Single pilot IFR without autopilot

[AndreiC]

Today was a great IFR training day around here (ceilings 400'-800', no turbulence, solid IFR, good VFR weather nearby if ceilings were to go down more) so I went to do some IFR practice by myself. The first two approaches, flown with the help of the autopilot, were great, rock solid, very happy with them. (RNAVs; the A/P is STEC 30 coupled to GNS430W). Then I thought let's try a few flown by hand. 

Boy, was I in for a surprise. I felt behind the airplane most of the time, occasionally finding myself on a heading 20 degrees off what I wanted, etc. The 4 approaches I tried worked out eventually, but I was definitely not proud of myself. I felt all the time like a juggler who is at their limit because even one simple thing I was doing (turning a knob on the GPS or adjusting the DG) was bound to throw my scan off.

I know what I am supposed to do, go back up under the hood with an instructor and shake off some rust. But is it just this, being rusty? I did an IPC maybe a year ago, and I did very well on it, all without an autopilot. I don't fly actual IFR all that much, probably just barely the 6 required approaches in 6 months, and mostly just going through a layer on my way up or down, almost always with the A/P on.

Advice? How do others handle this?

[Falcon Man]

Andrei, what you reveal is a very common issue that I and my pilot friends deal with. FWIW, here is my experience:

I had this problem right after getting my instrument rating. I struggled for years until I had instructors that were from the Fed Ex method - they utilize the TLAR (That Looks About Right) system. The idea is that trying to fly an approach with absolute precision in anything less than still air is not realistic for most pilots when hand flying. More importantly, the FedEx safety analysis revealed that absolute precision was less safe.

Gradually I became more successful and confident as I I learned that it is OK to be fairly close to the vertical and heading course. Technique wise I learned to use very minor nudges on the controls and to verbally call out my next step. Like you my actual IFR flying is limited. About 80% of my flying is 3-4 hr cross countries in the mountainous west and I hand fly some type of practice IFR approach on every VFR flight (without hood) when allowed by the controllers. I don't ever hand fly hard IFR in my K model, never to minimums and only through layers. Icing is almost always in the forecast in winter storms across the western US and only my prop is deiced.

A couple of times over the years my autopilot (KFC 150) stopped working mid-approach in actual IFR and the hand flying was easier than practice flights under the hood with a safety pilot or instructor in the right seat. Maybe "performance anxiety" makes things harder.

Practically speaking all of my instructors and kerosene burning pilot friends say that hard IFR single-engine and solo pilot flying is a fool's errand (esp. in the mountains). I think the safety record supports that concept.

Thanks for bringing this subject up. I am sure others will chime in and provide their experiences which may be helpful.

 

 

[Ibra]

Thanks for bringing this on the table !

You are never happy with performance of single pilot hand flying IFR

Unless one is flying some usual instruction sector or local procedure at home-base, it always feels “sub-optimal” when going elsewhere. So one have to accept that they will be “behind the aircraft” and “not happy about precision”, no matter how good they are 

I am fine hand flying but those flights require lot of currency and preparation which I can rarely afford on higher minima while being very familiar in easy airspace, while on auto-pilot I can push things further: complex airspace, plates minima, brief and prepare inside aircraft…

I can fly radar ILS in complex airspace usingcoupled (or HDG/VS modes) on autopilot doing 180kias with all config and speed changes on short final doing to typical system minima but on hand-flying I would ask for procedural and 90kias final fully configured at FAF/FAP, stick to easy airspace down to typical circling minima. Same I would not fly long cruise without auto-pilot, the arrival and procedure while being tired of chasing needles for 2h tends to be interesting  

Go-Around and preparing diversion without auto-pilot is another nightmare.

Edited yesterday at 09:00 AM by Ibra

[midlifeflyer]

My best clients are those who come to me complaining that they are concerned they have become too autopilot dependent. When I fly with them, these are the top three things I see, in order if their prevalence and seriousness  

1. Loss of anticipation. The basic, “what’s next laterally and vertically?” that slows things down and keeps us ahead. I wrote about that this past year in IFR Magazine. It’s not only #1 on my list. It is so far above the others it could  arguably be the only one on my list.
.

