Holding Question

[Rob Schaffer]

Hope this isn't an easy one, but I have a question that came up today while flying.

I left Wings field, heading north-northwest towards Pottstown-Limerick (Heritage field) tracking the East Texas Radial. I did a direct entry into the hold at GOOGL, flying outbound and learning that I had some more wind from the North than I thought, and adding correction. Turned inbound, intercepted the Localizer which is the inbound leg. I flew over GOOGL, and decended for the first time down a Localizer course,.. wow, those are sensitive! Ended up making some S's as I went down, to much wind correction, then not enough. Anyway, went missed, climbing turn to 2100, heading 90*. This sets me up for a parallel entry into the hold to reverse course.

As I cross GOOGL again, I turned parallel, then I cut in a little to get centered back on the LOC outbound,.. which is now backwards, but keeps me on track with the wind from my left out of the north. My CFII said I shouldn't do that, since it is a parallel entry.

I said it's a tool to use to keep me at the proper location and not drift while flying parallel.

I continued, made the long left turn, ended up re-intercepting the LOC but going slightly thru from my now tailwind, then flew the inbound on the LOC and down the course again after GOOGL.

So,.. am I wrong? Can I not use the LOC as a tool to keep me from moving off course? I can't find anything that says I can't.

Approach is the PTW LOC 28. I'll post my GPS track soon.

 

[rpadula]

So,.. am I wrong? Can I not use the LOC as a tool to keep me from moving off course? I can't find anything that says I can't.

Approach is the PTW LOC 28. I'll post my GPS track soon.

I don't have any problem with that.

 

[HPNPilot1200]

As long as you remain in the protected airspace (which at 2,100 feet tracking the localizer outbound near GOOGL you should be), that should be just fine. Since GOOGL is a fix on the localizer that would seem logical to me to turn outbound to 096*.

 

[nosehair]

I continued, made the long left turn, ended up re-intercepting the LOC but going slightly thru from my now tailwind, then flew the inbound on the LOC and down the course again after GOOGL.

Your CFII was probably trying to get you to not have to do all that twisting and turning at the end of the minute.

Getting on the inbound outbound while going outbound will keep you from drifting, but causes a need for extra turning to get turned around going inbound.

If you would have stayed off to the right of the inbound course, as the holding pattern is designed to do, you would be in a position to make one turn inbound intercepting the inbound course.

Knowing you had some wind from the left, it would have been good to correct a few degrees, maybe 10, but not to worry, you cannot drift out of your holding airspace in one minute. You may drift a little further away than you'ed like, causing a slow intercept, but it is better to undershoot the intercept than overshoot.

 

[AdamZ]

Well I think it is a parallel entry which IIRC would call for a left turn out upon passing google then a left hand 210* turn back to cross google and perhaps another loop in the hold or inboud. I think what you did worked but I can't see how you reversed course 1 min out of google because the lines merge together. So what would you call that entry? I dunno but it worked.

 

[Rob Schaffer]

Well I think it is a parallel entry which IIRC would call for a left turn out upon passing google then a left hand 210* turn back to cross google and perhaps another loop in the hold or inboud. I think what you did worked but I can't see how you reversed course 1 min out of google because the lines merge together. So what would you call that entry? I dunno but it worked.

Parallel. See my diagram below for some clarification. The recommended parallel entry (AIM 5-3-7) says "turn a heading parallel to the holding course outbound on the nonholding side for one minute, turn in the direction of the holding pattern through more than 180*, and return to the holding fix or intercept the holding course inbound" I was intercepting the inbound course.

This flight had more wind and was bumpy compared to my last flights, so I had some more workload going on. I figured that the Localizer is a tool, since in this case it defines the inbound course, and I could use it outbound to keep my wind correction correct and not drift away from the parallel, possibly causing a problem when I would go to turn inbound. The lower half of my sketch is what I did, but was told I shouldn't do. I intercepted the ETX radial, turned, intercepted GOOGL, turned outbound, but continued the turn till I centered the LOC needle again and held it for one minute, then proceeded with the turn to the holding side, around, and intercepted the LOC to proceed past GOOGL and down the Localizer for the runway.

So I guess my question is, am I wrong, that I cannot do this, because it is stated in the AIM to be on the nonholding side?

 

[Rob Schaffer]

Your CFII was probably trying to get you to not have to do all that twisting and turning at the end of the minute.

