Corvair Engines - The Justification Series

[daleandee]

* * * * * * * * CONTENTS * * * * * * * *

1) Preface - http://www.sonexbuilders.net/viewtopic.php?f=9&t=4273&start=20#p32792

2) Introduction -http://www.sonexbuilders.net/viewtopic.php?f=9&t=4273&start=20#p32837

3) It's Legal - http://www.sonexbuilders.net/viewtopic.php?f=9&t=4273&start=30#p32881

4) Data - http://www.sonexbuilders.net/viewtopic.php?f=9&t=4273&start=30#p32910

5) LSA Cleanex -http://www.sonexbuilders.net/viewtopic.php?f=9&t=4273&start=50#p32949

6) Structure History - http://www.sonexbuilders.net/viewtopic.php?f=9&t=4273&start=50#p32988

7) Design Quality - http://www.sonexbuilders.net/viewtopic.php?f=9&t=4273&start=50#p33013

8) Performance - http://www.sonexbuilders.net/viewtopic.php?f=9&t=4273&start=60#p33029

9) Safety (Engine) - http://www.sonexbuilders.net/viewtopic.php?f=9&t=4273&start=60#p33049

10) Parts & Service - http://www.sonexbuilders.net/viewtopic.php?f=9&t=4273&start=60#p33081

11) Epilogue -



___________________________________________________________________________________________________________________________


I would like to begin by posting a statement from Mike when he lamented the sad way we proceed on this list sometimes:

It's a bit of a shame that we can't respectfully talk about Corvairs here and share information; please everyone, let's try to keep it civil.

... and perhaps we can heed his advice and try to be a little more courteous to each other.

We have a thread that is dedicated to Corvair knowledge called "Corvair - General Info" and a lot of great information is found there. But during the course of that thread a question was asked that needs to be addressed. At first my thoughts were to just let this go on by but some off line emails from other builders, some that are considering using Corvair power, as well as those that were curious brought me to rethink that approach. You see it was suggested that my putting a Corvair on this air frame was done without much thought, consideration to the factory warnings, or any research at all.

I took a bit of offense at the unwarranted and rather abrasive way the question was put forth by suggesting that I had used "blatant disregard for the design limits" in my choice to do this and that I only did so because "It's experimental and I can set any gross weight I want." None of that is even remotely true.

So after consulting with a moderator for the group and the Foundation I was told that the members of this forum do have the right to post relevant information for the airplanes we fly. After much consideration I have decided that I would begin a thread that will address the question of "justification." I have no intention of arguing or debating anyone over the information I will share over the next few days. The information I will offer only pertains to me and the aircraft I have built.

But before we get started I have a few questions for the group. Did you know:

1) More than one Sonex has been flown over the Light Sport Limit of 1320 lbs.?

2) A Sonex has been flown in excess of 250 mph?

3) A Sonex has flown several flights with a FWF weight almost double the factory limit?

4) The Sonex main wing was tested to 9 Gs positive & 4.5 Gs negative and that this information was on their website?

5) At least one Sonex has an empty weight of 840 lbs.?

Curious,

Dale Williams
N319WF @ 6J2
Myunn - "daughter of Cleanex"
120 HP - 3.0 Corvair
Tail Wheel - Center Stick
Signature Finish 2200 Paint Job
171.9 hours / Status - Flying
Member # 109 - Florida Sonex Association
Latest video: https://www.youtube.com/watch?v=1VP7UYEqQ-g

[x3 skier]

I’ll be reading your series. No intention to put a Corvair in my Onex but I’m always interested in general aircraft information and Sonex designs in particular.

[lutorm]

But before we get started I have a few questions for the group.

This is not meant as an argument either way for the Corvair or any other modifications, but I think those questions aren't really relevant if we're discussing design modifications.

Let me counter-ask these questions:

• How many times did the Space Shuttle fly before the Challenger blew up?
• Knowing what we know after the fact, were those flights safe or not?

