Sunday, July 22, 2007

Basic Jugs - I

HEADZUP! The fins on your cast iron cylinder barrels are brittle. Drop one, the fins are going to snap off and you're gonna have to buy another set. So put down some cardboard. And be willing to sacrifice a toe if a jug gets away from you.

After-market VW jugs come in two basic flavors and a variety of sizes. The two flavors are A-types, meaning they're to be used with the stock crankshaft throw, and B-types for stroker cranks. A-types have a greater crown height; the distance between the center-line of the wrist-pin and the top of the piston. Stroker cranks shove the piston right out of the bore. By using a lower crown height the piston sits lower in the bore and will not protrude as far, meaning you can use a thinner spacer under the jug, your heads won't be pushed so far away from the center-line and your valve-train geometry won't be as badly out of whack. Of course, that also means the pistons will sit slightly deeper into the crankcase at BDC, where the piston's skirt may interfere with the flange of the opposing cylinder. So B-types also have shorter skirts in that area, although not short enough for a really aggressive stroker. As the Mechanic-in-Charge, part of your job will be to ensure there is no interference.


Today, buying pistons by mail order is a crap-shoot. To see why, go read...

When buying pistons what you wanna do is stand right there at the counter, open the carton and inspect the color-dots on the tops of the pistons. If all four don't match, don't buy. It's hard enough building a good engine with good parts; it's virtually impossible if you start with bad parts. And mis-matched jugs are bad, bad parts. (When you do find an honest dealer, buy as many sets of jugs as you can afford. [Figure 1] Good investment if nothing else. [I'm sad to announce that OVW will be closing its doors on 25 Dec 2007. Nancy will continue to accept drop-orders but the friendly little store with its honest and honorable people will be no more.])

But let's say you've found a suitable set of jugs, no broken fins (feel them), color-dots are okay like the set of Mahle's in Fig 2. (Yeah, they're cast. No, they'll do fine at propeller speeds.) The first thing you want to do is number them. Use a marking pen or crayon to put a big, bold number on the inner flap. This is called a work number; it doesn't have anything to do with the engine's method of designating cylinders. But the pistons are matched to your barrels and the rings are matched to your pistons. You can't let them get mixed up. And they will if you don't mark them. (See Fig 3)

Go find a container that will hold and protect four pistons. Turn on your air and rig a die-grinder with a narrow cut-off wheel. (No got? Then use a rat-tailed file.) Find your vibrating scriber and have it handy. (No scriber? Then get a round-nosed punch. [Make one out of a nail or something.] ) All tooled-up? Okay, pull a jug outta the carton and use a hammer-handle to push the piston out of the cylinder. Put the piston into the plastic bag and fold it over on itself; the next step calls for spraying abrasive grit around and we want to keep it off the piston. Find the flat section of fins and use the file or die-grinder to transfer the jug's work-number to the jug. It should look like Fig 4 when you get done (only prettier). That is, three notches means '#3,' one notch means '#1,' and so on. Before you put it back into the box, mark the piston. Herezhow:

Lookit the face of the piston. Find the arrow. That shows you which way the piston has to be installed on the engine. Turn the piston over and scribe the work number on the wrist-pin trunnion under the arrow. Herezwhy: If you're building a really good engine you're going to have your valve heads, combustion chambers and piston-tops treated with a ceramic-metallic thermal barrier coating. When the parts come out of the oven the coating will have obscured any markings on the crown of the piston -- you won't know which way to install the thing until you measure the off-set of the piston-pin. So you put your work-number on the trunnion that must point toward the flywheel-end of the crankshaft.

So... write the work-number onto the underside of the piston under the arrow, put it back into its plastic bag and put it into the second box you've provided. Now you can put the barrel back into the carton. (But there you are, die-grinder at hand, and you just know you could do a better job cleaning up the parting-line flash on those fins.... Don't. No air flows across the parting-line. [Think about it.] It doesn't matter if the fins are open or closed at that point. Grind on the things and you'll just be spraying lotsa abrasive grit around.)

Okay, got the gen? Then do the same procedure to the other three jugs and whistle when you're done; I'm gonna go cop a smoke.

You'll love this next step! A chance for you to exercise your artistry in paint. Herez whatcha need: Some flat black paint. If you don't have any, make some by mixing a tad of naptha with glossy black paint. If using Rustoleum Flat Black in the half-pint can as shown in the photo, you'll need to thin it with about an ounce of mineral spirits. But before opening the can, make sure it is at room temperature. Then shake it for at least two minutes. Don't guess; check a clock and give it an honest two-minute shake-up. Provide yourself with a stirring stick, open the can, add the mineral spirits and stir for at least one minute.

