
One of the most basic instrument procedures is the procedure turn, used to reverse course as part of an instrument approach. They have been around since the earliest days of instrument flying—think DC-3s and NDBs. In the modern era of radar vectors and GPS, they are not required as often as they once were, but you may still be called upon to fly one as part of an instrument practical test or proficiency check. While they may seem rather simple, there are a number of pits in which to fall while executing one, both with and without a GPS to help.
Turning It Around
Essentially, a procedure turn has three segments: The outbound leg, the reversal maneuver, and the inbound leg. The outbound and inbound legs are flown by tracking a course, typically a VOR radial, but it can also be a localizer (LOC) or an NDB bearing. The reversal maneuver is done by displacing yourself off the course, turning around, and re-intercepting the course inbound. Figure 5-4-18 from the FAA Aeronautical Information Manual (AIM) shows a generic procedure turn both from above (“planform” view) and the side (“profile” view).

The protected area shown in that picture has two purposes. First, the altitudes in the procedure provide obstruction protection. The amount of obstruction clearance provided varies and is established in the United States Standards for Terminal Instrument Procedures, or TERPS (FAA Order 8260.3, currently on edition G). Protection is based on several factors including altitude, distance from the navaid, nature of the navaid, and aircraft approach speed. In addition, ATC provides protection from other IFR aircraft based on that depicted zone.
Maneuvering zone protection is provided on both sides of the procedure course, so there is no “protected side” or “unprotected side,” but protection is provided further from the course line on the reversal side indicated by the barbed arrow on FAA Aeronav charts to allow room to perform the reversal maneuver. For this reason, it is essential that the reversal maneuver be performed toward the barbed side.
The figure shown here, as well as both Aeronav and Jeppesen charts, shows the classic 45-180 reversal maneuver performed by making a 45° turn to the barbed side, flying that heading for one minute, and then making a 180° turn in the opposite direction back to intercept the inbound course. However, it doesn’t matter how you do the reversal other than performing it toward the barbed side. The AIM mentions several other options, including the racetrack pattern, the teardrop procedure turn (typically a 30° angle), or the 80° ↔ 260° course reversal. While you are free to choose from those options (or any other method you may develop that will stay in the maneuvering zone), I generally use the 45-180 shown on the charts, mostly because the headings are already there, saving me the math in my head while trying to keep the plane on course/heading.
For me, the exception to the 45-180 is when I’m arriving at the anchor fix from the side opposite the barb with more than a 90° turn to the outbound, much like the airplane to the outbound side of the anchor fix in the figure above. Trying to make a turn anywhere from 120° to 210° for the outbound intercept and then stabilize on the course in time to execute the reversal maneuver without exceeding the distance limit gets to be something of an exercise in basic instrument flying. When I’m arriving in that lower-right quadrant in the figure above, I’ll generally use the racetrack entry, turning the procedure turn into an extended holding pattern with a direct entry.

When To Turn It Around
The issue with which I see the most struggles on IPCs and refresher training is how far out to go on the outbound leg before initiating the reversal maneuver. The profile view on the approach chart will always state the outer limit for your maneuvering, typically 10 nm, but as we’ll see shortly, it can be shorter. It is critical to remain in that limit as exceeding it compromises both obstruction clearance and separation from other IFR aircraft. There is no regulatory reason not to use it all, but avgas isn’t cheap, and there may be other aircraft waiting to execute the same procedure or depart that airport, so saving a few minutes on the approach by shortening the outbound leg is generally desirable. On the other hand, we don’t want to be rushed during the maneuver so we end up back at the anchor fix too high for a comfortable descent the rest of the way. How do accomplish both goals?
This requires a careful analysis of the approach, preferably on the ground during preflight planning. One goal I have in that process is to plan so I will not be descending during the reversal maneuver. I’ve seen too many pilots either blow through the limiting altitude, or bust out of the airspace, or simply miss a heading or intercept while trying to do that. My process is to look at how much altitude I have to lose on the outbound leg to reach the reversal altitude before starting the reversal maneuver, and then how much altitude I’ll have to lose from there to the next fix or MDA, and let my outbound leg length be driven by the greater of the two.
As a side note, on procedure turns, you must not leave the reversal altitude shown on the profile view until you are “established” back on the course inbound to the anchor fix. The FAA defines “established” as “To be stable or fixed at an altitude or on a course, route, route segment, heading, instrument approach or departure procedure, etc.” This isn’t particularly specific, but the ICAO definition in Doc 8168—Procedures for Air Navigation Services—Aircraft Operations. Volume I—Flight Procedures is: “An aircraft is considered established when it is: a) within half full-scale deflection for the ILS and VOR; or b) within 5 degrees of the required bearing for the NDB.” That means, at least in theory, if you’re intercepting the inbound and the needle has reached within half-scale (assuming you’ve got it under sufficient control that it won’t blast through and exceed half-scale the other side before you’re stable), you can begin your descent to the inbound altitude.

