Corvair Engines and the Sonex
Updated November 27, 2004
Friends,
We are putting this on our Web site as an extremely frank discussion of the concept of installing a Corvair
engine in the Sonex airframe. We have previously posted brief discussions discouraging the combination. As
time went by, I found that it was necessary to expand with more technical reasons why I felt that it was
a poor combination. Here is the latest update on my evaluation of the combination. In a nutshell, my opinion
has not changed. Out of respect for the readers who are thinking people, who are not inclined to take "it's a
poor combination" as a complete technical discussion, we're expanding this section of the Web site. I ask that
you read it carefully, and consider it in its entirety. It is a plain discussion. There are no hidden points
or agendas, and there's nothing to be read between the lines.
John Monnett, the designer of the Sonex, has stated that the firewall forward max weight for the aircraft
is 200 lbs. A handful of Corvair engines have been built slightly below this weight. It takes a special effort to lower a
Corvair engine below 200 lbs. and would not include electric start, etc. When the weight of a motor mount, cowling, prop,
exhaust, etc., is considered, it is simply not possible to build a Corvair engine package for the Sonex which
will meet the designer's requirement of 200 lbs. or less total firewall forward weight. This is the beginning
of why the Corvair is not a good match for the Sonex airframe. Initially, it was the only aspect on which I chose to
elaborate.
About a year and a half ago, the first Corvair powered Sonex airframe began its test flights. It belongs to
Del Magsam of Wisconsin. To date, this remains the only Sonex airframe to be powered by a Corvair engine. Del
is a customer of ours, but was on his own to work through the difficult process of fitting the Corvair engine
to the airframe. Out of respect for his accomplishment, as we would for any of our other customers, we wrote
a story about his airplane in our own newsletter, The Corvair Flyer. I did not then,
and still do not now, consider this acknowledgement as endorsing or promoting the combination. Over the years,
we've done a lot of work to develop firewall forward components to install Corvairs on various airframes, and
people who follow my work will clearly see that I haven't developed or endorsed a single component specific to
the Sonex airframes.
The reason for this is simple: I know the Monnetts fairly well, and they have presented a number of good technical
reasons why it's not a good match. Let me be clear and say that no one from the Monnett family has ever
asked me to put any of this on our Web site. It's just easier for us to have it here and avoid repeatedly answering the
question individually as private e-mails come in. Below, you will read several more reasons why people who are
thinking about the combination may want to reconsider. These points are my own, and are not the opinion of
the Monnett family or Sonex Ltd. If you have further questions or issues with these points, it would be properly
directed to me and not the Monnetts.
Although I did not follow the progress of it closely, the discussion of non-approved engines in the Sonex
airframe (approved engines are the Jabiru 3300 and 2200, and the Aerovee 2180) has occasionally become very
heated. I have not taken part in this, and have done nothing to fuel it. Keep this track record in mind when
you read my thoughts here. If anybody thinks of taking something I've written here and using it as ammunition
in an Internet squabble, you've completely missed my tone and intent.
The official position of Sonex Ltd. is that only airplanes that have one of the three factory approved engines
are really Sonexes. While this may seem like a small point, in many ways it isn't. I understand their position
entirely, and it's not a new position in this industry. People who have been around sport aviation for a while
will remember that Burt Rutan was adamant and very vocal about people using unapproved engines in VariEzes and
LongEzes. He felt that he had designed a complete airplane with careful consideration given to the powerplant.
The record showed that airplanes built to his plans had a vastly superior track record over those which had
been modified. He did not want the blame for work which was not in the plans, nor did he want unapproved
installations polluting his insurance statistics. While a handful of people thought if they'd purchased plans
from the Rutans that they had a right to name anything they finished a VariEze or LongEze, most reasonable
people saw Burt's position. Likewise today, I think that most reasonable people understand Sonex Ltd.'s position
about the name of the aircraft. Thus, if you plan on putting a Corvair engine on the front end of a Sonex, it's
no longer going to be a Sonex.