 

2. Poor division of attention. Autopilots get us used to not having to look at the instruments. We can take our time looking up frequencies, loading approaches, copying instructions. This includes a wide area from the way most of us tend to pressure the controls when we do something else (that’s your 20° off) to not following a “3 second rule” (when doing something else, check the instruments at least every 3 seconds) to fixation to failing to prioritize  

3. Scan breakdown. Loss of basic attitude instrument flying. It’s last and way behind the other two because (a) it’s the fastest to come back and (b) it’s usually caused by 1 or 2. I can’t count the number of times I see a CDI needle moving left and a “rusty” hand flyer turn away from it rather than toward it. But I think that’s  more about poor division of attention and the rushed feeling we get when we don’t anticipate adequately than not knowing how to interpret instruments, 

[N201MKTurbo]

The bulk of my IFR flying was in my M20F which didn’t have an autopilot. It did have a functioning PC system so you could let go of the yoke without it rolling over. I took my ATP checkride in an airplane without an autopilot. The hardest thing to do without an autopilot is dealing with a complex reroute where you have to write it down, figure out where those waypoints actually are and spelled and update the NAV while hand flying the plane. I personally found this a lot easier back in the day of paper charts. There was not as much button pushing, all you had to do was switch frequencies on the nav radio and spin the OBS. And you probably didn’t have to do that for a while.

[A64Pilot]

I hear people who say “I’d never fly IFR without an Autopilot” and I think what fools, what their plan when it fails? Because one day it will.

It’s all proficiency based, I used to fly my Maule without an Autopilot and a Maule isn’t a good IFR platform without issue often to ILS minimums.

But I flew real IFR a LOT, now I won’t fly IFR anymore because I’m not going to put the work into it that’s required to gain and maintain proficiency, plus I don’t do the cross country like I used to that demanded IFR.

My belief is if you have an Autopilot you need to fly without it often enough that your comfortable hand flying. I fear many can’t they never really trained without one.

Edited yesterday at 12:48 PM by A64Pilot

[midlifeflyer]

5 minutes ago, A64Pilot said:

My belief is if you have an Autopilot you need to fly without it often enough that your comfortable hand flying. I fear many can’t they never really trained without one

Agree about the need to hand fly periodically to maintain proficiency. The other is to treat the autopilot as an idiot and do the same anticipation mantras as when hand flying.

Training is a separate issue. You really need both and while some DPEs will allow it's use most of the time, others will limit it. From an ACS standpoint, at least one of the nonprecision approach must be done without the AP.  When I give an IPC, I do a mix. 3D approach is typically a fully-coupled ILS and missed. One 2D approach without either primary flight instruments or autopilot. One 2D approach pilot's choice unless I see an issue with one of the other two.

[Schllc]

25 minutes ago, A64Pilot said:

I hear people who say “I’d never fly IFR without an Autopilot” and I think what fools, what their plan when it fails? Because one day it will.

It’s all proficiency based, I used to fly my Maule without an Autopilot and a Maule isn’t a good IFR platform without issue often to ILS minimums.

But I flew real IFR a LOT, now I won’t fly IFR anymore because I’m not going to put the work into it that’s required to gain and maintain proficiency, plus I don’t do the cross country like I used to that demanded IFR.

My belief is if you have an Autopilot you need to fly without it often enough that your comfortable hand flying. I fear many can’t they never really trained without one.

You know the cirrus training regiment requires you to engage the ap at about 300’ and to keep it on until the absolute last legal moment. 
when I was getting checked out, after each flight they did a review, and the only comment they had was to leave the AP engaged longer on approach. 
I responded that the perspective g1000 was a lot different and when I felt like I was getting behind with the interface on approach, I would cancel the AP and hand fly the approach. 
he said this was wrong and I should never do this. 
I said, my “go to” when on approach is me,  not the AP. 
he didn’t really push back on this, but I could tell he didn’t like the answer. 
while I do use the AP on approaches, if anything isn’t exactly how I expect I cancel and fly it by hand. 
I still believe this is the correct reaction. 

[midlifeflyer]

17 minutes ago, Schllc said:

You know the cirrus training regiment requires you to engage the ap at about 300’ and to keep it on until the absolute last legal moment. 
when I was getting checked out, after each flight they did a review, and the only comment they had was to leave the AP engaged longer on approach. 
I responded that the perspective g1000 was a lot different and when I felt like I was getting behind with the interface on approach, I would cancel the AP and hand fly the approach. 
he said this was wrong and I should never do this. 
I said, my “go to” when on approach is me,  not the AP. 
he didn’t really push back on this, but I could tell he didn’t like the answer. 
while I do use the AP on approaches, if anything isn’t exactly how I expect I cancel and fly it by hand. 
I still believe this is the correct reaction. 