Some Examples
Starting with the O08 VOR-A, I’ll descend to the 3500 reversal altitude before starting the reversal. At that altitude, I have 1300 feet to lose after I’m established inbound to make that 2200 minimum altitude at ILA inbound so I’m in position for a reasonable descent down to the runway after ILA. In a light single, I don’t want to be descending at more than 500 fpm, so it will take about 2-1/2 minutes to lose that 1300 feet. That means I need to complete the reversal 2-1/2 minutes from ILA. Since the 45-180 reversal maneuver will put me back on the radial about a minute farther out than where I started it, that means a minimum of 1-1/2 minutes outbound before making the displacement turn, even if I’m already down to 3500 before that time is up.
On the other hand, on the SGF VOR or TACAN RWY 20 approach, once I’m down to the 3400 reversal altitude on the outbound, there’s no more altitude to lose on the inbound leg, so once I’ve leveled off and stable at 3400 I’ll start the reversal. Of course, I do want to give myself every chance to be established and stable, so I’ll always go at least a minute out before starting the reversal even if they put me at 3400 across SGF initially.
The 2W2 VOR-A requires a loss of 700 feet on the inbound leg, requiring about 1-1/2 minutes inbound. However, we have the shorter-than-normal 5-nm limit for the procedure turn. Keep in mind that this does not mean you can go 5 nm before starting the reversal maneuver. Rather, you are not to exceed 5 nm from the EMI VOR at any point in the maneuver. Of course, since there are only Category A minima, you shouldn’t be flying this approach at more than 90 KIAS. At that speed, you’ll probably only go a maximum of about 2 nm farther out, so you need to start the reversal by 1-1/2 minutes, which should leave time for the descent inbound. It gets trickier if you’re arriving westbound so Potomac brings you over the VOR at 4000 feet. The three minutes it would take to get down to 2600 would run you out of the maneuvering zone. You’d want to be prepared to tell Potomac that any altitude over 3000 feet crossing EMI outbound is unacceptable.


One last point is that if you’re using your GPS to fly through a procedure turn, you may be surprised by how early it commands the displacement turn, often well before you want. If you ignore an undesired turn command and continue out as far as you like, the sequencing will catch up with you once you turn inbound.


Hi Ron,
Excellent article, but I do want to clarify one thing. In the second paragraph, you stated:
“The outbound and inbound legs are flown by tracking a course, typically a VOR radial, but it can also be a localizer (LOC) or an NDB bearing.”
This is true for the Inbound Course, but the Outbound Course is really a Heading that is the reciprocal of the Inbound Course plus or minus the estimated wind correction.
I guess either one could be true, but in my long career I always tracked the localizer or VOR radial on the outbound leg also. It just helped me have a more precise starting point when starting the procedure turn. Probably it just comes down to personal technique.
For holding patterns, yes, the outbound leg is a heading (preferably adjusted for wind). And the reversal maneuver is also performed based on heading. But for the outbound and inbound legs, you are expected to track the localizer course/VOR radial/NDB bearing. Obstruction clearance is predicated on this.
And the procedure turn should be performed on the side depicted by the chart.
IF you are too high to conveniently lose altitude you can also ask to descend in a holding pattern.
Good review, but with modern avionics the need to fly a PT has all but been eliminated. And if you did, just look at the moving map. As for a hold, also on the electronics, but who has had a hold in the past 10 years. I’ve only had a hand full of holds in 27,000 hours of flying, all with a moving map and an autopilot. Even with the invention of the ancient HSI, the above procedures are dirt simple, and I was the first kid on my block that bought one.
Holds are still very much in use, both as a mechanism to put airplanes on a “sky hook” awaiting their turn in the traffic flow and as a holding pattern in lieu of procedure turn (HPILPT). I probably get put in a hold at nontowered airports 8-10 times a year to wait for the airplane ahead to cancel IFR or to allow an aircraft which has been released to get airborne and in radar contact. And any time the winds are from the east at PHL, their acceptance rate is significantly reduced, and I’ve spent time in holding patterns a few times a year while arriving there in seat 12C. In addition, I probably get 2-3 HPILPT’s a month, especially on RNAV(GPS) approaches, when vectors to final isn’t an option.
As for PT’s, I admit you don’t get them often, but they are still often part of an instrument or ATP practical test or IPC. I certainly don’t sign off anyone for an instrument rating ride without having done them at least in the sim as well as (if available) in flight.
Those of us who fly into airports where the approach control is a distant Center and not a TRACON still do frequently fly procedure turns, holds and other rarities like DME arcs. I would need to travel 200 miles before I could get a vector to final.
Jon
It’s not real often that you get a hold these days, but you better know how when one is assigned. High traffic environments can throw you a curve pretty fast.