First, this means that you may have a very difficult time getting it insured for a reasonable price. This
especially applies to hull insurance. I've dealt extensively with the EAA's insurance provider to obtain
reasonably priced coverage for Corvair engines in 601 airframes. When you install an unapproved engine in an
airframe that can no longer be grouped with its original brethren, you're now in an insurance group of one.
Furthermore, it will have a big effect on resale value on the plane should you choose to sell it. To aid in
the discussion, please take a look at the following photographs. Keep in mind as you look, they're merely
an acknowlegdement that some people are trying the combination, along with a frank discussion about it. If we
chose not to put the photos on the Web site, a reader might falsely assume that someone had solved all the
issues and I was unaware of it. I keep close tabs on most things going on in the world of Corvairs, and I
present this to illustrate that we're aware of it and even this aircraft, which has not flown, does not address
the installation issues to the satisfaction of Sonex Ltd.
Most people looking at the above photo would say it's a William Wynne 601 Nosebowl on a Sonex, probably with
a Corvair engine in it. Considering the above paragraphs, anybody who said this would be off the mark. This
is a plans built airplane. Obviously the builder started off with a set of Sonex plans. But because he finished
off with a Corvair engine, the plane is realistically no longer a Sonex. The builder of this aircraft refers
to it by the name of "Cleanex" because he understands the logic of Sonex Ltd. I additionally want to point out
that the nosebowl on this aircraft started off life as a 601 Nosebowl, but underwent some serious surgery to fit
this airframe. We didn't do it. The builder did. Nor do I have any intention of re-creating our 601 Nosebowl to
fit the Sonex. Should anyone see this airplane in the future, it would be most accurate to refer to it as The
Cleanex, and understand that it does not actually have a 601 Nosebowl on it.
Here's a photo of the Cleanex and its builder in front of my hangar. A few people who saw this photo made
jokes about the "Builder Protection Program" with a nod toward John Monnett's reportedly sharp temper. In reality,
the Monnetts probably know who this builder is. The reason he chose to wear the paper bag is to avoid
answering any questions about his plane, or being drawn into any type of Internet controversy surrounding his
choice of engine. The decision was his, and he does not want anyone to feel that his choice should legitimize
theirs or encourage others. Although he's wearing one of our shirts, I can assure everyone he's not a member of
my staff, and before you take the trouble of e-mailing me, let me assure you I will respect his wishes and
won't divulge his identity.
The engine in the Cleanex is a highly modified Corvair powerplant. This particular installation is a hybrid
engine utilizing VW cylinders machined to fit the Corvair case. While standard Corvair engines can be built from
our Conversion Manual for approximately $3,000 to $3,500 in parts, this is no standard
conversion of a Corvair. The builder intended to lower the weight as much as reasonably possible, and increase
the horsepower to get as close as possible to the Jabiru 3300's output. This conversion cost about $7,500 in
parts, and the builder had to take a lot of time to assemble it. High end engines like this, with hybrid VW
cylinders, are more expensive and far more difficult to assemble correctly than our standard 100hp model,
straight Corvair conversions. My Manual discusses both engines, but it's important for people to note that
standard Corvair engines will not come even close to meeting the Monnett's weight restriction firewall forward.
This engine weighs approximately 8 pounds less than a standard Corvair, but it's still not going to meet the
200 pound limit. When this airplane is done, the builder has offered to weigh it in public to demonstrate the
reality of the combination. It has been falsely reported many times that there is some modification that can
be done to the Corvair to meet Sonex Ltd.'s weight restriction.
Here's a view of the underside of the Cleanex's motor mount. The Corvair uses a bed mount. The approved engines
for a Sonex all use a Lycoming-style mount. This is another factor in why the firewall forward weight with a
Corvair would be outside the limits. Bed style mounts on reverse sloped firewalls are going to be inherently
heavier. Del's engine installation made adapters to mate the Corvair to a standard Sonex mount. I did not like
the way that turned out. It functioned, but I felt it overstressed the Sonex mount. Additionally, it precluded
the use of our ignition system, which I feel is the most proven and reliable ignition system for Corvairs.
