EAA - Experimental Aircraft Association  

Infinite Menus, Copyright 2006, OpenCube Inc. All Rights Reserved.


Tools:   Bookmark and Share Font Size: default Font Size: medium Font Size: large

EAA Experimenter

[ Home | Subscribe | Issues | Articles | Q&A | How To | Forum Review ]
[ Hints for Homebuilders | Glossary | Polls | Around the Web | Submit an Article]

Perry Harber's Glasair SII Taildragger

Nothing like a V8!

Story and photos by Patrick Panzera, EAA 555743, ppanzera@eaa.org


Father and son team Charlie and Perry Harber of Pacific Grove, California, completed their plane in their own backyard – in a single-car garage – and then disassembled it and trucked it to their hangar in Salinas, California. It first flew in May of 1994, powered with a 3.8-liter Ford V6 with a Blanton 1.6:1 redrive.  After logging about 650 hours in a little over 10 years’ time, the crank in the little V6 gave up the ghost. Following an otherwise uneventful landing, a new engine was installed, one that the team had already been working on to replace the Ford. This time, however, two more cylinders were added, increasing the displacement by nearly a factor of two in the form of a 441-cubic-inch, all-aluminum Chevrolet small-block V8.

The plan was to swap in the V8 sooner or later, but as fate would have it, sooner came first.

The original engine was a Dave Blanton Ford 3.8-liter V6 conversion, probably good for a solid 160 ponies, but the guys just wanted a few more – just as one would expect from any red-blooded male. They thought that 20 to 30 hp more would do the trick, but why settle for 20 when 100 is just as easy to make? They were already hitting the fixed-gear numbers published by Stoddard-Hamilton, 200 mph in a low fuel-flow condition, but the desire for a steeper climb attitude at a higher airspeed won out. Although 1,000-plus feet/minute was indicated regularly on the vertical speed indicator with two on board, 3,000-plus was what Perry and Charlie were after, and ultimately achieved. Cruise settings with the 1.6:1 Blanton and a ground-adjustable, three-blade Ivo Magnum propeller netted 4,200 to 4,400 engine rpm. Most trips were under 200 miles.

But it wasn’t a simple plug-and-play situation. Like any first-time engine installation in any airframe, some tweaking had to occur, especially with respect to cooling. In this case, after two to three iterations, things were under control.

The Blanton V6 was carbureted, fed by a standard automotive-type downdraft Holley fitted with the traditional McNealy leaning block. Perry was (and still is) interested in performing mild aerobatics and soon came to grips with the idea that a float bowl carburetor wasn’t the right choice. A little bit of zero or negative g’s and the engine would get real quiet real fast. After about a half dozen or so of these episodes, Perry was done with carbs. There was also the nagging suggestion in the back of Perry’s mind that these abrupt engine-outs contributed to the premature crank failure.

How It All Began
Motorcycles, go-carts, and modified bicycles were Perry’s inspirations as a kid, but he’s sure that his father’s service to his country in the 8th Air Force during World War II was how he garnered his mechanical abilities. Charlie worked in a B-17 ground crew and ended up passing along his skills and knowledge, as well as his love for aviation, to his son, Perry, who started flying in 1967 fresh out of high school. He flew for about two years before moving to Oregon where he resided for 15 years. Not much flying was done during that time with the exception of stick time in other people’s planes. Until bitten by the building bug, not much in the way of experimental aviation was experienced. Glasair had one of the first kits that caught Perry’s attention. He looked at a few others but received a chance invitation from Glasair to visit its Arlington, Washington, facilities and partake in a demonstration flight. This was in 1991, shortly after the introduction of the SII.

“I knew as soon as we took off that this was the plane,” Perry told us. A check was written and within three months a 4-foot square by 24-foot long box showed up on Perry’s doorstep. Included were all the skins, nuts, bolts, rivets, and a thorough manual; they started putting it together immediately and it’s been nothing but fun ever since – even with all of the modifications.

Perry and Charlie took 6 months off from work and began working every day, 7 days a week, at least 8 hours a day. They modified their workspace so the whole plane (without wings) could be built inside a one-car garage. The spinner ended up nesting about half an inch away from the light switch.

The whole plane was ready to go in a little over two years, but the team couldn’t find a home for it at the airport. In an act of desperation, Perry put up a notice at the fixed base operator, (FBO) and in short order space became available. The plane was moved, assembled, inspected, and flying within a few months – most of that time was spent checking and double-checking systems. They did a lot of ground testing before it flew, taxi testing, chaining the tail down, and running full throttle 10 to 15 minutes at a time, just to make sure things stayed together. Later, they did this same routine when they installed the current engine.

