobe (false glide slope) above the actual glide slope because it forces them to find new power settings (like in a strong tail wind). I think that the first false glide slope is 9 degrees, but don’t quote me on this.

  • Rick

    >> so I am not sure I have the below-slope agreement I am looking for from the controller<<

    FAA Order 7110.65, “Air Traffic Control”, paragraph 5-9-1 has a note which states:

    “A pilot request for an “evaluation approach,” or a “coupled approach,” or use of a similar term, indicates the pilot desires the application of subparas a and b.”

    Subparagraph b states:

    “For a precision approach, at an altitude not above the glideslope/glidepath or below the minimum glideslope intercept altitude specified on the approach procedure chart. “

  • Len Register

    Hi Bruce, I briefly spoke with you while I was preflighting a citation in Asheville NC a couple of months ago. I am routinely vectored onto the ILS 34 at AVL above the glideslope. I normally plan to hand fly approaches into AVL or at least hand until the FD has ‘captured” the glideslope. In fairness to the controllers in Asheville, high terrain southwest of AVL probably prohibits them from getting me down without vectoring me across the localizer to the southeast. Since I always arrive from the SW, I prefer the slam dunk to the extra vectoring as long as it’s expected. I doubt I would accept a clearance like that in unfamiliar high terrain.

  • Gerard Field

    First of all, it’s nice to have an interactive site to talk to people like you about issues. Most of the timethe communications between experts and readers is just one way.

    About vectors to approaches. At some airports, approach will usually put me on the correct (published) altitude for the FAF or GS intercept. But my experience is that many approach controls (not the towers-the tower just gets you at the FAF) seem to frequently make stuff up. A lot of times they keep you high and you’ve got to slam dunk to get to FAF. Sometimes I ask for lower and sometimes they give it and sometimes they dont . Sometimes they take me far enough beyond the FAF to get well established and sometimes they turn me in right at the FAF. Sometimes they give me a downwind parallel to final and sometimes a converging path leaving no base leg. Sometimes they vector me to final course OK, and sometimes they take me through the course and way beyond and I have to come back although they usually give me a suggested heading to re-intercept. All I know is that when I am on vectors to an approach to an airport, and in the hands of approach control, they can and will take me anywhere around that they want to depending on traffic or what mood they’re in. I dont think I’ve ever had the same vectors twice in a row. That’s one reason that the picture display on a GPS is great. It really helps keep your orientation to an airport.

    Gerard Field

  • Bram Tilroe

    I am a retired Canadian Air Traffic Controller. 33 years as an approach controller and 6 years as Tower Chief. I am also an active IFR pilot and aircraft owner. The Canadian regulations are very clear and unless a pilot asks for a close in vector the controller “shall vector an aircraft to intercept final at an angle of 30 degrees or less and 2 miles or more prior to glide path intercept from below” Anything less or from above was not acceptable and subject to reprimand.

  • Richard Phillips


    I am an operations supervisor at GSP ATCT. The approach plate excerpt you gave is from the ILS RWY 5 approach to SPA (Spartanburg SC), which is within GSP airspace.

    Of course, the requirement to intercept the glide slope from beneath is to ensure that the pilot has the opportunity to fly a stabilized approach. However, I disagree with your interpretation that the pilot should be at FREST at 2400.

    Notice that the runway 5 threshold elevation is 793 (from Jeppesen), the threshold crossing height is 48 AGL (or 841 MSL), and that the elevation of the glide slope at MEANS is 2067. A little math shows the descent rate on the glide slope to be 323 feet per NM. Taking this out a little further, the elevation of the glide slope at FREST is 3424. (According to the AIM, paragraph 1-1-9d, glide slopes are normally usable out to 10 NM from the antenna.)

    FAA order 7110.65S, paragraph 5-9-1, requires controllers to vector arrivals to intercept the ILS at an altitude not above the glide slope. This altitude will vary depending on weather, traffic, minimum vectoring altitude, and pilot requests, but an intercept at FREST at 3400 is perfectly acceptable.

    By the way, the minimum vectoring altitude for this approach is 2500.

    Hope this information helps your readers. The documents above are available on the FAA’s website.

  • http://NA Robert W. Smith

    I have not had problems, but have noticed times when the autopilot KFC150 would not capture glideslope. Your explanation helps a lot for future flights.

    I was on a flight last year with another pilot not familiar with the term “Vectors to Final” and he made a critical mistake by turning inbound at the final approach fix. He was cleared for the ILS approach, not vectors to final. ATC was expecting him to go outbound for the proceedure turn when he turned in at the fix. He was given the phone number to call when landed!!!

    Thank you for your explanation of glideslope capture.