There are definitely those who believe that unless you are hand-flying as a proficiency exercise, the autopilot should always be engaged when not contrary to limitations. That's not a Cirrus-only thing. Gary Reeves who specializes in avionics has been preaching this for several years. 

I agree with you - if in the real world, the AP is not doing what you expect hand fly. The only problem is, with modern systems, if something on the approach is not exactly as you expect, the chances are the error is yours. Sometimes, the pilot expectation is incorrect. Other times, the pilot has failed to set things up properly; Still other times, the pilot has failed to monitor progress and fails to notice the flight plan sequencing isn't where it should be. The system simply not doing what the pilot told it to do happens but is pretty rare in comparison. That's the reason I require a coupled ILS on an IPC. I have seen too many pilots screw a simple approach up. 

 

[AndreiC]

Thanks all for the mix of comments, this is very helpful. Here is my main question. Given the type of flying I do -- about 100-150 hours per year, all of it recreational, I can cancel any flight if the weather does not seem right to me, probably 95% of it VFR, very rarely approaches to minima -- what is a good frequency with which to go up with an instructor to shoot approaches hand-flying?

One option: with what I do I can stay current (barely) by going out once every couple of months when the weather is like yesterday and shooting 3-4 approaches in actual, with the autopilot on (to be safe), maybe shoot one approach by hand and call it a day. Since I don't ever plan to fly actual IFR when I don't need to without a functioning autopilot, this seemed (until yesterday) like a reasonable option. I prove to myself that if the autopilot kicks the bucket in hard IFR I can get myself on the ground by hand, treating it more like an emergency. I would ask ATC to divert to a place with the easiest approach possible (LPV or ILS, never anything without vertical guidance), with good weather well above minima, etc.

Option two: go out regularly with an instructor, enough to stay current on instruments to be able to fly by hand confidently. Of course this would be better, though with the existing availability of instructors it seems hard to do. Also, I found that under the hood work does not feel the same way to me as actual IFR, and I am not confident that if I do everything right with an instructor under the hood, this will translate to perfect flying in actual. (And good IFR weather is not often happening here -- either it is good VFR, or bad TSRA or icing...)

What do people do in situations like these? Should I call it a day on IFR flying, like @A64Pilot? That is very limiting... Is it ok to say I have an IFR-lite rating?

[Hank]

1 hour ago, A64Pilot said:

My belief is if you have an Autopilot you need to fly without it often enough that your comfortable hand flying. I fear many can’t they never really trained without one.

 

1 hour ago, midlifeflyer said:

Agree about the need to hand fly periodically to maintain proficiency. The other is to treat the autopilot as an idiot and do the same anticipation mantras as when hand flying.

My personal attitude is two-fold:

1. In IMC, use whatever assistance you have available. My job is to watch it, anticipate the next move, and make sure everything is as close to perfect as possible. 
2. For practice approaches, make some of them with no help, just me holding the yoke and following the needles. My goal here is to be at least as good as my last practice approach, but probably not up to the level of #1 above. 

If in the clag and hand flying, my goal for that approach is to be close enough to land at the end. Sure, expect a miss on every approach, but really, don't we always land at the end of one, and don't we usually shoot only the one?

Sometimes being off isn't all my fault, like when the new tower brought me in, "maintain 3000 until established," and my vector brought me over the IAF, charted at 2700 . . . Fought altitude & speed the whole way, ended up a little right as I broke out just under 1000 agl, and saw the runway to my left. Asked and was given right traffic, circle to land, because I didn't see the benefit of going missed and doing the whole thing again, likely with another poor vector. Isn't hand flying in the soup fun? Tops were just under 3000, ASOS said 900 broken, I thought it would be easy.

P.S.-- @AndreiC, if you want goggles to feel "more real," go out with a CFII and do approaches in the evening, out in the country. You won't know where the sun is, because there isn’t one, and any peeks you get from behind them wont help you any.

[A64Pilot]

There is a push, most likely substantiated by statics to rely more heavily on automation, automation when it works is often safer than a person.

The stats for instance have been showing that automation is safer, for instance Tesla Full Self Driving is five times less likely to have an accident than hand driving and a Tesla due to the other automation is half as likely to have an accident than the average US car.

https://www.torquenews.com/14335/tesla-fsd-releases-safety-data-5x-safer-normal-driving

Having said that I still think it’s important to be competent driving myself and while I haven’t tracked ours yet, I’d like to just to find it’s and my limits, because whike I use FSD extensively, I don’t trust it.