Obtain an inexpensive 1" paint brush; something cheap enough to throw away after using. (Why? Because it the cost of the mineral spirits needed to clean the brush is more than the price of a new brush. Using a pair of heavy shears cut off about two-thirds of the width of the bristles. You've now made a fin brush :-)

As you can see in the photos I've threaded the barrels onto a piece of extruded angle supported at both ends; a broom stick or piece of plastic pipe would work as well. Do not use anything that can scratch the barrels.

This particular set of pistons & cylinders was free of cosmoline. Some sets are not. If the jugs are greasy they must first be washed in solvent to remove the grease.

All set? Then go ahead and paint the cylinders. Try to keep the paint off of the machined surfaces. Use a paper towel dampened with mineral spirits to wipe off any mistakes; the machined surfaces must be perfectly free of paint when we assemble the engine. Removing it now is easier than removing it later.

It generally takes me about half an hour to paint a set of jugs. If this is your first set, you won't need that long. When you're done, wrap your paint brush in plastic and allow the paint to cure for about half an hour. Now take a strong flashlight and inspect your work for holidays. Surprise!

Okay, so it's a bit harder than it looks. (Which is why it takes me about half an hour.) Unwrap your paint brush, stir up the paint and do the job properly.

Once the jugs are painted it will take a day or two for them to dry sufficiently to be baked. Baking hardens the paint. It also causes it to shrink. If the paint is not fully dry, baking will cause the paint to crack, rust will form in the cracks and your engine will look like hell. And run hotter than it should, since rust makes a dandy insulator.

Another reason we must have a good paint job is because the next step in prepping our jugs is to give them a bath, complete with lots of scrubbing and a boiling water rinse. If you've missed a spot with your painting, your jugs are going to start to rust even before you get the engine assembled.


NOTE: You also need to check the length of the barrels within the set. All should be the same to within about .001" This may be checked with a surface gauge (ie, you don't have to actually measure the things -- just make sure they are all of equal height between the sealing surfaces). If the height is out by no more than .0015" you can live with it. Up to about .003" you can shorten the three longest cylinders. But any error greater than .003" is simply too much work; it would be best to find a more accurately made set. When checking for height measure at least three points of the circumference. You will occasionally find a barrel in which the sealing surfaces are not parallel to each other. If you have a shop full of equipment you can re-machine the barrel to make it square then adjust the length of the other barrels to match. But on the whole, you'll be miles ahead if you start with more accurtely made parts. -- rsh

Friday, July 20, 2007

Crankcase Painting

The Volkswagen crankcase is cast from a magnesium alloy (about 96% magnesium). A critical characteristic of magnesium is that it’s highly reactive; it likes to corrode. Upon manufacture the crankcase is usually treated to a chromate bath but the protection is defeated by time and heat.

Since a properly assembled engine based on VW after-market components can give twenty years or more of reliable service, it’s vital that the magnesium alloy crankcase be given some form of protection. For the home-builder, painting the crankcase has proven to be the most practical solution.

Although any good oil-based enamel will serve to protect the case, all forms of paint act as thermal insulators. To preserve the function of the crankcase as a thermal radiator a thin coat of flat-black paint will provide the best results, since flat black has a higher thermal emissivity than any other color.

If flat black paint is not available you may use gloss black. Mixing a small amount of naphtha or even gasoline (!) with the paint will kill the gloss.

Paints intended for high-temperature applications should be avoided. Often called ‘stove paint,’ barbecue’ paint or ‘exhaust’ paint and advertised as being able to withstand temperatures as high as 1200 degrees, these paints get their high-temperature qualities from clay, metallic salts or ceramic frits, all of which make excellent insulators. The use of such paints will reduce the engine's ability to rid itself of waste heat.

Common oil-based enamels can withstand temperatures up to about 400 degrees Fahrenheit, far higher than your normal crankcase temperature.

When painting the crankcase it’s important to keep the paint where it belongs, which is on the outside of the crankcase. This can be accomplished by careful brush-work. If using spray paint, you should mask any area you don’t want to paint. As a general rule that would include any sealing surface or threaded bore.

When building just a single engine the masking is usually done with tape. Threaded bores may also be protected with corks, plugs of various types and even dowels. However, people who normally build more than one engine at a time often make up a set of re-usable masks. Some of these can be quite elaborate but I've found cardboard to work well enough.

In Fig 3 you can see the front of the crankcase with the cam plug and the #1 main bearing area masked off with cardboard salvaged from a cereal box (I think :-). Dowels are used to mask the tapped holes to the lifter oil galleries while the main oil gallery has been sealed with a small cork.

The cardboard is held in place by rubber cement. Applied to the cardboard and allowed to dry, it remains tacky enough to stick to a clean crankcase, peeling away without leaving any residue after the paint has dried.