The father and son team proved the advertised 2,000-hour build time to be accurate. They kept track of their hours from uncrating to painting, and it worked out to be spot-on 2,000 hours. Once they hit that point, they quit keeping track, but they weren’t done; they had plenty of modifications to go.

An all-metal fixture was constructed to dimensionally represent the Glasair firewall, such that after testing, the engine mount and all could be removed and installed directly on the airplane verbatim. The engine was run on the test stand for short sessions without a propeller simply to check the operation and function of the systems. Once they were convinced all was well, the engine was removed from the fixture and installed on the airframe for the final high-power runs which were accomplished at the airport and with a the propeller installed. This system of building and running the engine on a test stand was used for both engines. After installing the new V8 in the airframe and making the high-power runs, Perry thought his brakes were failing. At full power, the plane crept forward no matter how hard Perry stepped on the brakes. Once shut down, Perry saw that he left 25 feet worth of skid marks on the ground. The brakes were working fine – he was just making too much power…if there actually is such a thing.


Flying Experience
Although Charlie’s flying experience is limited to copilot duties beside his son, he’s responsible for setting the speed and g record in N2013. They were out doing aileron rolls one day when Charlie decided to pull it into a split-S. That’s when they both found out that this plane goes real fast when pointed downhill. “The airspeed indicator just went ballistic; I chopped the power – we started pulling,” Perry told us. To commemorate the experience, the g-meter reading hasn’t been reset.


The New V8
Starting with a fully custom all-aluminum engine block from Dart Machinery, www.DartHeads.com, Perry built the engine from the ground up using all (off-the-shelf) heavy-duty racing-quality parts, including but not limited to: forged steel crank, H-beam forged steel connecting rods, forged aluminum J.E. pistons, roller rockers, and solid lifters. The block is far from stock; the deck height has been increased by a half inch and the cam location is moved up, away from the rotating mass of the crankshaft, by the same half inch. Moving the cam and spreading the oil pan rails farther apart than normal allows for extra clearance for the increased stroke of the crank. The cam bore is sized for big-block bearings, allowing the use of a high-lift cam. The piston bore is “square” at 4.125 inches for both the diameter and the stroke. Also used are 6.125-inch connecting rods, and the sum total works out to 441 cubic inches at 11:1 compression.

Dart blocks for small- and big-block Chevy (and Ford) engines are designed from the ground up for hardcore racing; the claim is that all of the known weaknesses of the factory castings have been addressed. Dart blocks are brand-new, fully machined, and virtually ready to assemble with off-the-shelf components. Extra-thick cylinder walls and decks, four-bolt main bearing caps, and a “competition” oiling system are just a few of the details put into these castings. With CNC machining and strict quality control measures, optimum out-of-the-box performance is promised. Increasing the small-block V8’s bore centerline dimension from 4.4 inches to 4.5 inches allows for an increase in cross-sectional area, providing more material around the cylinders to enhance the head gasket seal. The cylinders utilize pressed-in cast-iron sleeves, and the crank is retained by billet steel four-bolt main caps.


Ram-air from the cowl first enters the induction system through an RSA-10AD1 Bendix mechanical fuel injection servo, donated from an IO-540. Fuel pressure is built and maintained by a Lycoming mechanical fuel pump designed for high pressure use but altered by Perry with the use of a marine-quality laminated steel arm. By changing out the arm, the unit becomes a direct bolt-in application, fitting in the side of the engine block per normal. In addition to the mechanical pump, a Dukes 12-volt high-pressure pump lives underneath the seat and is merely used for priming and reserve.


The flow divider (spider), perched on top of the engine, was purchased from Airflow Performance of Spartanburg, South Carolina, a well-known high-performance aviation fuel injection source. When Perry got to the stage of needing an injection system, he called Airflow and ordered it with a kit containing enough stainless and 0.028 injector nozzles to get it running.


The addition of 0.5 inch to the deck height of the crank precluded the use of any off-the-shelf intake manifold. It’s a matter of simple geometry: As the heads become located farther away from the centerline of the crank, they move away from one another, necessitating an increase in runner length. Perry and Charlie spent almost a year experimenting before they whipped out the Skil saw, and with a carbide disk, cut the manifold in half. Then they just had to weld it back together with a bit of a spacer in between.

An intake plenum that was built to support the injector servo is bolted to the top of the modified manifold, with the intake opening facing rearward. A 4-inch-diameter SCAT tube brings the ram air from the cowl-mounted filter box rearward, then up and forward (between the engine and firewall) to the servo. 