Getting on the inbound outbound while going outbound will keep you from drifting, but causes a need for extra turning to get turned around going inbound.

If you would have stayed off to the right of the inbound course, as the holding pattern is designed to do, you would be in a position to make one turn inbound intercepting the inbound course.

I still made one turn, typical of the parallel entry diagram, just did my outbound leg co-incident with the inbound, not offset to the non-holding side.

Knowing you had some wind from the left, it would have been good to correct a few degrees, maybe 10, but not to worry, you cannot drift out of your holding airspace in one minute. You may drift a little further away than you'ed like, causing a slow intercept, but it is better to undershoot the intercept than overshoot.

Good point. Thanks.

 

[John Collins]

What you did is perfectly fine, although more work than required. The holding entry procedures suggested by the AIM are just that, recommendations, albeit good ones. Any method you choose that keeps you in the protected airspace is OK.

Yes, the localizer is more sensitive than a VOR course. Anytime there is a significant crosswind at altitude, you will have a tendency to S turn, particularly as you descend. This is due to the fact that in general the winds change direction and fall off as you get closer to the ground. So the wind correction angle you determined while tracking the course prior to the FAF, will be changing as you descend and require a corresponding change in the wind correction. If you have a GPS that can display the DTK (Desired Track) for the approach and TRK (Track), you can see the wind correction in action and adjust your heading as you descend to keep them the same. Otherwise, you have to be very aware of CDI needle movement and adjust before you get too far off of center.

By the way, under the same conditions, if you are using an autopilot to fly the localizer, it will have the same S turning tendency, especially as you approach the MAP.

 

[rpadula]

This flight had more wind and was bumpy compared to my last flights, so I had some more workload going on. I figured that the Localizer is a tool, since in this case it defines the inbound course, and I could use it outbound to keep my wind correction correct and not drift away from the parallel, possibly causing a problem when I would go to turn inbound. The lower half of my sketch is what I did, but was told I shouldn't do. I intercepted the ETX radial, turned, intercepted GOOGL, turned outbound, but continued the turn till I centered the LOC needle again and held it for one minute, then proceeded with the turn to the holding side, around, and intercepted the LOC to proceed past GOOGL and down the Localizer for the runway.

So I guess my question is, am I wrong, that I cannot do this, because it is stated in the AIM to be on the nonholding side?

I like it and agree that the localizer is another tool for you on this particular hold. Tracking it outbound would help you to figure out the wind correction angle if you hadn't already figured it out on the first hold.

 

[mantakos]

... The lower half of my sketch is what I did, but was told I shouldn't do...

I can certainly imagine a CFII stressing "by the book", to at least impress upon the student what the book says.

In the "standard" definition of a parallel entry I see some intent for you to be a little outside the hold, paralleling the inbound leg, because this helps to keep you from drifting too far in the opposite direction when you make your turn back to intercept the inbound. This also eases the intercept, reducing the likelihood of intercepting too late, or requiring too great an intercept angle, which can result in overshooting.

I'd say that you're working too hard, but that it's difficult to declare what you did as "wrong", as long as you do understand the standard definition of each of the holds, in order to parrot it back during a checkride, and how your maneuver differs from that, and that you should probably do a standard entry on a checkride.
-harry

 

[poadeleted20]

There's no "intent" that you be either on or off the inbound course line during the outbound portion of a parallel entry. The textbook procedure is to turn to the outbound heading, and hold it for one minute without regard for alignment with the inbound course. I went round that flagpole with a DPE once, when he asked my trainee how to enter a hold when arriving on the inbound course from the opposite direction. My trainee said he'd to a parallel entry by continuing straight ahead for one minute, then executing a 210 degree turn to the holding side to intercept. The DPE told my trainee that he was required to displace himself to the nonholding side of the inbound courseline during the outbound leg -- at the fix, he had to turn towards the nonholding side long enough to get some distance from the inbound courseline before turning back to the outbound heading. My trainee got that "deer in the headlights" look, and I intervened, saying that I'd taught him the way he'd answered, so if there was a problem, to lay it on me, not the applicant. The DPE and I agreed to disagree for the test -- he said "I won't hold this against your trainee, because this is a point every IR applicant gets wrong and I always have to tell them the right way to do it." My trainee passed, and I called the examiner's POI at the FSDO the next day. The POI's comment was, "He said what?"

 

[jxdrl]

My trainee said he'd to a parallel entry by continuing straight ahead for one minute, then executing a 210 degree turn to the holding side to intercept.

