The intended Sonex Powerplant


Exploded Drawings of the Type 4 Long Block (Aug. 2nd, 2001)

If you have some strange looking parts left over and don't know where to put them on, the following sketches may be helpful.

All pics are of high resolution and most are larger than the usual computer screen. So it's best you download the pic and use your favourite picture viewer to examine it offline.

Engine Case (GIF, 97KB)

Cylinder Head (GIF, 59KB)

Crank (GIF, 66KB)

Cylinder (GIF, 28KB)

Cam & Lifter (GIF, 69KB)

Oil Pump & Filter (GIF, 53KB)

Breather & Dipstick (GIF, 83KB)

Inition (GIF, 64KB)

Starter (GIF, 163KB)

Fuel Pump (GIF, 56KB)



A Commercially Oriented Powerplant Reality Check

I have researched the German 'Aircooled VW' scene now for quite a while. I have read several books, studied intensively the internet and have collected catalogs of the most reputed aircoolded VW rebuilders/tuners. This is what came out so far:

If machining is being done by a specialist rebuilder, then the price will be somewhere around

DM 10 000 (or about USD 4 500)

for the basic engine. This is with exhaust and muffler but without ignition and carburetion (and wiring and small parts and baffling and and and ...).

The engine should provide a solid 90HP takeoff power (which I may need for my expected overweighted flying machine).

Compared to the 2200 Jabiru 80HP engine the end result may be in par in money, however the type 4 will consume another several hundred work (or fun) hours. The Jabiru may be the more proven/reliable product, but the type 4 I'll know then in and out by heart (and it hopefully will deliver a few more horses).

Maintenance for the type 4 is expected to be cheaper than for the Jabiru.

The Jab I will have to pay in one sum, the type 4 I can 'stretch' in several smaller amounts over the time of the rebuild.

The Jab 3300 is financially out of my reach. The Corvair engine is not available/unknown in Germany. The Type 1 (Great Plains) is too weak. Rotax 912 is too expensive and there's no support from the airplane designer (Sonex-Ltd).

The type 4 will be my on creation.


This is a GREAT program for developing a 'paper'-engine


(and to refute a lot of HP claims from several auto-aviation converting firms)


When removing the case bolts and cylinder studs (M10x1.5 rolled fine thread), the threads were covered with a very stout varnish.

Very hard to remove. A good idea came from Bob Hoover (Veedubber) from the VW email list (highly recommended). He proposed to split a nut and use this one as a thread cleaner.

Works great. A regular threaded die should not be used for this cleaning job, because the rolled bolt threads might be hurt.


If you decide to go with a Type 4 VW engine (which I think will make sense...), then check the following when you buy the core for your rebuilding project:

The engine case has an identification stamp near the breather case (fan end). At least in Germany the ID is extended with an 'X' if the engine has already been rebuilt. THIS IS A BAD SIGN! The engine case can be rebuild/bored TWO times, then it's ready for the junkyard. So look out for a normally worn-out engine without an rebuild history!



This is an interesting link for a homemade EFI system (I believe more in this

technology than 100-year's old dripping carburetor)



this one is a very nice collection of engine calculation programs. Here I found that compression ratio of 7.5 instead of 8.0 only cost me 2 hp for an 80 hp engine (but is much nicer to the cylinder heads) - HOWEVER - a 2.2L engine with compression ratio of 8.0 (Great Plains VW) gives me only a little of 62 hp at 3200 rpm (what I always have been afraid of).



nice collection of VW aircooled engine rebuilding and maintenance tips



here is a complete Porsche 914 ('VW'-Porsche) Type-4 engine parts-list in microfiche form. This could be the engine of choice for my aero-engine conversion...



Universal Formulas


Im Allgemeinen wird die Schallgeschwindigkeit nur in Abhängigkeit von der
Temperatur angegeben (die allerdings z.B. mit der Flughöhe abnimmt):


Hierin sind
a: Schallgeschwindigkeit in [m/s]
k (kappa): der Isentropenexponent der Luft = 1.4
R: die Universelle Gaskonstante = 287 [J/(kg*K)]
T: die absolute Temperatur in [K], wobei T(K)=t(°C)+288.15

Bei Raumtemperatur (20°C = 308.15 K) also: a = sqrt(1.4*287*308.15) = 351.87
[m/s] = 1266.74 [km/h].


HORSEPOWER/THRUST/ calculations / estimations

Horsepower = torque * RPM / 5252


There is a formula for calculating horsepower from CFM....... displacement
divided by 2 times RPM divided by 1728 gives theoretical CFM. Actual CFM
will be about 85% at max rated power for most engines, so multiply by .85 to
get a realistic actual CFM. Divide CFM by 1.62 to get SAE horsepower.

The easy way is to multiply displacement in liters by RPM in hundreds and
multiply by .926.

A 1600 engine naturally aspirated turning 4500.....
1.6 * 45 = 72
72*.926 = 66.67 HP

H.W.,, Jul 24, 2000


Here's some good words on thrust
It seems like it's back to front - but in a recip engine developing constant
power, the thrust produced by the prop decreases as speed increases - in
order to maintain a basic relationship, that
speed times thrust = a constant number.

There is a limit to this rule however - else at a ground tether (speed = zero)
the thrust would be infinite and it isn't. You would want the prop's
max angle of attack to be happening just under stall speed, for best escape
from stalling, and have the prop more and more stalled at slower speeds,
so the prop drag is higher, even if the thrust is high.

