Any engine equipped with a carburetor is subject to carburetor icing. One purpose of the carburetor is to control the flow of air and fuel. The other is to cause the atomization of the fuel and to mix it with the air. The means used to move the fuel into the air is a pressure difference between the fuel supply and the air--this pressure difference is caused: 1) thru the use of a venturi that increases velocity, thereby reducing pressure; and 2) by the restriction caused by the throttle plate or slide. Fuel enters the air thru an orifice (i.e. "jet") that begins the atomization process which is then further developed as the fuel is broken into small droplets by the moving air. The amount of fuel is more or less "controlled" by change in air flow with change in throttle position somewhere in the range of 13-15 parts air to 1 part fuel.
When air is accelerated thru a venturi, the pressure is reduced as is it's temperature. When pressure is reduced by the restriction of the throttle plate, temperature is reduced. Hence air flowing thru the venturi that impacts the downstream throttle plate is colder than ambient and air downstream of the closed or partially closed throttle plate is even colder.
While the air actually cools the atomized fuel, the relatively volatile gasoline will still begin to evaporate. Gasoline is a mixture of many different fluids that evaporate at different temps. Some of the compounds will evaporate at temps as low as 40F below zero. The evaporation of the more volatile gasoline components will cool the air and surfaces of the carburetor significantly.
The amount of moisture in the incoming air will determine what happens when the air and carburetor surfaces are cooled. If the air is cooled below the "dew point" then water vapor will condense to liquid and wet the inside of the carb. If the surface/s of the carb are below freezing, then ice will form. IT IS THIS LIQUID WATER IMPINGING ON A CARBURETOR SURFACE THAT IS BELOW 32F THAT FORMS THE CARB ICE.
So, first there must be enough water vapor in the air that a reduction in air temperature causes the water to condense. Second, the surface/s of the carb must be below 32F.
The chart above shows the various temperature and humidity levels and the corresponding risks of carb ice. Note that carb ice is always more of an issue during "glide" with a closed or partially closed throttle.
Note that the difference in the amount of air flowing thru the carburetor from 40% throttle to full throttle is only 10-15% but falls off below 40% throttle. Ice forming below 40% throttle is therefore much more due to fuel vaporization than due to reduced pressure and velocity in the venturi. The reduced pressure causes a higher rate of evaporation and more temperature drop than any temp drop due to the reduced air pressure itself. A large part of the fuel in this "intermediate" and/or "idle" stage is from the low speed jet downstream from the throttle plate.
The above assumes a careful pilot that avoids flying in clouds and conditions of very high humidity.
The Tillotson HR197a Diaphragm Carburetor shown below and used on the MZ201 is a bit different than the Bing 54 Carburetor variable sliding venturi on a Rotax or other “slide” carbs like Mikuni’s on Kawasaki 440. The Bing and most “slide types” are much less susceptible to carb ice. On the Tillotson, carb ice is likely to occur downstream of the throttle plate butterfly. But, any carb where fuel is evaporated and not just atomized inside the carb is susceptible to carb ice under certain conditions.
(Butterfly throttle plate carburetors have different performance characteristics as compared to slide type carburetors. The butterfly types are thought to have better performance during the transition from 60% to full throttle. The butterfly throttle plate is less ideal with less than 50% throttle.)
To offset the heat lost from the carb surface and air to the evaporated fuel in order to raise the carb temperature there are five sources of heat input to prevent or melt ice: 1) Heat from an external source heating the air or the carb itself: 2) Heat added from air flow on the outside surfaces of the carb assuming ambient temps are above 32F; 3) Additional ambient temp air into the carb caused by opening the throttle assuming ambient temps are above 32F; 4) Heat conducted from the engine to the carb via the manifold; 5) Heat radiated from the nearby cylinder head cooling fins.
So, assuming ambient temps are warmer than 32F, increasing the load and RPM by opening the throttle will add heat and increase the carb temperature. Granted more fuel will be added, but the increased air flow at 13-15:1 ratio will overcome the heat lost due to evaporation.