That's also how I was taught, Ron.

On my checkride, I actually did parallel the course but it was just a matter of how I came in and my initial turn. I would never make it a point to do that though. Holds don't need to be hard. Visualize it, fly to the fix, turn towards the protected side and (wait a minute before) intercept the inbound.

Of course, be able to to explain and differentiate between the AIM preferred entries.

 

[midlifeflyer]

The POI's comment was, "He said what?"

How many times has the FAA had to write in the Designee Update publication about the AIM-recommended holds not being requirements? I recall at least two and that there may have been three.

And here you find a DPE who comes up with the required way to do a non-required maneuver.

My favorite line is, "this is a point every IR applicant gets wrong and I always have to tell them the right way to do it."

 

[bobmrg]

At 2000' agl the protected airspace on the non-holding side is 2.6 nm wide. Unless you had extremely strong winds blowing you away from the holding pattern you could easily stay within 2.6 nm, couldn't you? Just parallel the inbound course.

Bob Gardner

 

[Rob Schaffer]

At 2000' agl the protected airspace on the non-holding side is 2.6 nm wide. Unless you had extremely strong winds blowing you away from the holding pattern you could easily stay within 2.6 nm, couldn't you? Just parallel the inbound course.

Bob Gardner

Yes, I can, and I have done this hold with a typical parallel entry before. The particular day in question it came to mind that I have the localizer as a tool I can use, hence, the question that arose because of it.

Thanks for all the input from everyone. It helped my understanding and also sparked up some good discussion at the airport.

 

[Ted]

I don't see a problem with how you did it. In this case, your 090 radial, due to the wind, had you intersect the LOC before the ETX radial. Had you just blown through the LOC, you would've intersected the radial on the south side of the LOC incorrectly.

Ideally, you would've caught the wind and perhaps corrected a bit further north. Otherwise, seems like good situational awareness to me.

 

[poadeleted20]

And here you find a DPE who comes up with the required way to do a non-required maneuver.

Not only that, but he was wrong about how the FAA says to do an AIM-standard parallel entry.

My favorite line is, "this is a point every IR applicant gets wrong and I always have to tell them the right way to do it."

At that point, I bit my tongue so hard it was bleeding for two days.

 

[Rob Schaffer]

As a follow up, I did holds today at the Modena MXE VOR, direct and teardrop entries, and did them fairly well. Even with the crazy wind today, it's proof that practice makes things better. Keeping the wind direction known and what each adjustment was, and what I should think about adjusting to (inbound crab vs outbound crab) is getting better.

I'm also getting much better at following the localizer without making S-turns all the way down to MDA.

Thanks for the discussion here.

 

[Ted]

I'm also getting much better at following the localizer without making S-turns all the way down to MDA.

Don't worry, eventually you'll be able to shoot it at 180+ over the ground while having a screwed up entry from ATC and still keeping it nailed.

 

[Rob Schaffer]

This morning doing the ILS to Chester County, I hit 74 knots groundspeed inbound at one point,.... 134 knots GS after turning missed back to MXE !! My first entry back into the hold at MXE came up fast!

Today I did the ILS at MQS twice, the VOR-A at Brandywine once, then the GPS 6 back to Wings, circle to land on 24. Below 2000 feet it was a bumpy ride and really made me work, but I had a surprisingly good day and kept those needles pretty well.

 

[Ted]

One idea (for down the road a bit)... once you get the approaches going pretty well normally, try to do some with a stiff tailwind. It makes the standard approaches easier, and there will be times when you'll want to do an approach with a tailwind. Plus it makes step-ups to faster aircraft that much easier.

I seem to recall that PNE/TTN have ILSs that allow you to do a figure 8 of sorts. ILS on one, go missed, make a turn, they set you up for the other ILS, do it, etc.

 

[Rob Schaffer]

One idea (for down the road a bit)... once you get the approaches going pretty well normally, try to do some with a stiff tailwind. It makes the standard approaches easier, and there will be times when you'll want to do an approach with a tailwind. Plus it makes step-ups to faster aircraft that much easier.

I seem to recall that PNE/TTN have ILSs that allow you to do a figure 8 of sorts. ILS on one, go missed, make a turn, they set you up for the other ILS, do it, etc.

Cool, I"ll keep that in mind. PNE/TTN would be good radio work as well bouncing between the two.

