# New Optional Task-based Phase 1 Flight Test Program for E-AB

### Help Support Homebuilt Aircraft & Kit Plane Forum:

#### PagoBay

##### Well-Known Member
Excerpt --
April 1, 2021
– After years of hard work and advocacy by EAA, the FAA has published draft guidance to implement an optional task-based Phase I program for Experimental Amateur-Built (E-AB) aircraft. Under the program, once an aircraft completes a flight test plan that meets FAA standards, Phase I is complete. The standard 25- or 40-hour flight test period for Phase I will remain an option for all E-AB, and Experimental Light-Sport (E-LSA) continues to carry a 5-hour test period.
The program is part of an upcoming update to Advisory Circular (AC) 90-89B.
The Advisory Circular is in draft form and comments will be accepted through April 29. Please note that the relevant language on Task-Based Phase I is housed in Chapter 1, Section 1 of the draft. The rest of the document contains advisory information on flight testing and is not part of the task-based program requirements.

#### narfi

##### Well-Known Member
Log Member
Unfortunate release date.....

#### gtae07

##### Well-Known Member
Unfortunate release date.....
True, but we've known this was a real thing coming for a while now and it is officially on the FAA website.
I like the idea and hope that it will lead to better Phase I testing by the general E-AB population. I also welcome the additional guidance on preparing a full flight test program, having only peripheral involvement on "real" flight test operations beyond operational check flights.

Some people are hoping it will reduce the total flying time for Phase I but if it's done right I don't think it'll shorten it that much.

#### Hot Wings

##### Grumpy Cynic
HBA Supporter
Log Member
Some people are hoping it will reduce the total flying time for Phase I but if it's done right I don't think it'll shorten it that much.
My quick read tells me that if done properly it may take longer than the 40 hrs. Done in 25 hours would make me think the task was very well planned, with no unexpected complications, or there was some 'creativity' involved.
The flight test cards can add a bunch of time and I didn't see (quick scan) of an option to use electronic data gathering in place of the manual method.

#### gtae07

##### Well-Known Member
My quick read tells me that if done properly it may take longer than the 40 hrs. Done in 25 hours would make me think the task was very well planned, with no unexpected complications, or there was some 'creativity' involved.
The 25 and 40 hour test periods remain an option. This is just an alternative means. The hope is that people will be motivated to do it even if it might take a tiny bit longer.

The flight test cards can add a bunch of time and I didn't see (quick scan) of an option to use electronic data gathering in place of the manual method.
The EAA test cards are a good start and you can likely use many of them verbatim. There is provision for the use of electronic data gathering; see 1-1.h(4). The test card shown is just a sample; you can make your own. Test cards are good ideas anyway. Note also that when it talks about what "should" be in a test card, "should" is non-mandatory. If it's required you will see "must". I would think that many of the "should" items for a test card (that is, data collection items, not procedural/test point configuration ones) could be automatically captured by an EFIS and therefore not have to be manually recorded.

#### Hot Wings

##### Grumpy Cynic
HBA Supporter
Log Member
The hope is that people will be motivated to do it even if it might take a tiny bit longer
Even though I'm generally against more regulations I'd have liked to see at least some of the basics incorporated as 'must document' during the 25/40 hr option. As is I doubt many will choose the task based option.
Some of us understand the reason behind the test period and will use it for that purpose. The others need some kind of 'motivation' to not just fly around for a few hours and make log book entries..............a certain 5 place canard project comes to mind as a good example.

All in all, one of the more positive things to come from the FAA recently....IMHO

#### tralika

##### Well-Known Member
Even though I'm generally against more regulations I'd have liked to see at least some of the basics incorporated as 'must document' during the 25/40 hr option. As is I doubt many will choose the task based option.
Some of us understand the reason behind the test period and will use it for that purpose. The others need some kind of 'motivation' to not just fly around for a few hours and make log book entries..............a certain 5 place canard project comes to mind as a good example.

All in all, one of the more positive things to come from the FAA recently....IMHO

#### N804RV

##### Well-Known Member
HBA Supporter
My quick read tells me that if done properly it may take longer than the 40 hrs. Done in 25 hours would make me think the task was very well planned, with no unexpected complications, or there was some 'creativity' involved.
The flight test cards can add a bunch of time and I didn't see (quick scan) of an option to use electronic data gathering in place of the manual method.
Anyone, not an experienced test pilot, who's taken up an unfamiliar EAB and tried to work out basic performance numbers could probably guess that. Add in the time to properly break in a new or overhauled engine, you're easily gonna be 40+ hours.