Remember, during conditions when carb ice is probable, carb temps will almost always be lower than ambient temps which often results in condensation inside and outside of the carburetor. Ice can only form on the inside of the carb if it’s inside surface is below 32F,
The existence of carb ice is difficult to “prove” or document as the “evidence” tends to disappear by melting before the carb can be expected. There are many anecdotal “stories” of “near death experiences” and engine out accidents blamed on carb ice. Some of these issues may have different causes such as contaminated fuel. But no doubt, carb ice can form and can block the low and intermediate jets causing engine shutdown with a restart being impossible while in flight.
It is also possible when ambient temps are low, in high humidity for ice on the outside of the carb without internal carb ice being present.
Most General Aviation planes with air cooled engines have a ductwork system that provides air heated by the exhaust pipes directed into the carburetor, controlled with a “carb heat” control cable. This is impractical for ultralights because of weight. Motorcycles and snowmobiles sometimes have an electric heating element attached to the carburetor body. This is possible and an elegant solution but again can be impractical due to the weight of the necessary large capacity battery.
The photo below from Jack B Hart shows a HR197a Tillotson Carb with a heating element installed. Jack does not share whether it was effective.
One necessary item that needs to be added to an ultralight is a carb temp gauge. I believe the GRT EIS had an option for this readout. The sensor that can be used is similar to those used for CHT. The carb temp gauge will alert the pilot to a cold carburetor where carb ice might form. Testing will determine the correlation of the carb temp gauge reading to the actual surface of the inside throttle bore just downstream of the throttle plate. In other words, a reading of 36F on the gauge may indicate 32F inside the carb.
I conducted a diligent research effort to find any scientific or technical research on the effect of using a typical two stroke cycle fuel/oil mix. I found none. But here is my hypothesis: THE OIL MIXED WITH THE FUEL SIGNIFICANTLY REDUCES THE AMOUNT OF FUEL EVAPORATION INSIDE THE CARB. FUEL EVAPORATION OCCURS DOWNSTREAM OF THE CARB INSIDE THE CRANKCASE WHERE THE MIXTURE IMPINGES ON THE CRANK AND ROD, AND IS VIOLENTLY AGITATED BY THE ROTATING PARTS BEFORE ENTERING THE CYLINDER. IN ADDITION, THE OIL CREATES A FILM ON THE CARB SURFACES MAKING THEM LESS SUSCEPTIBLE TO ICE ACCUMULATION. HENCE, TWO STROKE ENGINES ARE LESS SUSCEPTIBLE THAN FOUR STROKES TO CARB ICE.
Carb ice can form in a two stroke engine—ice will simply form more often and more easily in a four stroke. A four stroke needs a more aggressive system to prevent and overcome carb ice.
To prevent carb ice with a two stroke engine ultralight (particularly for engines having carb with throttle plate vs a slide):
1) Avoid flying when ambient temps are in the 30-65F range when the dew point is less than 15F lower than ambient. Keep in mind that temps and humidity are often different at higher altitudes. (Avoid the red zone in the chart) Avoid flying near clouds or in hazy high humidity inversion layers or areas with light ground fog. (Generally afternoon flying is safer than morning flights as the “dew point spread” generally is higher in the afternoon.) This “rule” applies to all engines-even those with slide controlled carbs like the Bing 54.
2) Avoid prolonged periods of operation with less than 50% throttle. Pressure downstream of the throttle is significantly lower with much higher fuel evaporation cooling at low throttle positions. Periodically apply full power to maintain CHT. Avoid sudden reduction in throttle. (Prop pitch is part of the equation—cruise level flight should only be possible with more than 50% throttle and ultralight should descend with 50% throttle. My preference for an ultralight is more a “climb” than a “cruise” prop. This limits top speed, but provides other benefits including faster climbs, less engine stress, and shorter takeoff rolls,
3) During “glide descent” for landings use forward slip to steepen descent so as to avoid prolonged periods with less than 50% throttle. Periodically apply full power to “clear” engine and maintain CHT. Do not reduce power below 50% until you are sure you can safely land with a possible hypothetical engine out, prop stopped scenario.