Whenever possible I try to use an existing part or gasket for the mask. In Figures 4 & 5 you can see the anti-splash baffle being used to mask the dynamo base while an old sump plate takes care of masking the bottom. Fig 4 shows a couple of corks and another dowel. Normally, I use an old distributor body and a fuel-pump block-off plate as masks but they've wandered off so I used masking tape.

The crankcase shown here was painted with Rustoleum Flat-Black in a rattle-can, which makes it rather expensive. An air brush works equally well and a quart of paint will do a dozen engines or more.

At my location it takes the paint about a day to cure well enough for it to go into the oven, where it will be baked at 170 degrees for four hours, after which it's remarkably bullet-proof. If you don't have a shop oven that will accept a crankcase you can line a cardboard box with aluminum foil large enough to fit down over the engine and rig a 100W incandescent lamp to go under it. The rising heat will be trapped by the box, raising the temperature high enough to harden the paint in about 12 hours.

If you live in a warm, sunny climate you can also pop the painted crankcase into a parked car standing in the sun. It generally takes two or three days for the paint to achieve the hardness and scratch-resistance it gets from being oven-baked for four hours. (Even a black car will reach thermal equilibrium at about 145 degrees F, a bit low for optimum paint-curing.) But be warned: As the paint cures a greasy residue will appear on the inside of your car's windows :-)


Wednesday, July 18, 2007

Crank Basics - V

Bagged and swaddled, the crank shown in Fig 1 gets laid gently in the back of my 1965 VW bus and we putter off toward Escondido. Traffic on Highway 78 is moderate so the eleven mile trip takes only twenty minutes with just two sessions of stop-&-go. The old bus has no trouble keeping up with the 300hp punkin seeds darting from lane to lane looking for what doesn't exist. Highway 78 is a linear parking lot about twelve hours out of 24, virtually empty the other twelve.

Don at HDS looks the crank over, we discuss the job, he writes me out a ticket and I'm free to return to the six-lane insanity. Barely an hour after leaving the house, I'm back.

As you can see from the photo the crank was delivered fully mantled save for the bearings. On one end of the crankshaft is the propeller flange and spool from Great Planes, on the other is the Harley Davidson permanent magnet dynamo, installed on a hub of my own design and manufacture. Drawings of the rotor hub and the stator mount will be posted to the Chuggers Group when I get around to it but previous posts will give you a hint as to what it looks like.

No one answers the phone in southern California any more. We live in what has become an up-scale ZIP code, targeted by politicians, telemarketers and evangelists to the tune of two dozen calls a day. We spin through the messages every evening and after a couple of days HDS leaves a message saying the crank is ready for pick-up.

This time I take surface streets. It's a few miles farther but takes less time.

The job-ticket is translated into a bill which I take to the office for payment. The yellow receipt is the crank's Ticket of Leave and Don haul's it out to the bus for me because it's kinda oily and I'm wearing a spiffy Hawaiian shirt. We chat for a bit, discussing the rattle from the spacer and the fact the assemblage was relatively clean, meaning he didn't have to do much work to remove the wobble. In fact, on the first run-up it was within spec for a stock VW crankshaft (8 gm/cm ) which he reduced to about .1 gm/cm. Once it's mounted in the airframe I'll go through the same procedure with the prop. No one ever believes how much this benefits the power-plant until they actually do it.

"Nice shirt," he sez, pretending to go blind. I've known him about twenty years; have three more crankshafts coming his way. I'll wear one of my really gaudy ones next time.

Back home via Stagecoach Road and Twin Oaks Valley, I lug the crank into the shop and start writing on it. The crank has survived its Rite of Passage. It is no longer just a crankshaft, it is part of an engine, serial number HVX0381, which I scribe onto one of the flanges with a carbide burr. I get one of the little red Engine Logs from the cupboard and start filling in the blanks. The log book will go into a big zip-loc along with the other documentation. Now I get to take the sucker apart for the second time, the first having been to trial-fit the main bearings, during which I also clocked the case to determine what size cam-gear I'll need. Fitting the cam-gears is one of those unimportant details the shade-tree types like to ignore.

Due to normal tool wear and tolerances, the distance (and sometimes the alignment) between the crankshaft and the camshaft varies slightly. The difference is small but significant, since it involves a gear-train. To accommodate the differences Volkswagen used nine different sizes of cam gears, from a -4 thru zero to a +4. They started out with just four sizes; -1, 0, +1 and +2. But factory-overhauled engines often required align-boring, which lead to the other sizes. The markings are on the inner face of the cam gear so as not be confused with the o used for the timing alignment.

The Factory Manual will tell you what checks to perform to see if you've got the proper cam gear. Or you can read all about in my two-part article "Dialing in Your Cam" that appeared in the 2001 October and November issues of 'VW Trends' magazine. (Sure to be a collector's item :-)

But right now I was busy taking the crank apart and writing, etching or stamping '0381' on all the bits & pieces. Well.... most of them, anyway.