#### tralika

##### Well-Known Member
In my case, building from a kit with no major modification, it would have reduced the 40 hour Phase 1 significantly. I followed the recommended flight test tasks in AC 90-89B. I scheduled about 35 flights with specific tasks and a flight test card for each flight. The majority of the specific tasks for each flight in the Advisory Circular only take about 10 or 15 minutes, even if you do everything twice. On each flight I would perform the tasks to complete the goal for that flight, record the info, then fly around or practice take off/landings until I had an hour of flight time. Some of the flights were longer than one hour when it came time to calculate x-country fuel consumption. My schedule included doing all the maneuvers at minimum weight with a forward CG, then repeating everything at gross weight with an aft cg. Even doing all of that I still spent a lot of time flying around to make the 40 hour minumin for Phase 1. My kit has been in production for quite a while so I had no surprises in the flight characteristics. Certainaly for a one-off design or or aerobatic AC it might take longer, especially if you have to repeat tasks after making changes or tweeking the AC. Based on my experience I believe for most of us this flight test option will make EAB planes safer while reducing the Phase 1 time and expense.

#### Marc Zeitlin

##### Exalted Grand Poobah
In my case, building from a kit with no major modification, it would have reduced the 40 hour Phase 1 significantly.
I will disagree that it can possibly take substantially less than 40 hours to TRULY test a new aircraft, kit or otherwise, modification or otherwise. While you may have followed the plans/kit instructions to the letter, and didn't change much, your plane was NOT built in a factory, using factory processes and factory jigs, so even if it's the 3,053rd instantiation of an RV-7, it's different from every other airplane on the planet and should be completely tested. If it doesn't take 30 - 40 hours to perform all the tests in Phase I, you haven't performed all the tests in Phase I.

My schedule included doing all the maneuvers at minimum weight with a forward CG, then repeating everything at gross weight with an aft cg.
I'm not sure why you chose those two (and ONLY those two envelope points for testing, but that's NOT worst case. Worst case will be full forward CG and MGW, and aft CG and lightweight. And since you never know what a plane is going to do before you've tested it, you should always start testing in the middle of the envelope, lightweight. And there are sometimes characteristics that can change based on weight at the same CG, or same weight and different CG's, so it's best to have at least 9 data points for testing climb, glide, rotation, stall, accelerated stall, stability, flutter, etc. - these would be lightweight, midweight and heavyweight, at mid, forward and aft CG's.

So you can see that if you only tested at the two data points indicated, you haven't tested the full envelope and worst cases, and that's probably why you were able to finish your testing in far below the necessary timeframe.

My kit has been in production for quite a while so I had no surprises in the flight characteristics.
See above - while the kit may or may not be identical to the one produced 5, 10 or 20 years ago, YOU built a custom plane that is identical to no other on the face of the earth.

Based on my experience I believe for most of us this flight test option will make EAB planes safer while reducing the Phase 1 time and expense.
It may be the case that the new flight test options will simplify testing for some folks. But I will stick to my position that the total amount of time it should take someone to perform all the necessary test in Phase I, whether certified engine/prop combination or not, is on the order of 30 - 40 hours. If you're doing it in less, you're pencil whipping it, or at the very least, missing necessary tests to fully prove out the envelope of YOUR plane.

My $0.02. #### Kyle Boatright ##### Well-Known Member <snip> While you may have followed the plans/kit instructions to the letter, and didn't change much, your plane was NOT built in a factory, using factory processes and factory jigs, so even if it's the 3,053rd instantiation of an RV-7, it's different from every other airplane on the planet and should be completely tested. <snip> My$0.02.
You really think so? I'd opine that with matched hole construction and a stock engine/prop, the 3,053rd iteration of the RV-7 might be more uniform to the type than most factory built aircraft, which aren't self-jigging during construction.

I'm not arguing that you shouldn't fully test and be comfortable with your new airplane in every conceivable configuration, but...

#### TFF

##### Well-Known Member
I think the best way to describe is most people verify factory numbers not test. Testing can be a little more exciting than most want to do as you are looking sometimes at failure to do something just as much as verification of doing.
An old article of the four seat Tailwind like plane Fourruner described the designer’s VNE test. Sandbagging passenger weights diving at VNE+ margin. Rigged door for removable pins in case he needed to jump. I don’t know if diving a plane to 250 with a decent g pull out to test the wings is in most’s ability. Especially if you did the calculations first.