4) Diligently monitor Carb Temp and EGT. Ice formation is often indicated by a drop in RPM and EGT.
To deal with impending ice or to melt ice formed: Raise carb temp by increasing load and RPM. In other words, full power and climb at Vx=40mph. This introduces a massive amount of warmer ambient air (if above 32F) and increases cylinder and engine temp. Move throttle to break any ice formed. Remember that the ice, if any is interfering with the low and intermediate fuel jets but not the high speed jet. Identify an emergency landing site just in case. Do not reduce throttle until a “safe” carb temp is reached, signs of ice are eliminated, and an emergency landing site is identified. Return of the throttle to 50% throttle may cause an engine out if ice is still present.
Another consideration is to use fuel with a relatively lower initial volatility. In other words a low Reid Vapor Pressure and a relatively high 10% Evaporation Temp. Combine this with a 40:1 fuel oil mixture ratio and temperature drop in carb will be reduced significantly from pump gas and 50:1 mix.
Fuel Reid VP 10% Evap
Sunoco 260 GTX 4.7 182 F
Sunoco Optima 7.0140F
Pump Gas 9-15 100-120F
Note: Remember that the formation of ice in the carb generally requires the water vapor to change to a liquid. In other words, the temperature of the air needs to fall below the dew point. When the dew point is below 20F, generally the vapor will change quickly to ice crystals and flow thru the carb with little adverse effect.
There are no regulations requiring a radio while flying an ultralight. It is prudent however to use a radio while taking off and landing from public airfields. It would be necessary to have a radio in order to ask for "permission" to land at an airport in "controlled" airspace--most likely at a Class D airfield.
The weight of a hand held aircraft spectrum VHF radio is minimal and even with it's low 5 watts of power, it is adequate for the needs of an ultralight pilot. Here is a picture of the ICOM A24 I use. Actually, the transmit power for voice is much lower than 5 watts.
The next issue is the decision of the pilot "interface". One common interface is a standard pilot's headset. (I have a simple David Clark "mono-passive" one) These require the use of ICOM's adaptor and an additional "PTT=Push to Talk button. Note the two pin connector for the radio and the dual GA plugs 0.25/0.206 dia. The small plug is for the PTT harness.
The traditional aviation headset requires a special type of helmet with clearance for the ear muffs. Another option is to use a full face motorcycle helmet with speakers and mic installed.
I prefer the full face helmet when flying in an open cockpit like an ultralight. It protects the face in an impact (accident or bird strike) and also better protects the nose and mouth from insects. In addition, the output impedance of the ICOM A24 is 8 Ohms vs general aviation’s 130. So the David Clark headset would have limited volume to the ears.
I attempted to use the Midland radio "kit", but the two pin plug is wired differently and not compatible with the ICOM. I attempted to rewire it, but the 28Ga wires were so small and fragile, I abandoned the effort and will go with a "NASCAR" stereo single 1/4" jack (RJS 600080174) setup with an adaptor to the dual GA jacks.
The "NASCAR" setup from RJS needs an adaptor cable to go from single 1/4" jack to the dual GA setup as mentioned. I ordered the one that converts the "ultralight" single jack to dual GA. The "ultralight" wiring is used by Comtronics in their headsets. The setup did not work! Lot's of research and I found a wiring diagram for the "NASCAR" wiring.