The Cleanex has one of my ignition systems installed, and I like the layout of this motor mount far more. But
keep in mind that this is far from an engineering evaluation. This mount was eyeball engineered. I did not
build it. There were no stress calculations done. Nor was it load tested. Pete Buck did the structural analysis
on the Sonex. He's one of the best engineers who ever made a contribution to sport aviation. It's a stretch
to propose that an eyeball engineered motor mount would be the perfect complement to his
airframe work. The Cleanex builder has no jig or plans for his mount. I did not put a tape measure on it, and
I'm not in a position to comment on its design. If you choose to go with a Corvair powerplant in this airframe, this is one of
the issues you're going to have to tackle.
This top view of the engine shows that the Cleanex builder chose to use a number of the products that we sell
to put Corvair engines in other airframes. In the foreground is our Front Starter
arrangement. This allows the use of a stock oil system. The Distributor shown is one of our dual ignition
units. The cylinder head intake logs are left intact, but modified with our weld on intake pipe conversion.
The point I'd like to make here is that this builder has chosen a difficult installation. To have the greatest
degree of success, however, he has chosen to utilize as many proven components as possible from our inventory
of parts. I do not promote the combination, however the reality is I would greatly prefer that anyone who is
determined enough to build their own airplane like the Cleanex at least utilize the R&D work we have put into
the Corvair to their full advantage.
The above paragraph may sound like doublespeak, but it's not. The reality is I recognize some people will
do this. I don't want anybody feeling that they are an outsider and cannot utilize our basic R&D if they're
going to attempt such an installation. Sonex Ltd. has never asked me if we have any customers in common. I
feel their major issue is with people who want to call their done project a Corvair-powered Sonex or with people who would
falsely argue the Corvair meets their specifications. I think most reasonable people would understand my position
of not promoting the combination, respecting the Monnetts' wishes, but not isolating any Corvair builder in
need of basic information. This simultaneously allows me to respect the wishes of my friends, and lower the
risk of people attempting this type of combination.
Everything above can be considered a complete statement on the combination. For a very detailed look at how
well I understand these issues, please continue to read down and consider the false claim that the cylinder
heads on a Corvair can be significantly lightened. The engine shown in the above photos weighs slightly under
220 pounds installed. This excludes the mount. Someone might later try to claim they could further reduce the
weight on the engine, but I doubt even a few pounds could be removed. Even a switch to aluminum cylinders, a
topic which has been discussed but never installed on a running Corvair engine, far less a flying one, would
not reduce the weight of the engine below 210 lbs. Any other starter arrangement will require modifying the oil
system, which will increase the overall installed weight and reduce the reliability.
Cylinder Head Weight
I have a lot of data that shows the Corvair has an installed flying weight of about 225-230 lbs. This weight
includes an 18 amp alternator and electric start. This is slightly below the installed weight of a typical
O-200 Continental. Recently, several people have suggested that excess weight might be trimmed off the Corvair
to allow it to meet the 200 pound weight limit of the Sonex aircraft. This is pure wishful thinking. Take a
look at any picture of a Corvair engine we've built. Consider how big a 25 lb. steel barbell weight is, and then
look at the motor and tell me where one of those has been left in place. 25 pounds of aluminum is roughly
double that volume. The Corvair is a great engine, but it's not going to be a sub-200 pound engine with all
of its accessories. Bare, stripped motors might approach this, but it is my strongest opinion that people
would find a Corvair engine with its six cylinders on a short aircraft like a Sonex a very difficult combination
to hand prop. To simply address the issue of the weight of the cylinder heads, follow the photos and descriptions
below.
I recently heard the false claim that 5.5 pounds can be removed from each cylinder head. This is false.