In Fig 2 you can see how little metal Don had to remove to achieve zero wobble. (There's a matching patch on the opposite end of the crank, on the other side.) Grinding away that amount of metal meant the crankshaft had to be fully dismantled, all the plugs pulled then cleaned to within an inch of its life. Or whatever. I used lacquer thinner in a wash-bottle as my solvent, plus a variety of brushes, one of which you see in the photo.

After being scrubbed and scoured the crank gets blown dry then undergoes a visual inspection, where you poke a grain-of-wheat lamp on a wand into this hole here whilst peering down that hole there to make sure there's nothing in the hole but hole.

Then you get to put it back together again, only this time with bearings.

Since this crankshaft is counterweighted it has about four more pounds of flanges than a stock crank. When balancing a flywheel or stock crankshaft the usual method of removing metal is to drill it out. That isn't always possible with a counterweighted crankshaft because the outer portion of the flanges is fairly thin, which is why the metal is removed by grinding instead of drilling. Figures 3 is the 1/3 flange of a new stock 69mm crankshaft that has been factory balanced, (Meaning it's 'way the hell out of balance by modern-day standards.) Fig 4 is the 2/4 flange on the opposite side. Such divots are the balancer's spoor, telling you that particular component has been balanced to... some standard or other. The stock crank is destined for a 1968 bug. It will be presented to the balancer wearing a flywheel & clutch cover on one end, all of its gears, and a stock steel fan-belt pulley on the other. It will then be balanced as a complete assembly and to modern-day standards. In return, the engine will typically produce about 10% more power for the same amount of fuel.

This article marks the end of the series on crankshafts. I'll probably cover installation of the rods using a different engine. Right now, I want to complete work on the crankcase and start prepping the jugs. And there's still a pair of heads waiting to be prepped. (Good thing I got all this time on my hands :-)


Sunday, July 15, 2007

Crank Basics -IV

The crankshaft goes down to the magnaflux shop bare. That’s so they can peer down into the Woodruff keyways and the bottom of the groove for the circlip and, most important on a Chinese crankshaft, into the groove between the pulley hub and the #4 main bearing. That’s because on Chinese crankshafts this groove has square corners, whereas it’s a smoothly radiused curve on real crankshafts. The snap-ring groove on Chinese crankshafts also wanders around a bit.

If you look closely (click on the picture), Fig 1 will give you some idea of what I’m talking about. That’s a Chinese stroker on the left (of course :-) and a stock Germany crankshaft on the right. But the scary bit is the square corners in the groove behind the pulley hub (Fig 2). Square corners act as stress concentration points and standard manufacturing practice, especially with hardened, high-alloy steels, is to avoid them. Figure 3 is a close-up of the German crank for comparison

- - - - - - - - - - - - -

After the crank comes back from the NDT shop, assuming it’s usable, the next step is to have it balanced. Which isn’t to say it isn't already balanced. But proper balancing calls for the crankshaft to be spun-up with with all of its attachments. That means we have to assemble the crankshaft, except for bearings, from the prop hub right through to the dynamo’s rotor, torque everything to spec and deliver it to the balance shop.


The first step in assembling the crankshaft is to install the large Woodruff key then the cam’s driver-gear, a spacer, the distributor’s bevel gear and the snap-ring that prevents the world from coming to an end.

The steel driver gear and brass bevel gear are a shrink fit to the crankshaft, meaning we have to heat them. The spec calls them to be heated to 80 C. or about 176 F. Everyone gets them hotter of course, assuming that if a little heat is good, a lot of heat is better. It’s not, but there you are.

Figure 4 shows a very plebeian method of properly heating your gears. Put about three inches of water in the big can, the gears in the small can, then spoon in enough lard to cover the gears. (The curiously bent wire is how you retrieve them.) Put the can full of gears into the water and put the big can on a stove and bring the water to a boil. Let it simmer for about twenty minutes then carry the whole shebang over to the bench holding the crankshaft, fish out the cam gear, slide it on (bevel toward the flywheel, please) and seat it with a few carefully aimed blows using a brass drift. Slide on the spacer then fish out the scroll gear and slide it on. Get rid of your tin cans & boiling water and install the snap-ring.

I donno why the thought of using lard makes so many people laugh. If you got something against pigs you can use salad oil. Or motor oil. The idea of the water is that it has a high specific heat – it will keep the parts hot while you carry them back & forth. Plus you can do all your heating with a ‘hobo’ stove and two charcoal briquets... or a camp-fire, if it comes down to it. The key is that you need a pretty good heat sink, because the minute you remove the gears from the heat they start to cool off. The shade-tree types solve the problem by heating the gear to the point where the steel changes color... along with its physical properties.