#### Marc Zeitlin

##### Exalted Grand Poobah
You really think so? I'd opine that with matched hole construction and a stock engine/prop, the 3,053rd iteration of the RV-7 might be more uniform to the type than most factory built aircraft, which aren't self-jigging during construction.
I do. And while I CERTAINLY agree that the 3,053'd RV-7 is going to be WAY more similar to the 15th RV-7 is than the 3,053'd Long-EZ is to the 15th Long-EZ, there are still a zillion things that are not identical. Did the builder install the rod ends correctly? Did he tighten all the jam nuts so that things can't move (Vic Syracuse's hot button, if you read him )? Did the builder use lock nuts correctly? Did they rig the control surface correctly? Did they drill the holes that WEREN'T match drilled straight and correctly? Is the engine set up correctly (mag timing, fuel injection, carb adjustments, hose connections)? Is the wiring secured correctly, or is it all spaghetti wiring? Etc., etc.

There are just too many steps where a builder can do something that the kit factory didn't expect or anticipate for me to think that it's acceptable to skimp on the testing period. I just never want a flyer to experience a corner of the envelope for the first time, with her husband and kids in the plane, that she didn't experience in the Phase I period. So even if the plane is PERFECTLY nominal out of the box, you still need to test all the areas of the envelope, and because you don't KNOW that it's perfect until you test it, you have to start in the middle of the envelope.

And if you DO all that, it's a 30 - 40 hour sequence, not a 10 hour sequence.

Again, My $0.02. #### BJC ##### Well-Known Member HBA Supporter Certainaly for a one-off design or or aerobatic AC it might take longer, especially if you have to repeat tasks after making changes or tweeking the AC. I can’t imagine performing all aerobatic maneuvers in a new E-AB within the first 40 hours. “Sure, go ahead and do flat spins at gross weight and at aft CG. It should recover OK.” “Yup, dive to Vd to validate Vne.” “Sure, go ahead and do multiple outside snaps at Va, it should hold together.” Some things should be approached cautiously. BJC #### Kyle Boatright ##### Well-Known Member Did he tighten all the jam nuts so that things can't move (Vic Syracuse's hot button, if you read him )? Again, My$0.02.
I know Vic very well. He was in my EAA chapter until he moved South of Atlanta. Super guy and yes, jamb nuts are his white whale.

No doubt there are a thousand things that people can screw up, but I was thinking more along the lines that performance characteristics might not vary between similar RV-7's - climb rates, stall speeds, glide ratios, etc, I wasn't commenting on the mechanical shake down of a new airplane. Certainly, you find the infant mortality issues in the 25/40 hour fly-off, but there are plenty of other wear and tear items that pop up years down the road. I got back from Oshkosh a few years ago right at dark after a long day circumnavigating weather. The airplane was fine, and we had all the lights on as we approached the home drome in fading daylight. Next time I started the airplane, I got an "Alternator Inop" light and it turned out that the B-lead lug had failed at the alternator, probably at start-up and maybe because the wire wasn't secured well enough. It took 1100 hours to find that one.

BJC

#### Rhino

##### Well-Known Member
I do. And while I CERTAINLY agree that the 3,053'd RV-7 is going to be WAY more similar to the 15th RV-7 is than the 3,053'd Long-EZ is to the 15th Long-EZ, there are still a zillion things that are not identical. Did the builder install the rod ends correctly? Did he tighten all the jam nuts so that things can't move (Vic Syracuse's hot button, if you read him )? Did the builder use lock nuts correctly? Did they rig the control surface correctly? Did they drill the holes that WEREN'T match drilled straight and correctly? Is the engine set up correctly (mag timing, fuel injection, carb adjustments, hose connections)? Is the wiring secured correctly, or is it all spaghetti wiring? Etc., etc.......
Last time I checked, those are inspection items, not flight test items. Flying more test hours won't do anything at all to find or alleviate such problems, except maybe to bring an inevitable failure closer to fruition simply by virtue of operating the aircraft longer.