Seems that the "NASCAR" wiring uses the center "band" for the speaker and the tip for the mic. I figured this was opposite from aviation, so I disassembled the adaptor cable and re-soldered the wires reversing the green and the red wires, essentially making the center band of the NASCAR jack connect to the tip of the aviation wiring jack/plug. It worked! I tested the mic by increasing the side tone and speaking into the mic to hear myself on the speakers. I will install in my full face helmet. I will need a 1/4" extension, female on one end and male on the other to run from my helmet to the adaptor cable and radio. I may be able to reach the transmit button on the radio, but if not, I will add a PTT switch that plugs into the ICOM adaptor wiring.
In addition to radio communication with controllers and other aircraft, I found it potentially useful to also enable cell phone communication. Bluetooth technology and voice control makes this feasible. I tested these over the neck J-Lab earphones that see to fill the bill. They have Active Noise Reduction. I can actually wear them inside the speaker equipped helmet and/or the David Clark headset.
Finally, another option for the radio use is to go with a handheld "speaker-microphone" used by police officers. Radio use is in an ultralight is usually much more limited than for general aviation. For years (headsets did not become popular for general aviation until the 1980's--I learned to fly in the 1970's using hand held mic and a panel mounted speaker in the cockpit.) pilots used a handheld mic for communication, And, even when wearing a helmet, the speaker is probably loud enough with proper placement in/next to the helmet. One issue is that you would probably not hear the speaker when "stowed" i.e. when not holding it, so it would be only useful when you intended to listen and talk rather than listen only. Could be an advantage in that you would not be distracted by non-applicable radio "traffic". We will see.
In flight testing will determine how these systems work while flying an Aerolite 103.
A third option is to use wireless Bluetooth technology from Sena.
SR10-10
SMH 10-11
Smart Helmet and Speakers/Mic from Sena
Navigation
Ultralights are required to obey FAA Regulations. One of the most important is the limits on airspace that require reasonably precise navigation.
This can be achieved by using regular on line maps, including Google Maps with GPS and free VFR Sectionals at http://vfrmap.com or SkyVector.com/. SkyVector and vfrmap.com and Google Maps would only be available when connected to the internet via cellular or wifi, In flight can be achieved without a cellular data plane by subscribing to ForeFlight or free apps AVARE for Android or FltPlan Go for Android or Apple. The free apps are a bit less capable in terms of all forms of flight info, but are very good for GPS navigation for any device where sectionals can be downloaded and stored.
I carry a leg mounted device that holds an IPAD Mini equipped with GPS and FltPlan Go for this purpose. I carry an iPhone and I also have an android phone with Avare available. The phone's display is really too hard to read when flying. The leg mounted IPAD Mini works very well.
Google Maps Satellite view provides another useful feature when flying an ultralight. You can identify clear farm fields along a route to insure you abide by regulations for avoiding populated areas and always having an appropriate contingent emergency landing site. Your GPS position will move on Google Maps and you can use the highlighted "favorite" dropped "pins" as a form of navigation for safety and to insure against flying into controlled space.
Here is a screen shot below. Notice the substantial limitations on airspace from: 1) Mode C Veil; 2) York Airport Class E to Surface; Lancaster Class D and Harrisburg TRSA. Still, with a range of only 50-70 miles on a tank of fuel, the range is limited anyway. In addition, ultralights are prohibited from flying over "populated" areas. In general, minimum flight over areas of this map should be at 2000-2250 MSL which will generally be 900-1200 feet AGL. The glide ratio of the Aerolite 103 is in the range of 6-7:1. e.g. 1000 AGL would allow a little more than a mile of glide to an emergency landing site. So generally on routes I expect I might fly, I identity sites no more than 2 miles apart. Over some areas, I will be flying higher at 2000 AGL (3000 MSL) to allow up to 4 miles between sites. (2 miles from center)
Here is a screen shot. The "pin" provide a safe and "legal" flight path to various airports in the area within "range". Shoestring is my home base.