I don’t know anything about a Cirrus, never flown one, perhaps their Autopilot system is so well integrated and new, modern etc that it has an extreme reliability, but it’s my experience that systems in our older aircraft aren’t, very often it’s installation or other issues not directly related to the Autopilot hardware, I do know that broken or unreliable Autopilots are pretty common. Most don’t trust Vacuum pumps, I don’t although I’ve yet to have one let me down. I had a Pesco pump on the Maule and an electric standby attitude indicator.

Maybe if there was a backup for the Autopilot I’d be far more likely to trust one, but so far as I know there isn’t.

[Max Clark]

I punched my instrument ticket in October.

Almost all of my training was S22T w/ Perspective+ flying what I called the video game. The Garmin autopilot is a dream to fly and even in real IMC was no problem flying down to minimums. Learning how to program (and more specifically reprogram) the avionics was way more complicated than I expected at the beginning. And of course remembering to switch the CDI based on the approach being flown has busted more than one check ride.

I ended up with 20 hours in an AATD and was forced to "hand fly" in the sim a lot of the time. I'm so glad that my instructor did this with me. Was the main reason I was able to get ahead of the aircraft in the real world. Holding course and altitude while fumbling with the flight plan get's you task saturated in no time.

My DPE had me setup for an ILS, then vectored me to a hold and had me re-setup for a VOR approach with Circle to Land. Oh and he failed the autopilot and PFD during this. Let me tell you I was happy I had the training I did, and hope to never have this situation in the real world.

When I went to the Mooney Safety Pilot Proficiency Program my instructor made me hand fly (VFR and IFR) my plane. It was some of the best couple of hours of instruction I've had and really highlighted how dependent I'd become on autopilot during my instrument training. Now I make it a point to setup the autopilot but hand fly the plane more. I've been talking with instructors based in DFW and a big part of that is to hand fly instrument under the hood.

"You don't rise to the occasion, you fall to the level of your preparation."

I hope to never need to hand fly an approach in hard IMC, but I want to be ready if that ever happens.

[Max Clark]

14 minutes ago, Hank said:

My personal attitude is two-fold:

1. In IMC, use whatever assistance you have available. My job is to watch it, anticipate the next move, and make sure everything is as close to perfect as possible. 
2. For practice approaches, make some of them with no help, just me holding the yoke and following the needles. My goal here is to be at least as good as my last practice approach, but probably not up to the level of #1 above. 

Agreed - I have the same thought

[Max Clark]

15 minutes ago, Hank said:

P.S.-- @AndreiC, if you want goggles to feel "more real," go out with a CFII and do approaches in the evening, out in the country. You won't know where the sun is, because there isn’t one, and any peeks you get from behind them wont help you any.

Really excellent tip

[Vance Harral]

10 hours ago, AndreiC said:

I don't fly actual IFR all that much, probably just barely the 6 required approaches in 6 months

There's a lot of good advice in this thread, but this is really the crux of the issue - you can't be good at something you rarely do.  I appreciate that frequent IFR training and practice are a challenge for people like us, who don't fly professionally.  But the challenge doesn't change the truth.

[N201MKTurbo]

An airport neighbor asked me to be his safety pilot for some approaches. He was flying a Rocket. On his third approach he was about to the FAF and I pulled the CB for his autopilot. He totally lost it and had to take off the hood. It took him a while to find the pulled breaker. He would have been dead in real IFR. He got mad as hell and was yelling at me. He said I had no business doing that and he was going to report me to the FAA. I wish he would have, that would be a fun conversation. He told me he would never fly with me again. Which was fine with me, I didn’t think he was safe. 
 

The funny thing is, when he sold his plane, the buyer hired me to ferry it. He had to hand me the keys to his plane.

[ProtoFly]

My Cherokee 140 didn't have an autopilot, and my M20F also does not.  I've never actually flown in IMC WITH one.  I don't have a bunch of time in IMC - probably 10-15 hours total, but what my limited experience show is that I'm going to steer clear of anything that looks like a cumulus cloud, even small ones!  Climbing or descending through a layer, no problems.  