You don’t have to ruin your cam gear by overheating the poor thing. It slips on just fine if heated to the specified temperature, assuming it’s actually at that temperature when you slide the thing on.

When I start assembling a crankshaft I put the gears atop a big hunk of aluminum (Fig 5), pop them into a little oven I’ve got out in the shop and set the dial to ‘200' (which is as low as it goes). When I’m ready for the gears I drape a welding glove over them, carry the block over to the bench and install them. The block of aluminum will keep the gears hot for about ten minutes.

No aluminum? Then use steel.

No metal at all? So use a brick, fer crysakes. And if you don’t have a brick, use a pie-pan filled with sand. Or rocks. Or whatever else that can serve as a thermal mass.

And about here you’ll see that lard (or whatever oil you prefer) and a coffee can of boiling water isn’t quite as funny as it seems.

Once you’ve got your gears installed you can assemble whatever else goes on your particular crankshaft. Figures 6 & 7 is the crank for an engine that drives the prop off the flywheel end. The rotor for a permanent-magnet type dynamo is attached to the pulley hub. Everything is indexed; there’s only one way for it all to go together. It will be spun up as an assembly, resulting in a smoother-running, more durable and more efficient engine.

(The spruce lath is to keep it from rolling away. The prop flange and hub are from Great Planes. The cam gear has been treated with a solid-film lubricant.)


Saturday, July 14, 2007

Crank Basics - III

(Be sure to read Crank Basics I & II)

Crankshafts from China, supposedly forged, supposedly of SAE 4340 steel, having been coming into the marketplace for some time now at prices ranging from under $200 to over $400. Same cartons with the same cranks inside them but from different retailers.

Inside the box you'll find the crankshaft wrapped like a mummy. The plastic wrap and cardboard box is its only protection against the rock & roll of an ocean crossing, to say nothing of the hazards of UPS, should you order one. (Stock, replacement crankshafts from Europe are shipped a molded styrofoam block.)

The plastic wrapping is in the form of a sheet. The crank is first rolled in a bundle of the stuff, the ends neatly wrapped in upon the crank, then thirty feet of the stuff is spun out like a pig's intestine to make a plastic rope, which is bound around the mummy.

Your first chore is to unwrap the mummy without dropping the thing on your toe. Unwrap because cutting this stuff is not only a thankless task, it leaves a hell of a mess. So try unwrapping before reaching for your sword.

Once unwrapped the mummy turns out to be a fully machined, counter-weighted Volkswagen crankshaft fitted with eight dowel pins in the Porsche/SPG pattern.

Using a new gland nut as a gauge, you should check the threaded bore in the flywheel-end of the crank. Ditto for the pulley hub, using a clean pulley-hub bolt. Use new Woodruff keys to gauge the keyways milled into the nose of the crank. Finally, use a snap-ring to test the width of the snap-ring groove.

As a point of interest the crankshaft shown in the photos weighs almost exactly 21 pounds whereas a stock crank weighs about 16-3/4. Longer rods and bigger jugs also weigh more than their stock cousins but the weight difference between bone-stock and the largest big-bore stroker is rarely more than ten pounds.

After cursory checks of the threads, keyways and weight, suspend the crankshaft and perform the ringing test for cracks. (Just give it a light tap with a small hammer. It should ring like a bell, where a cracked crank gives a dull clank.) Using a good light and a magnifier, if necessary, give the crankshaft a close visual inspection paying particular attention to the radius of the corners and the lips of the oiling holes. A simple radius gauge filed out shim brass will prove handy here. (The specs for radius and dimension are printed on the form mentioned below.)

In the Chugger's Group files archive, in the crankshaft folder inside the Engine file, you'll find a Crankshaft Check Form. Print a copy, add the date, the serial number of the engine and any details about your micrometer that may be germane. As with all professionally built engines, the blueprint check-list will become a part of the engine's documentation package, along with a copy of the crankshaft's bill of sale, balancer's report and magnaflux report. The package need not be elaborate but it must be complete.

A set of standard bearings will serve as a quick check for main bearing journals 1, 3 and four.

With bearings 1 and 4 installed on the crankshaft you can use one of the crankcase halves to support the crank while you measure the run-out at the #2 bearing. The alternative is a good surface plate and vee-blocks.

To mike your crank you'll need to make a minimum of two measurements per journal and record the results. Provide yourself with good light and sufficient time; this isn't something you want to rush through. (The magnifying glass is because I often forget my eyes are nearly seventy years old :-)

Once you've miked the crankshaft, stand it on its nose using a spare pulley then bag the thing. A large zip-loc bag can be tightened around the nose of the crank to provide a dust-free environment suitable for a few day's storage. For any longer period the journals need to be protected with grease.

If you're satisfied with the crank's dimensions the next steps are to have it magnafluxed and balanced.


