Flying for Fun
D.B. Mathews
909 N. Maize Rd., Townhouse 734, Wichita KS 67212
IN THE JANUARY 2005 column I used two photos of Bob Erdman's wire static models. Unbeknownst to me, one of the photos was not taken by Bob's son.
Additionally, I directly quoted some paragraphs without crediting the source. The newsletter from which I duplicated the words did not include a byline, only initials, and was sent to me by a friend on the Internet. As a consequence, I thought this was public domain and did not credit the words to Paul Fiebich, who was quite upset and asked for an apology. I sincerely offer it here.
More About Those Engines: Last month I wrote about John Thompson's fabulous nine-Cox-cylinder fake Pratt & Whitney engine. I also included a teaser photo illustrating the other two engines that I will attempt to describe this month.
The nine-cylinder radial I will cover is made from nine Saito 120 four-stroke cylinders, valves, and pistons joined onto a machined crankcase. This engine is geared, but I don't know the ratio. It turns the scale Hamilton Standard 27.5-inch-diameter, variable-pitch propeller at a top rate of 8,500 rpm.
A video of the engine being started and run shows an easily hand-started, smooth-running, 24-pound monster with a wonderful sound. Each cylinder's exhaust stack is installed in a collector ring using the stock muffler fittings. Similar to last month's nine-cylinder Cox unit, the glow plugs are lit by a common shared ring just behind the heads.
Ganged twin stock Saito throttled carburetors feed into the base of the crankcase. I don't know the low-end rpm number, but on the video this engine runs slow enough to chug away merrily. It's a wonderful sound.
As can be seen in the rear view, the engine is mounted to the firewall with four shock-absorbing rubber mounts. The Thompsons had started constructing a Giant Scale North American AT-6 for this engine, but it was not completed before John's death. Imagine what it would have looked and sounded like.
Taking a close look at the machining and welding on these engines impresses, with the exceptional workmanship, engineering, and dedication involved in this project. They fit into the "Holy smoke!" category.
The four-cylinder, horizontally opposed engine is also an incredible project, involving some clever design and fabrication. It is made from four 30cc four-stroke cylinders, which, though unmarked, I presume to be Honda weed-blower units. Carburetion is with four, what appear to be, Walbro pumped-type units. The engine's throttles are coupled to use a CH Electronics capacitive-discharge ignition system featuring throttle-coupled spark advance.
Lubrication is via an oil reservoir in the crankcase with a sump pump. It's not visible in any of the photos, but a dipstick is used to check oil levels. Exhaust is routed from the cylinders through a set of header pipes into a horizontally cross-mounted muffler.
I have performance figures of 7,700 rpm top and 600 low rpm, but no propeller information. The engine has obviously been run; note the heat marks on the heads and headers. I can only imagine the sound this thing must make and the brute power it puts out. I have jokingly said it looks adequate to power a full-scale ultralight.
These three engines are a living tribute to John Thompson's skill as a machinist and engineer, and to his enthusiasm for a hobby. One must admire a person with the determination and dedication to spend several years of spare time fabricating such intriguing projects. They surely beat a chunk of engraved marble for a lasting memorial.
John's son Mitch is selling the machinist's drawings for any of the three engines. His address is Box 513, Douglass KS 67039.
More About Gorilla Glue:
Tom Mullen of Cleburne, Texas, wrote to mention more uses for this expanded polyurethane adhesive.
"I am doing plain old fiberglass layups in a mold for the fuselage using Gorilla Glue instead of resin; this greatly reduces the weight. It can also be used for small castings.
"There are two main brands of polyurethane glue: Gorilla Glue and Elmer's. The main noticeable difference is in the foaming properties. The Elmer's brand foams less than the Gorilla brand; about 39% is foaming is an issue. Both brands are available at most building supply and discount stores.
"Paint thinner can be used to thin the adhesive and does not seem to reduce its strength, but does give it a harder bead, but not a brittle one.
"The use of rubbing alcohol will clean up any residue before it sets. Do not use denatured or grain neutral spirits, as they have high water content absorbed in them. Water is the catalyst for the glue. Just like fiberglass resin, the more catalyst you add, the faster the cure. The faster the cure, the larger and lighter.
"I have used this method in repairing a missing piece of foam; sand, then use lightweight spackle or a covering to finish it off. It is strong as new. Additional quantities of alcohol will thin it, but also cause it to foam and cure more quickly and cause it to expand more in volume.
"I even talked to a company chemist and they never have considered any other uses than as an adhesive straight from the bottle. The fiberglass parts I've made so far are about 30% lighter than epoxy or polyester resin but have a strange feel to them and are so flexible they can be bent in half, and then will pop right back to shape with no cracks. But even with a balsa stiffener, they are still much lighter than epoxy fiberglass and much more durable than ABS parts."
I am merely reporting on these techniques—not recommending them since I have not yet tried them.
More About Slick Techniques: Back when Ace R/C was producing kits for several of my designs, we received reports of stabilizer failures. Since quality of his products was eminently important to Tom Runge (the owner at that time), he investigated even to the point of paying shipping for three builders to send him tail sections of their damaged models.
In all three instances, and by then asking the others, we discovered scored indentations in the wood where the covering had been cut with a blade and removed to uncover an area of the stabilizer for gluing to the fuselage.
The indentations didn't have to be really deep—just far enough down to interrupt the wood's grain, and in this instance cross-grain. The constant up-and-down flexing that any stabilizer encounters in the propswash and during aerobatics is more than enough to eventually cause the wood to fail across the scoring.
I recently watched Terry Holley and learned from him a neat way to remove the covering in preparation for adhering one surface to another. With the horizontal stabilizer temporarily positioned on the fuselage, he used a fine-line marking pen to draw the fuselage outline onto the stabilizer covering, top and bottom. Then he placed blue masking tape a smidgen inside the lines over the covering.
With a pencil-type soldering iron, Terry melted the covering to the underlying wood and simultaneously burned through it to allow removal of the excess piece. The soldering iron produces much more heat than a trim iron and cuts the covering while melting it to the underlying wood.
Leaving the masking tape in place as a guide and to prevent getting epoxy on the areas where it wasn't wanted, Terry adhered the horizontal stabilizer to the fuselage. After the epoxy cured enough to be solid, the tape was peeled off with no tendency for the covering to pull loose.
This technique will work well in any situation where covering needs to be removed to allow adhering to raw wood without scoring it. Cutting the covering over wing-servo wells immediately comes to mind. It's a much more satisfactory and safe technique than using a blade! MA
Transcribed from original scans by AI. Minor OCR errors may remain.