It's the nature of statistics that the presence of a Sonex that has flown with twice the FWF weight of the factory limit tells us little. In contrast, the presence of a Sonex with twice the FWF weight of the factory limit that had its engine fall off would tell us a whole lot.

Just so you know where I'm coming from with this: I work at SpaceX. Since I started there, we've had two catastrophic failures of orbital launch vehicles. I can't go into details, but they've both been traced to two failures of Helium pressure vessels. They were completely different failure modes, but what they had in common is that we had successfully flown the vehicle numerous times before these anomalies. Everything appeared to work fine, but in reality, due to larger part tolerances than we were aware of, we unknowingly had much less margin than the design aimed for. Eventually, we hit a situation where the variations stacked up against us and many hundreds of millions of $$$ went up in smoke.

The Sonex design has a certain safety margin designed into it. This is meant to account for tolerances in parts, builder technique, turbulence, pilot technique, design flaws, etc. We don't know what that margin is. John Monnett does, but he's not going to tell us. What we're doing when we exceed the design limits is eat into that margin. That may work. It may work fine forever. It may also only work fine until you hit severe turbulence, or do a hard landing, or pull a few extra Gs. You have no way of knowing unless you actually test it.

I think Dick van Grunsven of Vans Aircraft put it well (from https://www.facebook.com/notes/vans-aircraft-inc/what-price-a-masterpiece-by-dick-vangrunsven/237594966250883):

Along with gross weight increases, some builders take the same liberties with horsepower increases and speed increases, betting their lives on the assumption that the airplane is designed with a huge margin of safety---it is really far stronger than in needs to be. This is not really true. Certificated aircraft, and well-designed kit aircraft, are designed to withstand limit loads at specified maximum weights. During testing, they are subjected to ultimate loads, which are higher than design limit loads by a specified margin. Yes, there is a margin between the design and ultimate strengths. But that margin belongs to the engineer. He owns the margin. It is his insurance against the things he doesn’t know or can’t plan for, and the pilot’s insurance against human error, material variations, and the ravages of time. Wise pilots respect this design safety philosophy and leave this insurance policy in effect by operating strictly within established limits. They don’t try to steal the margin from the designers.

[Bryan Cotton]

It's the nature of statistics that the presence of a Sonex that has flown with twice the FWF weight of the factory limit tells us little.

I used to work at Sikorsky. I did a stint in ground test, where we do fatigue test and compare part fatigue strength against flight loads to determine component life. As part of that we assume the parts and construction are all good, and use some statistical analysis to derate the strength of the part to give margins in material properties and build quality. This makes it very improbable that there are failures in the field. But, they happen, both in the past and I am sure in the future too. That is a long winded way of saying I agree with lutorm.

[NWade]

Here's the short-winded way of saying I agree with lutorm: "The plural of anecdote is not data" [Kernaghan & Kuruvilla, 1982]

That quote, BTW, is not a statement about Corvair engines in any way. It is merely the acknowledgement that just because someone _has_ done something, doesn't mean its smart or advisable. Plenty of people build kit airplanes without understanding aerodynamics or structural engineering; so while we can have open and honest discussions on this forum, we should also be careful not to lead future builders down the proverbial garden path and ignore the compromises and risks in exceeding the design specifications of the airframe.

--Noel

[lutorm]

Here's the short-winded way of saying I agree with lutorm: "The plural of anecdote is not data" [Kernaghan & Kuruvilla, 1982]

True, but I think the real point is the slightly deeper "absence of evidence is not evidence of absence,"

From a point of reliability, a system can be in one of three states: 1: "has failed", 2: "has not yet failed, but will at some point", and 3: "has not and will not fail". Any amount of data does not help us distinguish between 2 and 3, it just decreases the proportion of 2s by turning them into 1s. The only thing you can firmly conclude is the 1s were not reliable, which is a good lesson but isn't enough. All you have for the things that have not failed is an absence of evidence of failure, which is not the same as evidence of the absence of failures. This is why data must be supplemented with independent design analysis.