The crankshaft shown in the photos was a bit of a disappointment. Main journals 2 and 3 miked 2.1642", which is the low end of acceptable specs, while #1 miked 2.1640" -- two tenths below spec. Assuming it magnafluxes okay it can be used but the bearing clearances are going to be more than I like. It will definitely be happier with 40W rather than 30.

The final step before cleaning & use is to attach everything that will rotate in the same plane as the crankshaft and deliver it up to a competent balancer.


(Ed. Note: In case you were wondering, this is not the same crankshaft as shown in Crank Basics - II.
These crankshafts are often advertised as being balanced and magnafluxed, made from high-alloy steel and so forth. Yet none of the two dozen or so I've examined have shown any of the marks characteristic of balanicng. )

Monday, July 9, 2007

Crankcase Ventilation

The Volkswagen engine holds exactly 2.5 liters of oil, which ain't much. Modify the engine for full-flow oil filtration you could count on the extra quart or so but back in the Good Ol' Days, whenever that was, you could blow better than a quart an hour when you were flying low through the sage brush. It wasn't uncommon to catch & pass your competition when they pulled up to pour another quart down the spout. And for them to do the same to you a few miles later. No big deal, since everyone had the same handicap.

Except for Charlie. He flew out of San Quintin running sixth overall and managed to pass everyone in front of him by the time he reached Catavina. He was better than ten minutes up when he blew through San Ignacio and would of won for sure if he hadn't tried to plow a cow north of Villa Insurgentes.

So what was Charlie's secret? He'd stuffed his dynamo tower with copper 'Chore Girl' pot scrubbers. They created a near-perfect labyrinth separator. No blown oil meant no stops to top-up.

Every crankcase has to breathe. You gotta provide some place for air to flow in and someplace for it to flow out. If your engine's running hot a lot of oil vapor will be mixed with the out-flow. If you don't do something about it, you'll blow it overboard. Which is less of a problem running the Baja than it is flying from here to there. Running the Baja, you can pull over & park.

On the Volkswagen engine the crankcase ventilation inlet is the annular gap around the pulley hub, which was machined with an Archimedes Screw to pump air -- and oil -- into the engine. Running off pavement you've gotta seal it up or you'll suck a lot of sand & grit into the sump. The outlet is via the dynamo tower, which usta have a road-draft tube that extended below the engine so your forward motion would produce a slight negative pressure at the outlet. (Later engines plumbed the outlet to the air cleaner, using the carb to provide the negative pressure that ensured good ventilation flow.

Once you'd installed a sand seal you had to provide a new ventilation inlet to the crankcase. Most guys plumbed a filtered line to their valve covers.

Since flying Volkswagens don't have dynamo towers you see all sorts of methods used to deal with crankcase effluvium; a lot of guys don't use any kind of oil separator. But then, a lot of guys still don't believe in oil filters, mom's apple pie or that Cheney is pulling the strings :-)

My approach was to try and use what's already there, such as the little shelf just below where the dynamo tower attaches, as shown in Fig. 1. The shelf is pierced with an opening over against the wall of the crankcase. (See Fig. 2 ).

To keep the mesh in place you'll need to tap a couple of 8-24 holes along the parting line and install a couple of drilled-head bolts to serve as a fence. (Fig. 3)

You want the mesh to pretty much fill the D-shaped hole for the dynamo tower but not to bulge above it. (Fig. 4) That will allow you to use an inexpensive oil pump cover as your outlet. Nor do you want it hanging out the bottom. This is the cam-gear gallery and the last thing you need is to have your mesh get sucked into the gear train.

If you shop around you can generally find a cast aluminum oil pump cover for a full-flow oil filtration set-up. These come with a threaded outlet.

As an oil pump cover, cast aluminum is about as durabile as a politician's promise. Enormously popular of course, at least to the Kiddie Trade. But as soon as youngsters see how rapidly an aluminum cover can wear they go for a cast-iron pump cover, which is why you can often find cast aluminum covers on sale. The two shown in Fig. 5 are idential, purchased for about $5 each. The one on the left is as-received, the one on the right has been bead-blasted then treated with a thermal dispersant. (Tech Line's 'TLTD').

On the flip side of the pump covers (Fig. 6) you can see the location of the outlet hole. Also note that one of the pump covers has been treated with a solid lubricant coating. (Another Tech Line product, although I don't recall it's name at the moment.) Coated in this fashion the pump cover holds up about as well as an anodized cover and can serve as a repair part. But it also makes a dandy crankcase ventilation outlet :-)

Since dynamo tower studs are closer together than oil-pump studs you'll have to open up the bolt holes in the pump cover. (See Fig. 7) You won't need the gasket nor the deflector plate. On my engines I run the outlet down to the carb-heat box, where the carb provides the required negative pressure.

