Engine Thoughts Dec. 25th, 2000

There are only a few requirements for my very own Sonex powerplant (the following list is sorted for priority)

1. reliable. reliable and reliable

2. affordable

3. available, doable

According to above list jewels like Jabiru or Rotax (or German variants like Limbach or French JPX) have to leave the arena because of cost. What remains? -of course the VW conversions. However my Sonex is expected to become a bit overweight because of several modifications. So what I need are about 90 reliable horsepowers. This will also throw the Type I (Sonex/Great Plains) out of the game. What candidates will remain then? It was the book 'How to Rebuild Your Volkswagen Air-Cooled Engine' by Tom Wilson (available from Great Plains, ISBN 0-89586-225-5) which enlightended me.

After reading this book I wonder why anyone converts Type 1 (1600cc) VW for aircraft use. The disadvantages are too obvious:

• crack-prone, cheap magnesium crankcase
• crank and rods designed for max 50 hp
• no oil filter
• machining and lots of modifications required for more than 1600cc, however then reliability questionable

So what's left then?

...the... >>>Type 4<<<

The Type 4's were used in busses, a few 411/412 cars and in all 'VW'-Porsche 914. The Type 4 is the 70's improvement of the 'bug' engine which was a design of the 30's (like other dinosaurs like Lyco/Conti).

Advantages of the Type 4 are:

• rock-solid aluminum case
• stronger crank/rod/bearings
• larger oil pan/oil capacity
• hydraulic valve lifters
• spin-on oil filter
• increased displacement up to 2 litres, even electronic controlled fuel injection

The Type 4 in any respect makes a much more solid design compared to the Type I. The Type 4 is plentiful availabe in Germany (lots of VW busses are still running along the roads), so as soon as temperature will rise I'll visit the junkjards (x'cuse they're called 'recycling plants' in Germany) for a core, and then this exciting rebuilding job will start. Lots of specialized VW aircooled shops provide any kind of sparepart. Even Great Plains offers a conversion kit for the Type 4. So be prepeared for much more about this subject in the near future.

This great example of a Type 4 installation is from 'Hawkeye' Doug Humble who published this at the VW mailing list.

============================================

Godd Tools - Bad Tools, Dec. 18th 2000

If a homebuild is advertised that only commonly available houshold tools are required -then be very careful! particularily in the case of measuring tools. Here's a short resumee about what I have learned so far.

My tools came to 90% from the hobby market. Questions about the precision came up when I recently started to mark the large skin and sparweb sheets. Particular the sparweb I wanted to cut as precise as possible (to 1/10 to 2/10 of an mm).

First surprise came up when I checked my large (2.5m) construction aluminum straightedge (or whatever this thing is called). I put up a tightened string at both ends and -surprise - found out that this ruler got an inside bend in the middle of about 0.3mm. Too much in my sense. Now my curiosity was sparked. I checked my 5m steel tape measure against my stainless 1m ruler. The tape measure was 3mm off because the 'L' stud was loose and misaligned. So I matched the tape measure to the stainless ruler. The stainless ruler itself was x-checked and found precise - however not straight. Never use any of this flexible rulers for marking straight lines!

Same problem with cheap woodworker's angles. The one I bought was absolute crap. So I ordered some industry-quality small and large angles.I x-checked then against each other and found then precise.

So what is the outcome of this all?

• In case of measuring tools I think essential is that all measures used are identical. Absolute precision is not so significant. I think it makes no difference in flying qualities if you build a 1:1 or a 1: 1.01 scale model.
• 90 deg angles are critical for a design like the Sonex which is built essentially on the tabletop. I would accept only a few 1/10 mm per 1 m leg.
• water levels from the hobby market are almost completely crap. I use a digital (smart level) which is precise to 1/10 deg (if calibrated properly).

... and what could be trusted in?

really precise are:

• the plumb bob for the vertical lines
• a tensioned thread for the horizontal lines
• a water-filled tube for horizontal levelling.