Engine Swap
Changing one engine for another wasn’t a huge effort. Building the engine at home over a relatively long period of time with no pressing timetable allowed for a thorough job. With the original fixture being proven, the new firewall-forward package was just as plug-and-play as the original installation. The biggest obstacle was modifying the cowl, but the glass was forgiving; Perry simply put the nose piece where he wanted it and built around that. “Dad and I are pretty comfortable with fiberglass right now,” he told us.

The MSD 6A is the base model of the capacitive discharge multiple spark system offered by Autotronic Controls Corporation, most commonly referred to as MSD. The system can be triggered using breaker points, a magnetic pickup, or the output of an electronic amplifier.


Two independent MSD modules are used in Perry’s system. Each one is triggered independently and is paired with its own matched coil. The coils are then joined via a simple coil joiner, also produced by MSD. This device allows two coils to feed one distributor without cross-feeding to one another. Although both systems can be run at the same time, Perry opts to run on just one system, leaving the other in reserve. Everything in the current system, with the exception of the eight-cylinder distributor, has been reused from the six-cylinder engine.

The distributor has nothing inside except a rotor. There are no points or any other timing device. All timing is done electronically by way of a crank position sensor and the dual MSD units.

Perry observed that some builders were mounting their coils inside the cockpit, and he took note. For a short period, Perry had his installed on the engine side of the firewall, but after one of them failed in flight, rather abruptly I might add, he opted to relocate them inside, away from the serious heat and vibration usually found under a cowl. The coil failure was benign; all it took was a simple flip of  double throw switch and the engine was running again.

A 40-amp, one-wire, mini Denso alternator is slung under the engine and is driven off of the stock harmonic balancer by way of the same twin V-belts that drive the all-aluminum, high-volume water pump. Aftermarket billet aluminum pulleys are attached to both the water pump and the alternator and help reduce their rpm.

The single alternator keeps the dual batteries topped off. Each battery can be cross-connected to either ignition system for absolute redundancy.

Engine Mount
The engine mount used with the V6 is another adaptation of a Dave Blanton design. When it came time to make a mount for the V8, the same principles were used.

The engine mount is pretty straightforward, and the attach points on the engine are simple, lightweight, and easy to align.

One unique feature Perry incorporated was the use of a large pair of longitudinal tubes for ducting cooling air from outside the cowl to the accessories at the rear of the engine compartment. Sharp eyes may have already noticed the two small holes in the front of the cowl, almost looking like machine gun ports. These holes align with the aforementioned engine mount tubes, and with the use of SCAT tubing clamped to the opposite end, fresh air is delivered to the fuel pump and the alternator.


The limiting factor at this point is the prop. Perry feels that they’re at a stage now where they definitely need a constant speed prop, possibly more blades, or simply more pitch. However, they’re going to tap the brakes on that for the time being. The next plane – their half-completed Glasair III – still needs an engine decision, so until they decide on which engine, prop, and propeller speed reduction unit (PSRU) to use on the III, not much will be altered on the II.

Currently the in-flight-adjustable three-blade Ivo Magnum works well enough other than needing more pitch. Perry leaves it set at full pitch at all times.


Perry is completely satisfied with the belt-driven Blanton redrive as installed on his ill-fated V6, but when it came time for the more powerful V8, he opted for a gear drive from Geschwender. According to Perry, the only tricky part was trying to get enough oil flow into the case as the rotation direction seemed to scavenge the oil out. He first ran it with a clear return line so he could watch the return flow, which seemed excessive at the time.

Geschwender has temporarily suspended production while they try to sell the company. Perry bought their last commercially available unit.

Oil System
Valvoline 20/50 full synthetic is the lube of choice for the engine and PSRU. The full-winded custom-made aluminum oil pan holds 8 quarts. Although Perry made this pan “full-winded,” the Glasair III taildragger he and his father are currently working on will have a fully aerobatic, inverted dry-sump system.


Temperature is monitored as the oil leaves the engine on its way to the filter, then the cooler, and once again after it leaves the cooler. A double throw switch allows the use of one gauge to display either reading.

When Perry built the V6, he followed a detail from the Blanton conversion instructions to route the crankcase ventilation to one of the exhaust pipes. It worked well enough for the V6, so he tried it for the V8. For one reason or another, the suction created by the exhaust this time around was far too great, resulting in an oil bath down both sides of the plane. The answer for the time being is a more traditional positive crankcase venting system, one that has been cobbled together with the use of copper plumbing fittings. Now that the system is proven, a lighter and more refined version is in the making. “Until you prove a concept, there’s no need to make anything real fancy,” Perry said.