  • travis holland

    This situation occurs with great regularity by spokane approach at ksff. The mva on the north side of the approach is rather high, but this is a recurring problem by approach not wanting to vector far enough out or onto the south side of the approach. I have flown ils approaches in dozens of countries worldwide this year and the problems of failure to vector under the glideslope seem to be only with usa controllers, especially in the west. Any assistance from aopa in this area is appreciated.

  • travis holland

    This situation occurs with great regularity by spokane approach at ksff. The mva on the north side of the approach is rather high, but this is a recurring problem by approach not wanting to vector far enough out or onto the south side of the approach. I have flown ils approaches in dozens of countries worldwide this year and the problems of failure to vector under the glideslope seem to be only with usa controllers, especially in the west. Any assistance from aopa in this area is appreciated.

    However all general aviation autopilots I am familiar with allow gs capture from above (kfc 150 kfc 225 kfc 200 kap 140 stec 55) using appropriate pilot technique.

  • Carl Nuzzo

    I flew fighters for 20 years in the USAF and do full time flight instruction for USAF and civilian now. The F-16 flight director would not give pitch steering if we intercepted the glidepath from above. USAF procedures require pilots to descend to Glide Path Intercept altitude as soon as cleared so they intercept the glide path from below, although few USAF aircraft have coupled autopilots.
    Maybe your friend was given an assigned altitude then cleared for the approach, in which case when on final he could descend to the GPI altitude. That is common.
    Also, contorllers instructions vary according to the weather. For example, on a VFR day they can get you on final at the FAF, on an IFR day they have to get you to final 3 miles before the FAF.

    Hope this helps,


  • Mikel

    Routinely while flying into MCO on the 18R ILS you are required by ATC to remain above 2500′ until over Orlando Executives ORL VOR. This puts you about 300 feet above glideslope, and it can be a little tricky getting back down to the glideslope, especially in a transport category jet. What’s interesting is that MCO approach will require this even when in solid IMC.

  • Bruce Landsberg

    Richard at GSP & All…..

    Thank you all for a great discussion. I hope the exchange is helpful – you just never know with blogs.

    I had the pleasure to be at SPA a few weeks ago during the only rain event I think you’ve had in several months. The GSP controller did a great job of vectoring and we intercepted outside of FREST, as I recall, at 3,400. In any event – point taken and it’s good that we all understand how the hardware works and to discuss with ATC as needed.

    ASF has developed a very good working relationship with NATCA and with some of the FAA’s management which facilitates open discussion without anyone getting in trouble.

  • Juan G Mendez

    A productive discussion and deserves some thought. In my experience of over 35 years of flying USAF Heavies and Corporate Jets and Turboprops, plus my personal Mooney, starting from above the glideslope is generally not a good thing. There are of course exceptions, but when being vectored we are less likely to miss the approach or impact the terrain if vectors are given so that an intercept occurs below the GS. The real question is not the Autopilot, but rather do we know where we are with respect to the published GS intercept altitude and DA or DH, or are we just trusting the controller. I have had many experiences wherein I was vectored to an intercept with the LOC while still above the GS. We pilots must remain situationally aware to avoid getting behind and requiring unsafe rates of descent to catch the GS on the way down. We could say just don’t do it, but occaisionally we have sufficient room to make it work safely. Situational awareness and familiarity with automation are the keys. I would also help if more controllers flew airplanes like in the old days.

  • David Heberling

    I fly Airbus 319/320 for business, and a V35B Bonanza for pleasure. Like the previous message, situational awareness is key to a successful outcome, especially when flying airplanes. Who is flying the airplane, you or the controller? You should never let a controller put you into a situation that you feel is unsafe. Plus, you always have an out…declare a missed approach and try again. Above all else, be stabilized by 1,000′ AGL.

  • jim denike

    one of the best preventives for avoiding the snare of a false glideslope is to recall the rule of “300′ per NM”. Say the airport touchdown elevation is 1500’MSL and you are precisely 3.0 NM out from the runway, you should be at an altitude of 900′ AGL or 2400′ MSL. If you’re much higher and you’re showing an extraordinary rate of descent to stay on glideslope indication, you’ve already suckered in. Have option of transferring to a LOC only approach or a missed. Never a good idea to play catch up here. ATC will gladly allow a vector or a 360 on final, traffic permitting.
    DoubleJay Aircraft

  • Marty Coddington

    Gee Bruce, you really know how to stir the emotions. While you know a bit of my background let me tell the others that I was a controller, ATC instructor, and Air Traffic Supervisor for 26+ years. I then went to flying and was a regional turboprop captain for 10 years. Total instrument flying comes to 52 years.

    Thess abuses can be experienced in the NAS. Some facilities are excellent and you seldom or never see it. Other facilities don’t seem to care how unprofessionally the work gets done, or how little the controllers understand of why their techniques are important.