I have approximately 100 Corvair cylinder heads in my shop at any time. We've constantly experimented with the
engine, and we have a number of special heads laying around. A little known fact about Corvair heads is that the
1964 and '65 heads are roughly a pound lighter each than '66-'69 heads. The later heads have additional
strengthening and webbing. In our application, the strength is not a crucial issue, and we use both models of
heads to build engines. It's important in cylinder head comparisons not to compare a trimmed 1965 head to an
untrimmed 1966 model. This would allow the false impression of a weight reduction. Although in these photos
you'll see the stock GM intake log trimmed off, I in no way endorse this. If you look closely at the head, you'll
see that GM very cleverly used this log to stiffen the top row of head bolts. With it removed, the clamping on
the head gaskets will be adversely affected. Additionally, I have numerous tests to show that the GM heads have
excellent mixture distribution, and few modified heads can match this. The best approach to a reliable, powerful
cylinder head is to leave it stock, or weld on an aluminum pipe to the stock log as we did on our 601 engine.
In the photo above is the lightest cylinder head we have in our shop, at 17.9 pounds. This is not a typical Corvair
head. It is about 8 ounces lighter than stock because it is bored out to accept 94mm cylinders. It also has been
bored out to accept 1.5" exhaust (an experiment not recommended). It also has had the intake flange removed from
the top of the intake log. Being a 1965 head, it is the lightest head with which to start. Looking at this,
you can begin to understand why someone claiming to have a 15 pound head may be exaggerating.
Here above we have another head pulled out of the experimental parts bin. It is also modified with 1.5" exhausts,
and gutted for 94mm cylinders. This is a 1966 head. The intake tubes welded on here are nowhere near as stiff as
the stock log was. This head can be considered structurally flexible. This is not desirable in any world.
Notably, this intake setup, for all the effort involved, did not have as good mixture distribution as a stock
log.
After measuring about 20 bare heads, this 1964 cylinder head stripped of its valves, above, was the lightest
head still in its stock form.
Here above is the heaviest bare head I had in the shop. It is a 1966 model. This photo and the one above
illustrate the one pound differential between 1964-65 and 1966-69 heads.
Above is a 1966 cylinder head. Before jumping to conclusions, please note that this head has valves, springs and
retainers in place. It does not directly compare to the above cylinder heads. These 1966 and later heads have
more material underneath the head bolts. While this is not a crucial flight issue, GM thought it was an
improvement. To understand how light the stock log is, compare this with the following photo.
Above is the exact same cylinder head with the entire intake log bandsawed off. Just so that you can see that it's
the same head, go back to the previous photo and note the saw curf and duct tape holding on the log for the photo.
This clearly demonstrates that the log weighs only 1.5 to 1.6 pounds, and you have to put something back in its
place. Even if this something only weighed half a pound, you would have destroyed the structural integrity of
the head, messed up the mixture distribution, and spent a lot of time. Your reward? One pound saved.
Here is the log sitting on the scale, above. If somebody tells you this weighs 5 pounds, use your common
sense to realize that a volume of 5 pounds of aluminum is a lot larger than this.
In conclusion, let the above photos illustrate the depth of knowledge we have from years of testing
Corvairs, and my adherence to the facts when it comes to teaching people about the engine. Over many years, I
watched as competing Subaru conversion companies, under extreme market pressure, exaggerated the horsepower
and minimized the weight of an EA-81. In the end, the public caught on, and never believed most claims about
the engine, even legitimate ones. With the Corvair, the situation is entirely different. I am the only commercial
provider of Corvair parts and information to the experimental aircraft world. I am absolutely adamant that
no one exaggerate the power or minimize the weight of the Corvair. It is a great engine, and it's plenty
powerful and fairly light. It can certainly stand on its own merits. Any attempt to stretch the truth only
damages the long term credibility of the engine, and undermines my work to make the engine the affordable
powerplant that so many people in aviation need in order to participate.
Corvair Engines in the Zenair 701
The Heintz family has told me directly that they do not want engines which weigh 200 lbs. or more in the 701.