From user feedback, I have concluded that the handheld radio using the "rubber duckey" antenna is inadequate, particularly for transmission. And, because of the challenge of providing an adequate ground plane for a whip antenna, I decided a dipole type is the way to go, and there are good reports about the homemade one in this article:
This dipole while effective is quite large and I was uncomfortable with the size and the mounting to the wing strut plate that is a very thin 0.080” aluminum. In addition, in order to mount the radio on the right side of the plane I would need more than ten feet of coax. Finally, I was warned that the wing location could be subject to severe interference from the engine's high tension ignition.
So, I begin my journey learning the "art" of antenna design! First, I bought a simple SWR meter. The SureCom SW33.
The rubber duck antenna with the ICOM A24 has a very good SWR when held in the person's hand. Essentially the radio's battery "box" and the person holding the radio form a ground plane--the SWR was an excellent 1.05.
But, in the plane, the person is not holding the radio and there is no room for a proper 300 sq. in. ground plane, so some form of "dipole" hybrid would be necessary to use the rubber duck remotely mounted. I came upon this article about a "tiger tail" in my search.
In essence, a dipole leg in combination with the rubber duck. I followed his suggested length calculation of about 5% longer than "theoretical" and decided on a 24.5" marine grade insulated wire. I added three 300 ohm snap on "chokes" to the coax. I securely mounted the rubber duck on the nose and extended the "tiger tail" next to one of the structural tubes. This did not work--the aluminum tube "detuned" or absorbed the antenna's output. But, the antenna came alive when I moved it upward away from the tube to the 120 degree angle used by the earlier dipole design. I reduced the angle to 115 degrees to compensate for the change in impedance with a 10 foot coax. The SWR was an excellent 1.08.
When testing on 121.50 "emergency channel" I spoke with two airline pilots--one more the 80 miles in front at 33000 feet (a 4 degree up angle) and another about 30 miles behind. So the radio is transmitting with good range in two directions. I was also hearing traffic from a airport 12 miles away, but transmission from ground level to planes only 500-1000 feet higher, 12 miles away (less than 1 degree up) was probably being absorbed by ground terrain interference.
Here is a picture of the antenna, and the radio mounting that holds the radio and the adapters:
Note angle between rubber duck and tail is 115 degrees
Back to the drawing board! Antenna that worked great with engine off, picked up so much engine ignition noise from the high tension wiring that radio was unusable. My guess was the "hybrid dipole" wire "tiger tail" was a perfect receptor for the noise--directly from engine and through wires in dash panel mounted EIS.
I disconnected the "tiger tail" and theorized that a wire mesh in the nose could serve two purposes: 1) Block the transmission of RF from engine since it would be above the downward pointing rubber duck-sort of a "Faraday Shield" ; and 2) Serve as a ground plane for the rubber duck. My friend Ricardo, tested the SWR as acceptable--in the 1.5-1.8 range for airband VHF frequencies. It was better for slightly higher frequencies in the 135-145 mhz. (By the way, the Surecom meter's SWR was inaccurate--the hybrid dipole had a SWR of 1.8, not 1.08) I used 1/8" steel wire mesh, connected to antenna ground. I "insulated" it from airframe using a neoprene sheet and covered edges with tape to avoid cuts and inserted it underneath my "glove box bag" in the nose. The neoprene and tape also protected the clear nose poly from damage.
It worked--Radio was usable with no noise using squelch of 18-19 of 24! Hearing transmission from 12 miles out. An in-flight test determined that my voice was not load enough when transmitting with my full face helmet (good with the David Clark headset)--assumed to be because the button mic in the helmet was too far from my mouth (David Clark boom mic has to touch mouth to work well)---modified helmet with a sculpted foam adaptor, attached with velcro moving the mic to just touch my lips which should correct this.
Any additional noise reduction will have to come from shielding ignition related wiring. The NGK plugs are resistor type.
At some point I may have a custom adapter cable set up made. Currently there are a lot of connections. The ICOM to GA .250/.200 dual; .250/.200 dual to single .250 female "ultralight"; 3 foot 1/4 male/male stereo jumper; 1/4 female/female adapter; NASCAR male 1/4 to speakers and mic in helmet. I may also consider using CEP "in the ear" speakers.