[x3 skier]

Lutorm discussion of design margin is spot on. Having spent some 50 years in the USAF aircraft development business from the F-111 to the F-22 and F-35, (losing most of my hair and gaining a lot of scar tissue) intruding into the structural or other design margins is not something to be taken lightly.

To this day, I still recall the wing failure in the early days of the F-111 and much later the static test failure of the C-17 wing, as well as many other “oh s***” moments, all of which led to major design changes and revised design guidelines.

Experimental Aviation is wonderful with the freedom to experiment. Just always remember the design is the way it is for some very good reasons and deviating from the design has risks. The bigger the deviation, the bigger the risks may be.

I myself have made and will make some minor mods to my Onex all tempered by my past history and an engineer’s perspective on why the design is the way it is.

Cheers

[GordonTurner]

....and now, back to our regularly scheduled program...

You guys are all correct, but in my opinion you are overstating the case wrt homebuilt aircraft.

We are starting with a good basic design but we are each building the plane we want. Engineering is by TLAR, construction is by “that’s about as good as I can make it”, and structural materials are hopefully somewhat controlled by taking the stuff with letters and numbers from spruce instead of home (aircraft Builder’s) Depot....

I think the basic engineering of the structure as per the plans is probably pretty well engineered and analyzed, but if you think it’s SpaceX or Boeing I think you’re deluding yourself.

Finally, what seems to kill people in homebuilts is very very few structural failures and a whole bunch of engine failures followed by stall/spin. Unfortunately Dale started off with a few not so relevant statistics, but good intentions. Let’s try and get back to that discussion.

[GordonTurner]

Basic info about Corvair motors is here:

https://flycorvair.net/2013/12/24/basic ... formation/

According to this source, all up flight weight of the Corvair 3.0 liter engine is 220 pounds. That’s firewall forward weight, complete motor and accessories, fluids, baffles mount and cowling. It’s not much heavier than the Jab3300 in a nosewheel configuration. It’s quite likely the same as the full up turbo VW with a nose wheel. The difference in weight is in the realm of a lead acid battery give or take. Engineer it all you want, it fits, it looks about right, and it’s in the ballpark on numbers. Won’t know until I get to the weighing part, but my “analysis “ of the weight and balance spreadsheet leads me to believe I’ll be in the 700 maybe plus a little range and with the Waiex the balance should work out pretty well. I’m thinking I’ll save the battery install until as late as possible and try to tune the balance with that.

[kmacht]

I agree that the questions posed don't provide enough data to help with a true risk evaluation. They are interesting data points but doing something once doesn't mean it will always work or work in every situation.

As far as the other comments, doing the stress calculations for an increase in firewall forward weight is not that difficult. Some of you on here seem to make it sound like it is beyond mere mortals. I have done the stress calculations and am comfortable with what the additional corvair weight does to the airframe and the amount of safety margin remaining. The theoretical has also been proven out by the hours that the 20+ corvair powered sonex out there have put on the airframe/ engine combination. Ideally there would be rig testing with various loads applied at different cyclic rates but that is beyond the realm and financial capability of most home builders. They call it experimental aviation for a reason.

With that being said I do not plan on increasing my gross weight. Doing so is well beyond the simple calculations needed for evaluating an increase in FWF weight. Not increasing the gross weight will limit me in the size of my passengers, baggage allowance, or possibly even fuel quantity at times. Since 90% of my flying is nothing more than hour long flights by myself around the local area this is a compromise I am willing to accept.

Nobody is going to force anyone to switch their powerplant out to a corvair. The aerovee can be made to work as shown by the few hundred out there flying. Every engine combination has tradeoffs. What is important is that you are comfortable flying behind the power plant that is keeping your airborne. For me switching to a corvair makes me the most comfortable. For others it may be the aerovee, jabiru, Rotax, or some other engine yet to be discussed.

Keith
#554