The curved top of the tank did cause out-of-plane warping which was more than a match for my metal shrinking skills (nil). I don't have fluting pliers either, so I modified the form block with a depression between each rivet hole on the top edge. After placing the side panel back on the form, I hammered flutes into the depressions. The sides are nice and straight again.
2 hr
[ add comment ] ( 2 views ) | permalink |
( 3.1 / 19 )Using the mylar patterns, I cut out the tank side panels and bottom from 0.040 5052 aluminum sheet. I roughed the parts out using a snips, then cleaned up the cuts with a laminate trimmer bit in the Dremel following a straight edge (or spline in the case of the curved top pieces).
3 hr
[ add comment ] ( 2 views ) | permalink |
( 3 / 19 )I designed and printed my wing tank pattern on mylar (dimensional stability). The tank is designed to fit above the drag bracing diagonals. The top of the tank forms the top skin of the wing. The tank fits inside a 14" wing bay, between the front and rear spars.
Each wing tank has a capacity of 7.25 gallons. In level flight at the stall angle of attack the sump will unport with just under two gallons in the tank. The aircraft will also have at least a 5 gallon header tank, so this is not a big concern. The tank should completely drain under cruise conditions.
The drawings do not show the location of the sump drain, drain fitting, filler fitting or vent fitting. I have purchased a flush type cap from Aircraft Spruce. I also bought my tank fittings there.
2 hr
[ add comment ] ( 2 views ) | permalink |
( 3 / 25 )After a considerable period of indecision, I have decided to make riveted aluminum fuel tanks for the RW-11. The wet wings of the RV series is made like this. R.S. Hoover (AKA VDubber) posted an informative article covering the basics. Several builder sites illustrate the steps involved. It looks do-able.
I thought about forming a "lost foam" mold out of polystyrene then wrapping it with glass impregnated with epoxy. But then I'd need to make a hotwire saw to get the airfoil shaped plug. I'd have to get set up to use laminating resin. All lame excuses, but it just didn't appeal to me.
I also considered a welded tank. Tom Marston (spelling) who flies a Dakota Hawk out of New Richmond, WI suggested these. He had his made for him, but I didn't care for the price. I do have an acetylene set-up, and I've watched the Kent White videos. With practice, I could probably manage it. However, I would like the top skin of the tank to form the top of the wing for that bay. I am not convinced I can make the tank without warping the top skin too much. Also, I am almost certain to face the frustration of leaks.
So I arrived at a riveted tank. A top skin, two rib shaped sides, and a wrap that covers the front, bottom and rear of the tank. Add a couple of baffles, a flush cap, and couple of fitting flanges for vent and outlet and it should set to go. The bottom of the tank is a couple of inches above the bottom of the wing to allow the drag/anti-drag strut to pass underneath. Each wing tank should have a volume of 7.5 gallons. Combined with a 5 gallon header, I should have 3.5 hours of fuel on board.
I was able to find 5052-H32 sheet aluminum in 0.032, 0.040, and 0.063 thicknesses from a local sheet metal fabricator. 5052 is the recommended material for fuel tanks. I'd read of people using everything from 0.025 up to 0.050 for tanks. 0.050 was recommended for top skins, lest they get banged up during filling operations. After feeling the 0.040, I decided that was plenty heavy. A 4x10 sheet cost be $90. I could have gotten away with 0.032 for everything but the top, but that would have cost me another $60, and saved me less than a pound.
I also ordered 1/8" Avex rivets - the same used on the CH601 - and a ProSeal clone. More as I move along - hopefully with pictures.
[ add comment ] ( 2 views ) | permalink |
( 3 / 45 )I spent the evening modifying the bell housing from my Corvair core. The manual from William Wynne, as well as half a dozen web sites, show how it is done. I used a bandsaw and then files and a grinder to clean mine up. Here is a picture of mine, ready to be blasted with walnut hulls:
1 hr
[ add comment ] ( 2 views ) | permalink |
( 3.6 / 40 )I've been toying with the idea of having fuel injection/electronic ignition for the Corvair. A month ago or so I went ahead and ordered a MegaSquirt kit. I got a v2.2 board with MS-1 controller. I've installed the MSnS-E code which has all the features I need. Resolution on the MS-I is plenty adequate for a motor that only turns 3000 RPM. The V3 board might have made some of the wiring for wasted spark simpler, but I can make do with the older board just fine.
Ultimately, I want two controllers in the airplane for redundancy. The weight from additional injectors is minimal, so I can actually have redundant fuel systems as well.
The last few nights I have been soldering the controller and stim kits together. The unit is now working flawlessly on the bench. I have the spark side set up driving three coils in a wasted spark arrangement. The injector section is controlling two injectors, one per head.
This eliminates the distributor, so long as you get one of those plugs that allows a modified dizzy shaft to remain. The shaft drives the oil pump, so it must be kept in place.
If I can work out a cam sensor, I can have a pseudo-sequential system that I can trim the injector pulse for individual cylinders. This allows better control of the AF mixture and smoother operation lean of peak. If I do a good job porting the heads though, this may not be a big deal.
About mixture control: I have spent a lot of time digging through Lycoming and Continental literature, articles pro/con lean-of-peak, CAFE reports, SAE journals, and numerous other articles. Here is some of what I learned. Modern cars run a stoichiometric mixture in most situations. This is what the narrow band O2 sensors are designed to operate at, and is where fuel controllers (MS included) are tuned to operate. This is because a stoichiometric mixture, generally speaking, has the least amount of noxious emissions in the form of unburned hydrocarbons or NOx formed by the excess O2 that can combine with nitrogen at high temperatures.
This stoichiometric mixture corresponds to peak EGT. This isn't where we typically run aircraft engines, however. In fact, above 75% power, aircraft engine manufacturers specifically prohibit AF ratios near stoichiometric. In the past, manufacturers specified well rich of stoichiometric - 200 degrees "rich of peak" or so. Lately some manufacturers have come around to the idea that lean of peak is OK too. Right at peak is hard on an engine. Combustion dynamics are ripe for detonation and the temperatures are high, weakening the metal parts.
Somewhere rich of peak is where the engine makes more power for a given slug of air. For climbout on a hot day or to just go as fast as possible, you may want to have this setting available. Somewhere lean of peak is where the best efficiency is in terms of BSFC. You'd want this setting to get the most distance out of a tank of gas. So we have two different operation points, but the MS does not have a mixture control like we are used to in aircraft, so our control is limited. However, MSnS-E code does have switchable fuel and ignition maps, so with the flip of the switch, you can transition from best power to best economy. This is the path I have decided to try.
All that said, Mark Langford has a narrow band O2 gauge in his plane, and has been running close to stoichiometric with no apparent ill effects. This is according to his web site, and when I get closer to turning a prop with this set up, I will pick his brain about the details. He has spent a lot of effort making his engine detonation resistant, so he may be afforded some leeway in mixture setting.
[ add comment ] ( 2 views ) | permalink |
( 3 / 146 )I won a 1965 164 CI 110 HP Corvair engine off Ebay for about $40. I've also purchased a 95 HP engine of the same year from a guy in Milwaukee for $150. By the time these two engines have had the necessary changes to convert them to aircraft usage, they will be essentially the same - producing 110 HP on takeoff, and 90 HP continuous.
[ add comment ] ( 2 views ) | permalink |
( 3 / 38 )
Calendar