Filtering the oil is done by an aftermarket in-line unit made by Peterson. It’s a reusable 80-micron cartridge tied to the system via AN fittings. At normal intervals, the unit is disassembled and the element is cleaned. This gives the mechanic an opportunity to actually see what was captured by the filter.
Cooling System
Coolant enters the engine via a single hose, just like any normal automobile system, but for the sake of space, two 1-inch hoses are used to route the hot fluid to a collecting/expansion tank before being directed to the radiator. This tank also serves as the filling location.

The belly-mounted radiator is all aluminum and custom built by Davis Radiator of Tucson, Arizona, a premier name in race car cooling. Measuring 18 by 21 inches with a 3-inch core, the tanks are “split” to create a double-pass system. (Please see the radiator article by Bud Warren in the June 2009 issue of Experimenter for more info on “splitting” a radiator.)

The business end of the belly scoop.

Getting the water from the engine to the radiator and back again is handled by two 1½ -inch aluminum tubes running (exposed) along the underside of the fuselage. Future plans are to clean things up a bit by enclosing them, but as with the PCVsystem, it’s all “proof of concept” at this point.

The inspiration for the system came from reading magazines where Perry saw a photo of an experimental aircraft with the radiator slung under the belly. At that point he realized that it wasn’t really important to have the radiator mounted 90 degrees to the relative wind as previously thought. “A lot of guys think we have to go straight through that radiator,” Perry said. “We found that it’s probably not that important. You need to have high pressure and a place for it to go, and it seems to work fine.”

The underside of the belly scoop

The learning curve was steep with the V6 installation. “We just kept messing with the inlet size and the outlet part of the radiator shroud, and I guess there’s a formula for that – outlet should be twice the inlet size for expansion – so we mimicked the same thing we had on the Ford when we designed this one, with the louvers.”

Perry admits than none of his work is really streamlined – yet. With 40 uneventful hours on the Chevy V8 and no cooling issues whatsoever, Perry considers the cooling situation a closed chapter. It’s now time to clean things up and call it done.

It doesn’t get any simpler than short, straight pipes, but mounting them as they are, while good for the radiator, creates conflict with the cabin-air NACA inlets. The fix is simple enough; tape over the inlets and try alternate locations for fresh air. Right now Perry has some small openings at the gear leg junctions where they meet the fuselage. In October of 2007, when we conducted this interview and test-flew Perry and Charlie’s plane, the weather was unusually cold, so we didn’t notice any loss of ventilation with the relocated inlets. I suppose we’ll find out for sure if it’s sufficient this coming summer.

The 3/4-inch stub welded to the aft exhaust tube was used to draw out crankcase vapors, but it worked too well and had to be abandoned.

Weight and Balance
Empty weight is 1,397 pounds, which makes it close to 200 pounds heavier than advertised and about 150 pounds more than with the V6 installation, which was only 40 pounds heavier than advertised.


But with a gross weight of 2,100 pounds, there is still 703 pounds for two good-sized individuals, full fuel, and another 40 pounds worth of gear. By moving the engine a little aft (prop station is only 1¼ inches farther forward than it was with the Ford V6), moving the dual batteries to the tail, and placing the radiator farther aft than with the V6, the empty center of gravity came out perfectly. All in all, the all-aluminum small-block V8 is a good match for the Glasair II.


Safety Modification
Halon fire extinguisher (seen in the photo above) has been routed to the engine compartment. Most air-cooled engines can’t benefit from such a system as the constant inflow of fresh air will simply flush the agent from the engine compartment. But with no major cowl openings and with the Halon nozzle in the uppermost portion of the cowl, there’s some hope that in the event of an engine fire the ensuing flames can be extinguished.

At the controls, Perry brings N2013 from base to final at Salinas (KSNS), California, as Pat mans the camera.


Future Plans
As mentioned previously, Perry and Charlie are currently building a Glasair III. It too will have conventional landing gear although they did think about retractable. At this writing, an engine decision hasn’t been made, but they’re strongly leaning toward a Chevrolet LS1 crate engine and the Geared Drives PSRU. Perry isn’t willing to go with electronic fuel injection yet. “And progress on the second plane is going way smooth,” Charlie noted, “You know we did a lot of learning on this one that transfers well to the progress of our new Glasair III.”

One thing that Perry and Charlie stressed to us while we were conducting the interview and photo session was that they credit part of their success to the help from a few friends, namely Keith Peterman, Hall Cross, and Tony Davis. Through their support, physical assistance, and overwhelming friendship, the project came together smoothly in a way that was fun and fulfilling as well as educational for everyone involved.

We certainly wish Charlie and Perry well with their future endeavors and look forward to getting a little stick time with them in their Glasair III when we meet to write the follow-up article once it’s done.



Copyright © 2014 EAA Advertise With EAA :: About EAA :: History :: Job Openings :: Annual Report :: Contact Us :: Disclaimer/Privacy :: Site Map