Although I originally used copper 'Chore Girl' scrubbers for this application I found they corroded pretty easily. For the last fifteen years or so I've been using stainless steel pot scrubbers, found in the paint department of the local Home Depot. Unfortunately, they are larger and springier than the copper jobbies which makes installing one like trying to shove a watermelon up a monkey's ass.

Over the past fifty years or so when converting a VW engine for flight I've found it best to keep things simple. I'm not an engineer and don't pretend to be one. But there's heaps of sound engineering out there, much of it embodied in components that are commonly available. Before resorting to something I have to buy and bolt on, I try to see if there's some feature already on the engine that can resolve the problem by simply configuring it differently. Such as the providing adequate crankcase ventilation without blowing all your oil overboard.

By devoting a bit of thought to the various problems involved in converting a car engine for use in an airplane I've managed to come up with reliable, inexpensive solutions, most of which can be easily duplicated by fellow home-builders. This approach is wildly unpopular, of course. Not because the engine's don't work but because Americans seem to have gotten out of the habit of thinking for themselves.


Wednesday, July 4, 2007

AV - Chugger's Progress - III

The ribs of a fabric covered wing are interesting things, their role in the lore of flight often as confused as the plot of a Russian novel. The ribs themselves are nothing more than sticks of softwood. Sitka Spruce is the preferred material because it offers an excellent ratio of weight to strength but it has become so expensive that most of the people who would like to fly can no longer afford to do so.

The shape of the rib is blocked out in a jig which holds the sticks in place while gussets are secured to the joints with glue and tiny nails. The nails are so small they are usually positioned with forceps or needle-nosed pliers before being driving into place with a small hammer.

This method of construction, using glue and nails to secure some kind of stringer to some form of plywood, dates from the earliest days of aviation and is known to provide light, strong and durable structures. But cults devoted to the particulars of this method have sprang upand their partisanship often obscures the basic goal. The Wood Cult insists the only stuff suitable is Sitka Spruce, the Plywood Cult says only a particular type of plywood can be used and there are similar cults for the nails, how to install them, remove them and so on. And we haven't even gotten to the Glue Cults.

In modern-day America flying is all about money. If you have a lot of it you can fly; if you don't, you can't. Low-cost alternative materials and methods are out there but to determine their suitability for airplane construction you're pretty much forced to emulate the Wright brothers and conduct your own experiments. But even that is discouraged. After all, there's no need to experiment now that we have all those helpful, friendly Cults. Of course, if you can't afford to follow the cultist's advice, why, then you simply can't afford to fly.

I don't agree with that. I'm not an engineer but I do quite a bit of experimenting in which I compare inexpensive materials against cultist-approved stuff. In doing so I've learned there's no such thing as a Bad Experiment. Things often fail in unexpected ways but there is knowledge to be gained from such failures since they tell you 'Don't Go There.' In fact, most of my experiments are failures in that the Cheap Stuff rarely equals the performance of the Expensive Stuff, with the rib shown in the opening photo being a good example. Although it will bear the same weight as an identical rib fabricated from Sitka Spruce and aviation-grade plywood it weighs more than twice as much, a clear Failure according to cultist criteria.

Of course, in a more practical vein, that horribly heavy Cheap Stuff rib, weighing-in at a massive 3.88 ounces cost less than fifty cents, whereas it's more elegant Cultist Approved cousin, at a svelte 1.8 ounces, cost more than $5. Or to look at it another way, the plane would gain three and a quarter pounds while the builder would save over $100.

Oddly enough, when it comes to Flying on the Cheap, cost isn't the only factor. See all those tiny nails? It takes about half an hour to assemble such a rib, regardless of the materials you use. Rich or poor, when building an airplane your time has no value but poor people tend to have less leisure time than rich people. Even if I could come up with inexpensive alternative materials the sixty-eight million or so Americans who earn less than the median $28k per year generally don't have the luxury of enough spare time to build an airplane. That meant methods of saving time as well as money are a valid area for experimentation. As a result of those experiments I have largely replaced aircraft nails and tiny hammers with other fasteners and different methods.

I recently described using a pneumatic brad-driver in a post to this blog that was meant to offer a bit of comic relief about the Evils of Experimentation. ( ) I was surprised by the vehemence of some of the messages it produced.

Didn’t I know that aircraft nails are always 20 gauge? (about .025"). Pneumatically driven brads are always 18 gauge or larger. Didn't I realize that a brad that size would split a cap-strip of Sitka Spruce to flinders?

Fig 1 will give you some idea as to what all the fuss is about. The 20 gauge aircraft nails are above the pocket-ruler. Below it is an assortment of 18 gauge brads, which are rectangular in cross-section and about .035" x .048". That's a fairly massive spike to drive into a quarter-inch cap-strip.