Here is a picture of the push to talk mounted on the yoke, using tubing mounted to seat as a "conduit" for the wire. And the finished full face modular helmet with NASCAR speakers and mic installed.
Here is a pic of the final modifications to the helmet. Speakers slid into pockets provided by helmet maker--wires tucked under padding. The green foam is the mod for the button mic to place it on the lips when modular helmet chin piece is lowered and locked.
Adventures of Buying A “Ready to Fly” Aerolite 103 and Preparing for First Flight
Note: It is more complicated than you think!!!!!
Contacted Dennis Carley at Aerolite 103 in June and was told wait list of 12-14 months. I asked about the “show unit” from Sun and Fun and he mentioned that plane would be for sale after Oshkosh first of August, equipped with a MZ201 engine and three blade IVO prop. (From pic at Sun & Fun, it appears it showed there with one cylinder engine and wooden prop.)
On June 25, I sent my check for $32,120; full price payment in advance. (No “show” or “demo” discount!) I also sent a check for a special instrument panel to accept additional UMA ASI and ALT gauges I intended to install.
Custom UMA Air Speed Indicator with V color coded
Standard Panel
On July 1, I sent $1,000 deposit to DeLand Barnstomers for transport from Deland, Fl to Stewartstown, PA. ($2500 total)
On August 5, plane was ready for shipment, but DeLand Barnstormers called indicating his trailer had mechanical failure and he was not sure when he could pick up my plane. One week later he called and said his trailer was repaired and he was going to pick up the plane on Sunday, August 14. He was not going to leave until Wednesday as he had other commitments.
Because of the uncertainty of Deland’s plans, I had investigated driving to DeLand FL in a one way rental car and picking up the plane in a 26 foot U-Haul. Cost would have been less than $2000; less than the $2500 DeLand charged.
Finally, DeLand Barnstormers departed on Friday, August 19. The following day his trailer suffered a 2nd spindle failure and he did not arrive until 6PM on Sunday, August 21.
DeLand’s rig is huge. Too tall and long to navigate the entrance to Shoestring Airport. He had complained about “not making any money” on a hotshot delivery of one plane, so he arrived with my plane, another Aerolite for delivery in Indiana, a racing motorcycle. In order to reload the motorcycle he needed a flat area, so instead of unloading at the airport driveway, he moved up hill 1/10th mile—now 1/4 mile from my hangar. It took my crew of four over two hours to move everything into the hangar.
All in all, DeLand Barnstormers was polite, careful and competent. Plane arrived in good condition. I did not care for the uncertainty as to arrival. Also, he essentially unloaded the plane and parts just past the end of his trailer ramp and really did not help moving everything into the hangar. (I figured he did not want to leave his rig unattended with his Great Dane dog companion inside.) He did request that I provide assistance in reloading his racing motorcycle, which I did. (Like the pic below) His trailer did not have a winch that could be used to move the heavy motorcycle up the ramp to the back of the trailer.
While waiting I constructed racks to store the wings and tail while moving them and holding them in hangar for installation. I also fabricated an adjustable height stand to support wings during install. I also used existing adjustable wing supports available to all hangar occupants.
On Monday, August 22, I began the process of putting the wings on and discovered that four “saddles” for the wing struts were missing. No answer at Aerolite so I contacted their dealer Mark Murray who sent the parts USPS two day. Dennis Carley returned my call that evening and overnighted parts UPS. Parts arrived late Wednesday PM. Wings installed 8/25.
On Friday, 8/26, started installing tail and found that incorrect saddles for horizontal elevator were shipped. Correct parts arrived Tuesday 8/29. Tail installed 8/30. Dennis Carley was very prompt and responsive to my questions and issues. In defense of Aerolite 103, Carley did offer to test fly the plane before shipment but I declined as I wanted to break in the engine myself—the parts “mistakes” probably would have been corrected if they had assembled the plane after returning from Oshkosh. (The wing fabric was damaged while in the trailer coming home from Oshkosh and was replaced with new before shipment to me.)