My first 'real' flight in IMC after getting my instrument rating:  Returning home from KRDU to KABE.  Filed, took off in 700 ft mist, which was burning off.  Given a 'depart via' clearance - something I had not yet done.  Had to look up the departure, and then figure out if I could even comply.  Took off, and about 700 AGL, entered IMC with limited visibility to the ground.  Ultimately climbed out to 4000, and was nicely in between layers.  Was beautiful.  Then came the closing in of the layers.  OAT was around 50 F, so no risk of icing.  In the soup, and what wasn't really practiced during training happened:  ATC: "Cherokee NBUGSMASH, have change to your clearance, advise when ready to copy".  Hand flying, IMC, and now have to do something for the first time 'for real'.  Very challenging.  Trying to juggle writing, flying, and getting initially on the new heading, while trying to get the GPS course updated.  Falling behind, and realizing what a great thing an autopilot would be.  Then ATC did it again.  And I got better at managing it.  Then a third time, and developed a flow that worked for me.  Like most things, the more you do it, the better you get.

My M20F is MUCH nicer in IMC.  More stable.  But still no autopilot.  Yes, it's on the short list of things needed.  I bought the Mooney so that I could visit family that lives too far to drive, but well within a morning flight to spend a weekend.  It's nearly perfect for the mission.  Typical trip to visit family is about 500nm.  What would have been 5 or 6 hours in the Cherokee, with a stop for fuel, is now a 3.5-4 hour trip, no stops (as long as I plan my pre-trip controlled dehydration).  The distances covered really emphasize the critical nature of flight planning for weather.  I plan for staying longer, or leaving earlier, depending on the forecast conditions.  

Some other observations.  When I was flying to pick up my newly purchased M20F last year, I took my CFII friend with me.  We alternated who was flying, as the Cherokee is slow, and there'd be multiple stops for fuel.  We got into an area over Ohio and Indiana where there was a layer of what I'd visually call 'popcorn' clouds.  Small clouds scattered, but cumuliform-ish.  We had filed, and the clouds were at our altitude, so rather than dodge them, we figured it would be good IMC practice.  Hey, they look pretty benign!  Not so much.  My buddy was flying this part of the leg, and he was sweating!  Bumpy inside those things!  The fun part was that each cloud we went through, he got better at it.  We'd bounce around a bit, trying to stay on altitude and heading, and then pop out the other side, getting a nice break.  In the end, it sharpened our hand flying skills, but made me realize that it's worth just staying away from cumuliform clouds, no matter how fluffy and small they look.

[midlifeflyer]

23 minutes ago, AndreiC said:

Thanks all for the mix of comments, this is very helpful. Here is my main question. Given the type of flying I do -- about 100-150 hours per year, all of it recreational, I can cancel any flight if the weather does not seem right to me, probably 95% of it VFR, very rarely approaches to minima -- what is a good frequency with which to go up with an instructor to shoot approaches hand-flying?

One option: with what I do I can stay current (barely) by going out once every couple of months when the weather is like yesterday and shooting 3-4 approaches in actual, with the autopilot on (to be safe), maybe shoot one approach by hand and call it a day. Since I don't ever plan to fly actual IFR when I don't need to without a functioning autopilot, this seemed (until yesterday) like a reasonable option. I prove to myself that if the autopilot kicks the bucket in hard IFR I can get myself on the ground by hand, treating it more like an emergency. I would ask ATC to divert to a place with the easiest approach possible (LPV or ILS, never anything without vertical guidance), with good weather well above minima, etc.

Option two: go out regularly with an instructor, enough to stay current on instruments to be able to fly by hand confidently. Of course this would be better, though with the existing availability of instructors it seems hard to do. Also, I found that under the hood work does not feel the same way to me as actual IFR, and I am not confident that if I do everything right with an instructor under the hood, this will translate to perfect flying in actual. (And good IFR weather is not often happening here -- either it is good VFR, or bad TSRA or icing...)

What do people do in situations like these? Should I call it a day on IFR flying, like @A64Pilot? That is very limiting... Is it ok to say I have an IFR-lite rating?

Don't be reluctant to fly "IFR light." That just mean high personal minimums. But be aware that the ability to hand fly, while absolutely essential, is the smallest part of IFR flying. Understanding procedures and your avionics comprises the bulk of it. 

There are a number of ways to maintain an acceptable level of proficiency.  

1. Watch for those "acceptable IFR" days and get out there. In reality, in many parts of the country, that can be difficult. 

2. Some recommend an IPC every 6 months. I have a couple of people who do that with me, to maintain currency, but I am trying to convert them to a non-IPC flight with me every 3-4 months. These, btw, are people who regularly fly in the system, but are faced with the reality mentioned above - they just don't get the weather that allows them to log the approaches, let alone holds. The problem with IPCs is that the minimum requirements can lead to some very unrealistic scenarios. Without having to check off those boxes, we can focus on proficiency more deeply.