Despite the fact that their 701 brochures have pictures of engines like O-200s and EA-81s, which weigh
more than a Corvair, they do not encourage installations like this. The Heintz family has a lot of
photos of customer built airplanes on their Web site. They obviously don't endorse everything that's in
these airplanes. I assume the airplanes are on the Zenith Web site as acknowledgement of builders' successful
efforts to produce a flying airplane. We have a handful of customers working on putting a Corvair in a 701
airframe, but I have not seen any serious progress on it yet, and no one has flown the combination. For anyone
who asks about reducing the weight of the Corvair below 200 pounds, I ask that you read the Sonex section
above carefully.
When I speak to people in person about the 701, I find that a number of them are attracted to the highly
efficient methodology the Heintzes have brought to building airplanes. These people often said that they chose
the 701 because they'd occassionally be operating off grass strips and wanted an airplane that was a good
slow flyer. When speaking to these people, I always encourage them to consider the 601. Having a built a 601, and
studied the 701 closely, I think they could be built in a very similar amount of time. They both use the Heintzes'
design and building philosophy of very simple, quick to build airplanes. For the type of grass field flying
that most people are interested in, the 601 is a surprisingly good performer. One of the places we stopped with
the 601 in the Summer of 2004 was the annual Pietenpol Gathering at Brodhead, Wisc. This beautiful airport
features modest natural grass runways. The 601 had no trouble operating out of this airport at full gross weight.
The XL's highly efficient wing, with generous span and area, got the airframe flying and off the soft, wet
grass quickly. The XL model features normal size aircraft tires, which do not have a problem with grass strips.
Once in flight, Gus demonstrated in the flyby that the 601XL with 15 degrees of flaps can fly slower than an
average Pietenpol. Builders who witnessed this realized that the 601 covers a broad performance spectrum.
People with the most extreme performance requirements will still need a 701, but most homebuilders, especially
ones sold on Corvair power, should take a good look at the 601. Its 44 inch wide cabin and 135mph speed are
serious factors that most builders can take advantage of on every flight, whereas few builders really need
the 701's radical short field capability. In either case, you'll enjoy the advantages of the Heintz family's
decades of experience in designing and manufacturing good, simple airplanes.
Corvair Engines in Vans Aircraft
Friends considering powerplants for RV series planes:
I have spent the past 10 years developing and flight testing Corvair engines. The overwhelming popularity of the RV aircraft and the emerging popularity of the Corvair often leads to the question: "Would a Corvair be a satisfactory powerplant in an RV?"
The plain and simple truth is NO. It has taken years of development to get the 164cid Corvair motor to economically produce 100hp with the reliability that is required from flight engines. While we now have a 120hp 190cid motor, it makes its power at an elevated rpm, 3,200rpm, and it cannot be considered a direct replacement for the O-235 in this application. The O-235 and O-290 make their power at several hundred rpm less. I believe the Corvair's thrust level to be below what would be satisfactory for the RV-9.
I have received reports of a geared, turboed Corvair installation in an RV-9, but I do not know if the aircraft has flown. The builder of the planetary gear set told me his cost was $2,500. The engine overhaul was perhaps $3,000 and the turbo system was certainly another $1,500. At this point, you've spent as much money as you would have on a mid-time 150 Lycoming.
While the Corvair has been geared and turboed to produce more power, this is done at expense which typically exceeds the cost of the equivalent Lycoming power with nowhere near the reliability Lycoming offers. Simple Corvairs have an excellent history that common sense tells you would be lost with a highly complex installation.
Only the single seat RV-3 in its most basic form would be a possible installation for a direct drive, simple Corvair.
In conclusion, the RV series of aircraft and the Corvair engine are great designs, just not for each other, with the possible exception of the RV-3. The other RVs are not designed to achieve their best performance on 164-190cid engines like the Corvair.
As a person who has done more work with the Corvair than any other living person, I suggest you consider my words carefully. Anyone who is trying to disagree doesn't have the experience I do or is trying to sell you something.
Thank you.
William