And if you DO all that, it's a 30 - 40 hour sequence, not a 10 hour sequence.
That's just not true. There's a whole lot more complexity to a Lancair Evolution than a Zenith CH 701, so saying their testing should take the same amount of time simply doesn't pass the smell test. I see what you're getting at. Skimping on test items, 10 hours was the example you chose, is absolutely the wrong approach. But that wasn't what tralika was saying he did, nor was it what he was advocating. There are simpler aircraft that will take less time to test certain items, or maybe not even require testing at all, and some complex aircraft that will take more time to achieve similar results. I agree that 10 hours certainly sounds unrealistically low, but nobody suggested that timeframe. The argument could certainly be made for no less than 30, as your range states, but it's not outside the realm of some very simple aircraft out there, even with complete testing, or even testing some things twice as tralika said he did.

I agree that most aircraft will probably take at least 30-40 to properly flight test, but by no means can that be stated as an absolute. Most homebuilts don't have retractable gear, and some have no flaps, so there goes Card 3. Some don't have a pitot static system, so there goes Card 4. Some have no trim, so there goes more test cards. The time and complexity of testing all the cards could vary dramatically between different aircraft. You get the idea. There is a huge variance in the complexity of homebuilt aircraft, so you can't logically claim it will take the same amount of time to flight test all of them. Just as some aircraft might require significantly more than 40 hours to flight test properly, perhaps significantly more, some could indeed take significantly less time.

I think what you were really getting at was the possibility that some pilots/builders will cut corners that they shouldn't be cutting just for the sake of simplicity. That is indeed a concern, though when you think about it, it may not be as much of a concern as it appears to be at first glance. My personal 2 cents is that any pilot/builder who's prone to cutting such corners would just as likely be prone to doctoring the hours they've flown on a standard, hour-based test plan. Willful stupidity comes in many forms, but it's fairly consistent, with laziness being an almost constant companion along the way. It's amazing how much effort a lazy person will expend just to avoid expending some other effort somewhere else. The only real danger I think the task based system would increase along those lines would be to maybe give those people more ideas on where they might cut those corners. That mentality, and the danger it represents, are going to be there either way.

BJC

#### Marc Zeitlin

##### Exalted Grand Poobah
<A lot of good points>
Agreed - for very simple planes, fewer than 30 hours may very well be required.

I think what you were really getting at was the possibility that some pilots/builders will cut corners that they shouldn't be cutting just for the sake of simplicity.
And what set me off on the first posting was that the poster only tested two of the corners of the envelope, and not the most important ones, then claimed that he needed fewer hours because he didn't need to test more. Which, in my mind, was incorrect - he didn't test the most important corners of the envelope, and didn't test the middle of the envelope.

That is indeed a concern, though when you think about it, it may not be as much of a concern as it appears to be at first glance. My personal 2 cents is that any pilot/builder who's prone to cutting such corners would just as likely be prone to doctoring the hours they've flown on a standard, hour-based test plan. Willful stupidity comes in many forms, but it's fairly consistent, with laziness being an almost constant companion along the way. It's amazing how much effort a lazy person will expend just to avoid expending some other effort somewhere else. The only real danger I think the task based system would increase along those lines would be to maybe give those people more ideas on where they might cut those corners. That mentality, and the danger it represents, are going to be there either way.
Interesting viewpoint .

While I don't disagree about the lazy folks, or the folks that cut corners ON PURPOSE, at least the folks that are just ignorant but WANT to do the right thing should have some concept of what's actually included in the Phase I test period, and what they should expect from it.

I don't think we're too far apart here... I mostly come at this from the standpoint of relatively high performance aircraft (Long-EZ, COZY, Lancair 360, RV-7, Velocity, etc.) so that's where my head is - not in the low, slow and fly around in circles crowd, where the planes are a lot simpler.

Thanks.

#### N804RV

##### Well-Known Member
HBA Supporter
I would think that even the "low-and-slow" crowd should be doing speed triangles, specific fuel consumption, best-glide/minimum sink, stick-force-per-g measurements and so on. And, the results have to be reproduceable, not just 'one-and-done'.

I have no problem with making Phase I testing task oriented. I just wonder if that's really going to increase the quality of phase I testing, or if its just going to make it easier to "gun-deck" the whole process.

#### Hephaestus

##### Well-Known Member
I've got 2 minds on this one...

Task oriented is one thing. But part of it is still just burning off hours to prove the systems reliability.

But then we have cases like Mr raptor - where it's clear he isn't running the cards just wants to burn off the countdown timer, and the systems are failing at an unacceptable rate but he's ok with that.

It's too bad there wasn't ongoing monitoring/check-in during the phase 1 - every major systems issue added hours.