 

[Ted]

Cool, I"ll keep that in mind. PNE/TTN would be good radio work as well bouncing between the two.

My instructor and I flew down there a couple of times to do the figure 8s there. It is good practice, because it's high workload, lots of radio work, etc. A good workout. Of course, you already know my philosophy on IFR training.

 

[dmccormack]

My instructor and I flew down there a couple of times to do the figure 8s there. It is good practice, because it's high workload, lots of radio work, etc. A good workout. Of course, you already know my philosophy on IFR training.

The biggest shock for me after earning the IR was how inactive *real* IFR flight was and is.

Sure, you need to prep for various emergencies, but for most of us flying in IMC it's an approach at the end with lots of straight and level before it.

 

[John Collins]

The biggest shock for me after earning the IR was how inactive *real* IFR flight was and is.

Sure, you need to prep for various emergencies, but for most of us flying in IMC it's an approach at the end with lots of straight and level before it.

And don't forget to bring your sun glasses!

 

[jxdrl]

And don't forget to bring your sun glasses!

Pilot: "It's a beautiful day out there today."

Non-Pilot: What are you talking about? It's raining and cloudy.

Pilot: Not above the clouds.


I like the world in the sky.

 

[Ted]

The biggest shock for me after earning the IR was how inactive *real* IFR flight was and is.

Sure, you need to prep for various emergencies, but for most of us flying in IMC it's an approach at the end with lots of straight and level before it.

Exactly. And most IFR flights actually have very little actual IMC.

Yesterday was a good example of a hard IFR day, though. We tried to fly into Linden, ceilings and visibility were too low there, couldn't get the field. Went missed, had to go to Caldwell and shoot the LOC in. TAFs forecasted OVC025, turned out OVC005 at Lidnen, and BKN010 at Caldwell. Then refile and get back out. Lots of working on your feet and being prepared for things to not go as expected, with improvising.

 

[bdlarkin]

The DPE told my trainee that he was required to displace himself to the nonholding side of the inbound courseline during the outbound leg -- at the fix, he had to turn towards the nonholding side long enough to get some distance from the inbound courseline before turning back to the outbound heading.

Ok, I'm confused. Can you help me understand this better?

A hold is a pattern with a "protected" (holding) side and an "unprotected" (nonholding) side. The protected side is...well... protected. Danger lurks on the unprotected side.

Why, oh why, does a Designated Examiner want me to fly my little spam can out where there is danger? There could be a mountain on that side for gosh sakes or some other monster. I know, not really, and there is some safe "area" on the unprotected side, but why go on the unprotected side if you don't have to?

And not to mention I have this perfectly good localizer to help keep me on a line that is "safe".

What purpose does it server to offset the inbound course? To make tracking back on the inbound leg easier? As long as I stay inside the TERP boundaries of the hold what does it matter?

I would have asked him how an autopilot is supposed to achieve that little offset when it starts tracking the magenta line on the GPS....

This flabbergasted me so much that I pulled out the AIM, and sure enough, AIM Fig 5-3-4 depicts the parallel entry a little offset on the non-holding side. I always assumed that was offset to clearly illustrate that it was supposed to track in line with the inbound course line (just outbound) and not actually be "offset". It's hard to draw a dashed line on top of a solid line.

Reading 5-3-3(a) however (emphasis mine):

[...]the parallel entry procedure would be to turn to a heading to parallel the holding course outbound on the nonholding side for one minute, turn in the direction of the holding pattern through more than 180 degrees, and return to the holding fix or intercept the holding course inbound.

It's in there. But WHY? It makes no sense.

If this was truly the intent why wouldn't they put some limit on it? Non holding side within 10 miles, 1nm, 100feet, 0.1 feet, width of a swallows tail feathers (queue Monty Python joke)....

 

[John Collins]

Brian,

There's a holding side and a non holding side of the racetrack, there isn't a protected side and an unprotected side. Both sides are protected, but the protected area is larger on the holding side than on the non holding side. The parallel entry is recommended to be used from 180 degrees from the inbound leg direction to 70 degrees from the inbound leg direction when approaching the holding fix from the holding side. So, with the exception of approaching the holding fix from 180 degrees opposite the inbound direction, and without using turn anticipation, you will always cross over the fix and be on the non holding side before the turn to parallel the inbound leg. Even when approaching from the 180 degrees opposite direction, you have a choice of doing a tear drop or a parallel entry.