Here's a sample calculation, 50 HP at 50 mph (in SI, for laughs)

We choose a VW that develops 50hp say = 50 x 746 watts = 37.3 kW
Plane is taking off at 50 mph say = 22.4 meter/sec

Remembering thrust times speed = power
so thrust x 22.4 = 37.3 x 1000
so thrust = 37.3 x 1000 / 22.4 = 1665 newtons

(Thats 373.4 lbs, found by multiplying newtons by 2.2/9.81)

That's about as much thrust as yer VW thust washers need to take care of.
Less at higher speeds.

Brian W,, Mar 19, 2000


The "limits" can come with a very small increase in rpm sometimes. Here is a
formula for horsepower that really helps the engine builder or modifier:


P is for PRESSURE - more specifically, Brake Mean Effective Pressure
(BMEP) which is the average pressure inside the cylinder.

L is for STROKE - the distance the piston moves from bottom to top

A is for AREA - of all the cylinder circles -
(bore / 2 * bore / 2 * 3.14 * numberofcylinders)

N is for RPM - what Gordon said! Double N and HP doubles!
****ALL ELSE BEING EQUAL**** which is very difficult.

k is finagle's factor - it is adjusted to accomodate such things as
whether measurements are in feet, inches or millemeters etc.
Hal Kempthorne,, Mar. 23, 1999


To get the speed increase you take the new hp's divide with the "old"
like 125/100=1,25 then the 3rd root out of that = 1,077 that's 7,7% increase
of speed with 25% more HP's.
that's if the propeller(s) is right!

Jan Carlsson @RAH, Mar. 09, 2001



All of the above cover the basics of tuned inlets and exhaust systems.
Smith gets complicated with theory of waves etc. Bell gives graphs of
measured optimum lengths for different engines. Simplest coverage and what
I think is the best read is in Campbell's book. He gives suggested inlet
and exhaust lengths at rpm:
RPM Inlet" Exhaust"
2000 46 100
3000 31 68
4000 24 50

All lengths measured from the valve. These values equate with the lengths
given in the other books.

As I understand things 4 into 2 systems can work well using the tuned length
system, but equal firing spaces are desirable, which requires the front
cylinders to be paired, separate from the rear.

JohnD,, Feb. 07, 2000





Im Allgemeinen wird die Schallgeschwindigkeit nur in Abhängigkeit von der
Temperatur angegeben (die allerdings z.B. mit der Flughöhe abnimmt):


Hierin sind
a: Schallgeschwindigkeit in [m/s]
k (kappa): der Isentropenexponent der Luft = 1.4
R: die Universelle Gaskonstante = 287 [J/(kg*K)]
T: die absolute Temperatur in [K], wobei T(K)=t(°C)+288.15

Bei Raumtemperatur (20°C = 308.15 K) also: a = sqrt(1.4*287*308.15) = 351.87
[m/s] = 1266.74 [km/h].

Philipp, de.rec.luftfahrt Sept. 27,2000


Für aerodynamische Kräfte
( F = q * c * A )
ist der Staudruck
( q = 0,5 * rho * v * v )
ausschlaggebend. Neben der Dichte kommt in der Formel noch das v vor.
Und dieses v ist die wahre Eigengeschwindigkeit.
Lars, de.rec.luftfahrt June. 07,200

If everything else is the same, it is easy to estimate the change in
rate of climb with change in power. At the same weight, and speed,
the drag is the same, so any extra power increases the climb rate.

Lets assume a propeller efficiency of 75%, so we only get to use 15
of the extra 20 hp (assuming that the 160 hp with constant speed is
the baseline).

Power = rate of doing work. 1 Horsepower = 33,000 ft-lb/min.

We are lifting a 1650 lb aircraft.

Extra rate of climb = 20 X 0.75 X 33,000/1650 = 300 ft/min.

Actual results will vary due to prop efficiency being a bit different
from the assumed value, variations in power output of engines and
variations between aircraft that affect drag.

Kevin Horton,, Mar, 24, 1999


Engine Certification

The document that governs any certification of any
aircraft engine (be it reciprocating piston engines or
turbine) is Advisory Circular AC33-2B, "AIRCRAFT ENGINE

To cut to the chase; only 150 hours are required to
certify ANY aircraft engine! And so as not to bore y'all
I'll give you the pertinent sections (brutally and
severely edited). Every combination you can think of is
covered in the manual. Single-speed supercharged,
double-speed supercharged, turbocharged, gear driven,
helicopter engines, etc. are all covered in the manual.
Prop, accessories and other good stuff are all addressed
in testing.

Section 33.49 Endurance Test

a.) General...during the runs at rated takeoff power and
for at least 35 hours at rated maximum continuos power,
one cylinder, must be...not less than limiting temp, the
other cylinders must be operated at not less than 50 deg
below the limiting temp...

b.) Unsupercharged engines. . . (1) 30 hr run...alternate
periods of 5 minutes rated take off power...5 min best
economy (2) 20 hr...alternate periods 1.5 hr @max...1/2
hr @ 75%&91% (3) 20 hr...alternate periods 1.5 hr
@max...1/2 hr @ 70%&89% (4) 20 hr...alternate periods 1.5
hr @max...1/2 hr @ 65%&87% (5) 20 hr...alternate periods
1.5 hr @max...1/2 hr @ 60%&84.5% (6) 20 hr...alternate
periods 1.5 hr @max...1/2 hr @ 50%&79.5% (7) 20
hr...alternate periods 2.5 hr @max...2 1/2 hrs max best

Ross,, Jan., 11, 1999