Epilog

Meanwhile I doubt if this jig-less building is really an advantage. The RV guys construct their wing in a very simple jig. The wing is always upright which allows the use of only a string and a plumb bob. This very simple approach is superior in my sense.

A Merry Christmas and a Happy New Year for all fellow builders.

(from rainy and much too warm for the season Germany)

 ============================================

Requiem for a Wooden Bending Brake, Dec. 9th, 00

Tja, this was a flop! My small test pieces bent quite nice. Tonight I tried a real thing (SNX-F16-05). This part requires a bent flange of 928 mm lenght. The ends bend quite well, but the middle section was bad (60 deg instead of 90 deg). The hold-down as well as the hinges are simply not trong/stiff enough to do such work. This was an experience, I learned some more, but don't try to repeat this!

============================================

WOODEN Bending Brake Part II, Dec. 4th, 00

The machine is complete so far.

Here's the brake in action. The moving part is usually in a horizontal position and moves the sheet down for bending (bends the sheet around the tables's edge). The clamped beam behind is for holding down the sheet. At first I thought to put a handle bar on, but the 0.025 and 0.032 stuff bends easily by pushing the moving part down with one hand.

a closer view to the left hinge. A aluminum block is bolted on the beam using four 10cm long wood screws. The hinges are two M8 machine bolts, which I cut the threaded portion off. The remaining bolts are pushed through a 5mm steel plate into the aluminum block. This way it's very simple to disassemble the brake, just pull the hinge bolts at both sides.

this is a side view on the hinge and attachment to the table. Locating the bolt holes for the hinge is somewhat critical. The pivot axis has to be flush with the table's top edge. This took some pilot drilling and reaming. I'm fairly satisfied now.

this is the way I attached the whole thing to the table. All bolts are M8.

...and the question now: how does it work?

this brake is good for bends less than 90 deg and produces a bend radius of about 1/8". The tight 1/16' bend radii cannot be done with this setup. I will use the brake for bending the seat, the glareshield and all the channels having 1/8" radius.

The remaining bends I'll have to carry to the metal shop.

============================================

I will build a WOODEN?! bending brake, Dec. 1st, 00

No, I had no beers (until now). So why would someone build a ~wooden~ bending brake?

I already own a small brake (commercial, made from steel), about 2 ft max width. If you do not buy the Sonex channel stock but if it's your aim to do (mostly) all by yourself, then about 70% of all 'L' and 'U' channels can be bent with such a small tool. BUT there is still 30% work left. From these remaining 30% most channels are around 100cm wide. This are all the fuselage cross ties and some spar channels of the tailfeathers. Now if I could do these at home then only the very long channels (wing rearspars, turtledeck joint) would be left. This I will let the metal shop do anyway.

So a larger brake is required! The alternatives are:

• buy a new one (nope, way too expensive)
• buy a used one (nope, very few available and risky buy)
• build one (YES!)

BUT how to build one?

• steel construction (no way, too heavy , takes precious space away, welding will ruin my nice wooden floor)
• wood construction (YES)

Concept of the wooden bending brake:

the idea (like most good ideas) is simple: the small side of my worktable is nicely straight an level. It's about 140cm wide, so a brake of 120cm will left enough space left and right for the hinge mechanism. The sheet will then be bent over the radiused edge of the table. So I need a straight and solid wooden beam of about 250cm. One half of it will make the moving part of the brake, the second half I'll use for holding down the sheet. Some scrap alumn and steel and some bolts and screws will make the hinges.

At first I will radis the tables's edge to 1/16" for the tight bends. Once these are done then I'll enlarge the radius for the 1/8" bends.

And another advantage: this brake will absolutely make no scrates on the sheet metal.

Cost analysis:

bending with homemade brake:

wooden beam: DM 15

some steel plates: DM 10, bolts and screws I already have

visit sheet metal shop's CNC brake

about DM 100 per hour or so. I estimate total cost of several 100 DM

So the decision is clear. On Sunday I'll post some pictures of this project.

============================================