It is interesting that the Garmin GNS430W flies the parallel hold entry closer to the way the OP did, but it uses turn anticipation to align itself with the inbound leg in the outbound direction and then when it reverses course, it turns 225 degrees to intercept the inbound leg.

 

[bdlarkin]

So, with the exception of approaching the holding fix from 180 degrees opposite the inbound direction, and without using turn anticipation, you will always cross over the fix and be on the non holding side before the turn to parallel the inbound leg.

Thanks for the clarification of my bad nomenclature, I use protected side to keep myself reminded that thats the site I want to be on. Obviously there are terrain clearances on both sides of the hold, especially for holds that would be on an airway, etc.

My question is more why we'd want to be deliberately offset of the inbound course line. If you didn't have a radial or localizer to track, then sure it might make it easier to just fly the designated heading and not worry that your maybe a little offset to the holding side. But in this case, and in the way the AIM is written it makes it seem that there is some benefit to being on the non-holding side.

It is interesting that the Garmin GNS430W flies the parallel hold entry closer to the way the OP did, but it uses turn anticipation to align itself with the inbound leg in the outbound direction and then when it reverses course, it turns 225 degrees to intercept the inbound leg.

Right, because it KNOWS the fixed point in space that is the Hold point and can do turn anticipation. The AIM says we shouldn't begin our turn until full scale deflection which could be after the station passage (or before, depends on the radio, the antenna's, etc, etc.).

I get holding, I'm just confused by what this DE (and the AIM) seem to think they are accomplishing by specifying the outbound leg needs to be offset from the inbound leg.

I guess in my GPS equipped plane I've always been anticipating the turn, since I have a countdown to the fix on the display. There shouldn't be anything wrong with that, "real world".

 

[John Collins]

Brian,

The DPE's assertion was wrong and the AIM just points out a fact that if you don't turn until passing the holding fix, you will be on the non holding side. If you read Ron's post, he checked with the POI at the FSDO who remarked, "He said what?" when referring to the DPE statement. There is no requirement that you be on the non holding side, and even if you were, wind could drift you in either direction. All this is taken into account in the TERPS to provide adequate obstacle and terrain clearance. The only leg you are expected to use navigation course tracking is on the inbound leg.

 

[nosehair]

Ok, I'm confused. Can you help me understand this better?

OK, I will!

A hold is a pattern with a "protected" (holding) side and an "unprotected" (nonholding) side. The protected side is...well... protected. Danger lurks on the unprotected side.

Well, I sense a little anxiety about the word "unprotected". You put it there. It's not really there, but I can see how it intrudes itself by the use of the word "protected".

Let's call them "more protected" - you can make a wider turn, and this is the side to actually do the holding turns on, but the other "less protected " side has pleeenty of room to make a turn.

So, first off, I wanted to get rid of any residual anxiety you may have over being on the non-holding side.

Because, there is a good reason not to fly the inbound course outbound.

What purpose does it server to offset the inbound course?

In my experience at getting actual holding instructions, at the IAF, more often than not, the approach clearance will come while I am going outbound.

I have learned to treat the inbound course like a one-way street. Don't go outbound on the inbound course.
Many times a controller will have to issue a sudden hold over the IAF because the prior IFR approach has not terminated yet, but as soon as you turn outbound, the other airplane terminates and the controller clears you.

I try to be off-set enough to make a one-turn inbound and I'm on final.

 

[bdlarkin]

OK, I will!
So, first off, I wanted to get rid of any residual anxiety you may have over being on the non-holding side.

It's not that bad I know that the guidelines are "really" big and you'r never quiet rushing to a holding pattern so I'm never going that fast anyway. It's more an affront to my "fly the line" mentality I guess. Even without GPS, there's a line in my mind that I'm trying to track. Here all of a sudden my worldview is shaken when I actually READ that section of the AIM, "with new eyes" if you will.

I try to be off-set enough to make a one-turn inbound and I'm on final.

Good example. The only hold I've gotten for "real" was in the enroute environment (shocked the hell out of me) instead of an approach.

Id probably just go around the hold another time It's only another 2 minutes and some change.

 

[bdlarkin]

Brian,
The only leg you are expected to use navigation course tracking is on the inbound leg.

Yeah but WHY, did the DE think that was correct? Not only correct, but required? Was he just looking for a legalese way to trip up students (and instructors), or was he just blind to the letter of the law.