A regular flying buddy to act as a safety pilot. The catch here is that you need to create challenging scenarios. Flying the same approach you know like the back of your hand over and over again will do wonders for currency but zero for proficiency.

Simulation. It doesn't have to be more than a BATD for currency purposes. Like the flying buddy, you have to be creative in what you do, but it does allow you to toss a virtual dart at a virtual map and pick somehing you are completely unfamiliar with. My own personal sessions have specific goals. It might be to fly something very unfamiliar (even a ODP that doesn't count for currency) or it might be to focus on hand-flying. Simulation is great for hand-flying. Especially with a static BATD, there's no biofeedback from the controls and the controls themselves tend to be a bit squirrely.  That's actually a good thing since it makes it so easy to lose control if you are not paying attention. 

[midlifeflyer]

36 minutes ago, A64Pilot said:

don’t know anything about a Cirrus, never flown one, perhaps their Autopilot system is so well integrated and new, modern etc that it has an extreme reliability

Modern Cirrus uses "Perspective." It's their version of a G1000/GFC700 integrated system. I've flown with it and yes, it's extremely reliable.

[Pinecone]

41 minutes ago, Hank said:

My personal attitude is two-fold:

1. In IMC, use whatever assistance you have available. My job is to watch it, anticipate the next move, and make sure everything is as close to perfect as possible. 
2. For practice approaches, make some of them with no help, just me holding the yoke and following the needles. My goal here is to be at least as good as my last practice approach, but probably not up to the level of #1 above. 

 

I agree with this approach.

Most of my flying career, I flew aircraft without an autopilot.  So everything was hand flown.  Including jets.  Nothing like 175 KIAS on an ILS hand flown.  At least most had a Flight Director, which helped.

I only do a coupled approach for practice, every so often.   If I am a bit rusty I do my practice in an AATD or in the plane on a VFR day with a safety pilot.  If I am feeling sharp, I have no problems on a "safe" IFR day and solo.

[midlifeflyer]

12 minutes ago, N201MKTurbo said:

He said I had no business doing that

IMO, he was right about that. 

[Hank]

1 hour ago, AndreiC said:

What do people do in situations like these? Should I call it a day on IFR flying, like @A64Pilot? That is very limiting... Is it ok to say I have an IFR-lite rating?

IFR Lite is fine, just be aware that the weather is what you get, and it may or may not match the forecast. 

The "off" approach I described above was at the end of a 505nm XC. I was above the forecast layer at 8500', and it rose near me over LA then fell back down over MS. Coming across AL, I could see the edge of the clouds below, marking where weather would magically clear up to scattered cirrus at 25k, visible above me.

Well, at the edge, there was another layer down below, murky gray. Finally within range, I listened to ASOS and it was calling 400 Broken, 2 miles in mist. Just then, ATC asked if I had the weather at home; when I responded affirmatively, they came back with "say intentions," my least favorite thing to hear. So I checked nearby fields, the only one with an ILS was calling 900 OC, visibility > 10, so i went there and stumbled into using the non-ILS runway and being vectored across the IAF in a turn, and too high. 

You gotta do what you gotta do. My wife had to drive an hour to get me, and I met several other nice pilots in the FBO which ad also diverted there from all around. Then we got to drive back later and pick up the plane. At least my truck was still in my hangar to get home. 

Make the best plan you can, hope ATC doesn't mess it up much, but be ready for the weather to not be what you were told to expect!

[EricJ]

We don't get a lot of IFR around here, so when I'm feeling like I need to fly just for proficiency I'll often go shoot an approach somewhere in VMC, no hood, just flying the approach and doing the buttons and all the equipment setup and manipulation.    My autopilot has been inop for many years, so I hand fly all of these.    It's not IMC, but it keeps you proficient with anticipating events and flying the airplane according to the approach procedure.   I do the usual proficiency flight every six months with a safety pilot, and no autopilot, and that winds up being pretty straightforward if I keep up with doing practice approaches regularly.

All that said, I'm not a fan of actual IFR without an autopilot.   There's too much multitasking required for me, especially if you get a reroute mid-flight in IMC or something.   Fortunately around here it's not been a problem since it's rarely an issue to not be able to get where you need to go VFR.

[N201MKTurbo]

1 hour ago, midlifeflyer said:

IMO, he was right about that. 

You don’t think an IFR pilot should be able to deal with an autopilot failure? Better to practice it in VMC than in the soup. If you anticipate it, it isn’t very good practice.