I’ve been experimenting to find out how large the stringer or cap-strip has to be to prevent splitting. But not when driven into $7.95 Sitka Spruce, which is actually rather forgiving when it comes to nails. I was more interested in learning what happens when you use pneumatically-driven fasteners in the far more brash $1.98 Douglas Fir and even lowly .89 cent Western Hemlock ( the stuff Bernard Pietenpol used for his spars ).

The fact the brads are rectangular caused me to wonder what would happen if I took the grain of the wood into account. So I made a test-spar -- just a pair of stringers with a shear web of luan ply -- and then broke it under controlled conditions. (That's the spar mentioned in the earlier article.)

My mention of foamy glue, meaning polyurethane, caused the Glue Cult to point out that it was not certified for use in aircraft and suggested I try T-88, which was ‘stronger.’ The fact T-88 is also uncertified somehow escaped their notice, as did the cost and availability. In fact, most of my experiments are with Weldwood ‘Plastic Resin,’ which does happen to be a certified adhesive. But the point missed by the Glue Cultists is that all modern adhesives have a higher shear strength than any of the softwoods normally used in aircraft. While there are many reasons to use a good epoxy, strength isn’t one of them.

The tricky bit with a water-mix glue such as Plastic Resin is that that it requires a significant amount of pressure due to the water causing the wood to expand. Polyurethane glues require a similar amount of pressure because the glue itself expands whilst curing. When properly clamped and cured each of those adhesives is stronger than the wood and while ‘Plastic Resin’ is far less expensive, polyurethane is more convenient to use, especially when doing spare-time experiments.

Some of my experiments were to determine how many brads are needed to achieve the required amount of pressure, along with what type of brad was best for different combinations of plywood & stringers, and how small the stringer could be before the orientation of the brad was significant.

Even the pneumatic brad-driver caught a bit of hell from the Cultists. Nails were meant to be driven by hand; pneumatic tools were evil incarnate, the airplane would fall apart and famine and plague would surely follow. Clamps are apparently okay. Ditto for wedges. But nobody used pneumatic brad-drivers. (Splits the wood, you know.)

Oddly enough, no one mentioned nailing strips, a standard method when applying plywood skins. Which caused me to wonder if the people who took the time to write were merely expressing Conventional Wisdom – stuff they’d heard was bad, rather than stuff based on their own experience.

Another fact offered by the Cultists was that aircraft nails were always removed. Because of the weight, according to one; because they work out of the wood, according to another. And of course, they always rust.


Perhaps the most important thing I've learned from my experiments is to ignore the Cultists. I'm sure their interest is well-meant but I'm equally sure it is misguided. The goal is not to build an unsafe airplane but an inexpensive one.

As a point of interest I've found most of the information offered by the cultists, while originally based on a kernel of truth, to be incorrect for most modern-day materials. For example, pneumatically driven brads come in sizes even smaller than aircraft nails and are available in stainless steel as well as bronze. And they are less expensive than aircraft nails. (My grandfather called them cigar box nails.)

Below you'll find a couple of photos of a test-piece assembled from hemlock, luan plywood (ie, doorskins) and 23 gauge pneumatically-driven brads 5/8" long. When driven atop a steel plate the brad neatly bends itself over and locks into the wood in a manner remarkably similar to the copper brads used on a clinker-built hull. On this end of the piece the spacing was rather random but for stringers up to 3/8" wide the glue-line exceeded the strength of the wood when fastened with one brad every two inches.

Chugger's fuselage is little more than a tapered box, plated with plywood forward, covered with fabric aft. The typical structural member is 3/4" square hemlock or fir. To ensure adequate strength I made samples of the various joints used in the fuselage and tested them to destruction. Experiments included plain joints, joints having filler blocks, joints having gussets on one or both sides, and joints having both filler blocks and gussets. The latter type of joint lead to some interesting problems in that I could not get it to fail with the test equipment I had at that time. When I was finally able to load them to the point of failure, the nature of the failure was unlike anything I'd seen before. A number of iterations proved the unusual mode of failure was not related to the test sample, which lead me to digging through books on structural engineering, where I eventually found that mode of failure described. Over all, it has been an interesting is rather slow education.

A basic tenet of flight is the ratio of strength to weight but materials having the optimum ratio, such as Sitka Spruce and aviation-grade plywood, are expensive. When building on the Cheap, using commonly available materials to achieve the required strength causes the resulting structure to comes out heavier than one built of aviation-grade materials. Experiments tell us how far we can go with our weight reduction efforts before the structure becomes unsafe. Experiments also show how the use of non-traditional tools can result in a significant savings of time when assembling a complex structure.

The Chugger will be inexpensive compared to virtually all other designs. And it can be assembled rather quickly. The result may not be very elegant but it will not be unsafe.