If I had been asked that on my IFR Checkride I'd have asked him why that the offset is needed, especially when I could track a radial or localizer to keep pointed in the right direction.

Plus, thats not how the GPS and autopilot would fly it.

I know its stupid and not required, especially in the real world. But I would think a GOOD DE would think about what he was asking before he asked that of a student.

 

[John Collins]

Yeah but WHY, did the DE think that was correct? Not only correct, but required? Was he just looking for a legalese way to trip up students (and instructors), or was he just blind to the letter of the law.

If I had been asked that on my IFR Checkride I'd have asked him why that the offset is needed, especially when I could track a radial or localizer to keep pointed in the right direction.

Plus, thats not how the GPS and autopilot would fly it.

I know its stupid and not required, especially in the real world. But I would think a GOOD DE would think about what he was asking before he asked that of a student.

DE's are human and make mistakes and sometimes have weird ideas that can't be supported by the facts or by the regulations. Your question would be a good one to have asked the DE at the time but we can't examine his motives at this late date.

In my opinion, the DE was wrong and has nothing to base his opinion on.

 

[John Collins]

Because, there is a good reason not to fly the inbound course outbound.

In my experience at getting actual holding instructions, at the IAF, more often than not, the approach clearance will come while I am going outbound.

I have learned to treat the inbound course like a one-way street. Don't go outbound on the inbound course.
Many times a controller will have to issue a sudden hold over the IAF because the prior IFR approach has not terminated yet, but as soon as you turn outbound, the other airplane terminates and the controller clears you.

I try to be off-set enough to make a one-turn inbound and I'm on final.

I don't follow why you treat the inbound course like a one way street. If you deliberately offset yourself when outbound to provide sufficient distance from the inbound course so that you can complete a 180 degree turn and roll out on the inbound course, you are using up more of the protected airspace on the non holding side than you need to. Since this offset is based on dead reckoning and does not take winds into consideration, you can drift even further away from the inbound side.

 

[somorris]

Unless you had extremely strong winds blowing you away from the holding pattern you could easily stay within 2.6 nm, couldn't you? Just parallel the inbound course.

I would vote with Bob on this. The way you did it is (I think) okay, but it adds a lot more maneuvering at a time when you have enough other things to do. I usually just start the time, turn to the outbound heading then get set up for the intercept.

 

[nosehair]

The only hold I've gotten for "real" was in the enroute environment (shocked the hell out of me) instead of an approach.

Id probably just go around the hold another time It's only another 2 minutes and some change.

With that in mind, I totally understand your thinking: To you it's all academic, and you are more comfortable being "on the beam". Me too. If I'm not 100% sure of where I am, then I tend to do more of what you suggest - at least keep the needle off the peg.
But after you do a thousand or so holds, you will find perfectly safe short cuts that keep the time down.

And sometimes, when you are fighting to keep it straight-and-level, you really don't wanna make another needless turn in holding. It's 4 minutes, and change, by the way.

 

[midlifeflyer]

Because, there is a good reason not to fly the inbound course outbound.

In my experience at getting actual holding instructions, at the IAF, more often than not, the approach clearance will come while I am going outbound.

I have learned to treat the inbound course like a one-way street. Don't go outbound on the inbound course.
Many times a controller will have to issue a sudden hold over the IAF because the prior IFR approach has not terminated yet, but as soon as you turn outbound, the other airplane terminates and the controller clears you.

I try to be off-set enough to make a one-turn inbound and I'm on final.

I'm trying to visualize how being "on the beam" on a parallel entry would prevent you from a one-turn inbound and just can't.

Outbound, get approach clearance, finish the outbound, turn inbound to intercept the inbound and fly the approach.

What's the difference?

 

[nosehair]

Outbound, get approach clearance, finish the outbound, turn inbound to intercept the inbound and fly the approach.

If you are on the inbound course going outbound, when you turn around inbound, you may not be able to intercept the inbound course before getting back to the FAF, requiring another turn in holding to get established.

I try to offset enough to have room to make the turn inbound and established on course before getting back to the FAF.

Example: Say you are on course going outbound, and the PT or holding side is to the right and you are having to hold a 20 degree crab to the left for a strong wind from the left, and you have to turn downwind to turn around, so you know you're gonna be blown so far south you'll take two minutes just to get back to the fix, and you won't be on course, so....

I avoid that by always off setting enough to make a turn "onto" the inbound course.