    Yes, some autopilots will intercept from the top but Mr. Gorham gave the best answer as to why this should not be done. You will never get a false glide slope coming from the “underneath” intercept. If you decide to accept the intercept from the top, you must carefully monitor your altitude vs your position and if your altitude does not match the published altitude (2067 @ MEANS), you must immediately disconnect the glideslope coupling or the whole autopilot because you are not flying a legal glideslope. You may be able to recapture the localizer, but whether or not you do is not the point. The point now becomes that if you continue following the localizer, YOU MUST USE LOCALIZER ONLY MINIMUMS !

    Reading all the thoughtful, technical, and wise comments indicates that our skies are full of really great pilots. For those who are accepting the abuse, please stop. Ask for a vector to intercept at or below the glideslope with an angle of 30 degrees or less (FAA manual JO7110.65S, paragraph 5-9-2). All of us will be better off if it is done the right way. The FAA asks us to fly our machinery the right way and we should not hesitate to ask for equal handling from ATC.

  • Christa

    While on my recent instrument checkride, ATC vectored me to about a mile out from the FAF to intercept the ILS. Problem was– I was over 800 feet high at that point. Even once I took the power all the way out, I couldn’t catch the glideslope. Given that I was on a checkride, I didn’t want to blow my airspeed target, but I really wasn’t sure what to do in that situation. I reached the MAP for the localizer, and went missed at that point, never having gotten closer than full deflection on the glideslope. Given what others have said– it sounds like I was correct in assuming that ATC really should not have put me in this position. At that point, is the correct action to still try and capture the glideslope? Or should I have called ATC back and alerted them to the issue? Luckily- the DE also thought ATC had screwed up my approach and did not hold it against me on the checkride; I still passed without an issue!

  • Bruce


    My opinion, and there may well be others, you should have advised the examiner that you would not be able to stabilize the approach, given the set up. Then politely call ATC and ask for a new approach as Marty has counseled above. As you get more experience, these miscues can sometimes be caught early enough to sort out. Do not let ATC push you into something that is uncomfortable. It’s rare but it happens.

  • Mark Smith

    Quite often I have been vectored to an ILS with a quick last minute turn to make the “within 30 degrees” of the localizer and as much as 1000 feet above where I should be to intercept the glideslope. I fly a PC12 always single pilot. The autopilot does not capture the glideslope from above, therefore I have the choice of disconnecting the autopilot and manually initiating a rapid descent to capture the glideslope or to increase the rate of descent on the altitude controller to expidite a descent to capture the glideslope. Neither alternative is desirable in IMC. Generally, if the ceiling is high enough that I will break out well above minimums, I will recapture the glideslope (usually manually). If we are close to minimums I will ask to be vectored around again at a lower altitude to intercept the ILS. A stabilized approach usually comes out to about 500 feet/min, having to go to 2000 feet/min to salvage an approach is not a safe alternative.

  • Mark Smith

    Please note that on my posting, I have the option (with weather above minimums) and without a useable glideslope indication, to transition to a localizer approach and keep that as an option. Then I still must make sure that the approach is stabilized at least 500 feet above minimums.

  • Steve Milberger

    Part 135 Jet Pilot – SoCal appch routinly vectors and clears you for the appch well above the glide slope “slam dunk” leaving you to perform a rather agressive dive to intercept. Very poor controller performance.

  • Ella

    I really liked your blog! i read 4 others that are on similar subjets, but they domt update very often, thanks.

  • David A. Strahle, MD, CFI-AI


  • preventing grey Hair in Men | CUREFORGRAYHAIR.NET

    There is no position avoid white hair young age in the international arena and further its
    economic and foreign policies this is what I do. Is it really statesmanlike to dye your hair!

  • Kfc Offers

    Just what do you think of this circumstance with my sweetheart? Essentially we went to an ice cream spot and I asked about a voucher and they didn’t take it so I asked if they took coupons and if they usually can be found in the mail.

  • Curve Fever Hack

    Hi, Neat post. There is a problem with your website in internet explorer,
    would check this? IE nonetheless is the market leader and a big section of folks will leave out
    your magnificent writing because of this problem.

    Also visit my site; Curve Fever Hack

  • OilKilm

    Physician Online Prescription Tramadol Cod 717 Tramadol Online Rash Tramadol order tramadol online no prescription. Tramadol Cheap No Rx Overnight Pee Drug Screen And Tramadol Ultram Long Term Effects Tramadol Mixed With Xanax Ultram Tramadol . Generic Tramadol Ultram Tramadol Weight Loss Drug . Tramadol Irrigation Tramadol Visual Studio Ketorolaco How Much Tramadol Nausea Vomiting Medscape Tramadol Tramadol History Tramadol Medication Fedex