I only had one issue with the manual. It is not as comprehensive as the one for kits and the instruction regarding the attachment of the down wires for the horizontal stabilizer was incomplete.
Not terrible having an assembled plane in my hangar about 2 weeks later than expected, but it was frustrating that the date of delivery by the transporter was so uncertain and then losing five days because of missing parts. (Sort of like Santa Clause coming two weeks late!) Still lots to do.
Next, the 62” Three Blade Ground Adjustable propeller came in a box unassembled. I was told to start with a pitch of 10.5 degrees, but instructions from IVO do not mention pitch angle at “neutral” rather, a call to IVO confirmed “neutral” is 35 inch pet revolution. That is where I set it for initial running. Torque to 200 inch pounds.
Neutral is defined by IVO as a specified distance of 1 7/16” from plate to end of adjusting screw. Tightening (clockwise) decreases
pitch by bending blade for clockwise pusher prop. Neutral is 35 inches per revolution. One full turn of the adjusting screw will change rpm by 200 -300 or about 3” per revolution. Adjustment is limited to + or - 5.5 turns from neutral from 18 to 52” or 3 to 17 degrees pitch.
Theoretical thrust calculated by prop rpm. The gear box is 2.55:1. So 4500 engine rpm is 1765 prop rpm or 5147 feet per minute, with pitch of 35”=10 degrees. 60 mph is 5280 fpm, so “cruise” is around 55 mph assuming 5% slippage; “max climb” with 20-22% slip would be around 45 mph or around best rate of climb. These speeds vs rpm will have to be confirmed and adjusted by flight testing but this looks like a good start.
Certainly 62” is max propeller diameter. Note the horizontal blade clearance.
Before weighing the plane, I decided to remove the nose wheel pant. I was thinking of best way to take hold of plane while moving it in and out of hangar. I decided the nose wheel yoke was a good “handle” but I was concerned that the 0.060” thrust spacer was inadequate, so I fabricated a stronger one and added two thrust washers. The spring appears to be 300# per inch, so with normal operation, less than 1/2” deflection is expected during landing. Deflection with pilot seated will be less than 1/8”.
Next I sat in the seat and found my head hit the overhead control switches: master, engine info, magnetos, flaps, and starter. I relocated all of the switches upward so they are above the main beam. Hard job with wings installed. Tested wearing my helmet I can move around OK..my helmet can bump into the overhead beam, but no chance that a switch will be turned off accidentally during flight. I will attach wing fabric buckles to overhead beam with Velcro and pad the areas where helmet impact is possible.
After with 1/2 “ above edge of beam
Full face “modular” motorcycle helmet
Comtronics with Aviation Headset
Not happy with lumbar support in the seat. Taking off the vinyl seat cover and soft foam, replace bottom cushioning with high energy absorbent padding and full 2.5-3” lumbar support. The cushion is only 1/2” thick to preserve headroom, but my calculation indicates it can absorb 8g force. I will also add anti-submarine crouch belts. Having experienced several auto racing impacts, I am very picky about seat ergonomics and protection for head and spine.
Cervical and thoracic vertebrae compression injuries are much more probable with a vertical or horizontal impact when sitting in a flat back seat without proper lumbar support, especially if pilot projects head forward to avoid hitting head on overhead structure.
The submarine straps were made from “soft ties” modified to get correct reach. They provide restraint to the thighs and are tightened up when the shoulder straps are tightened.
Their is clearance for helmet but not much. Still, with hard surfaces close to the head in rough air a helmet is mandatory.
Finally, plane is ready to leave the hangar, break in the new engine and do some ground taxiing. Engine started easily and ran well. First day about 15 minutes running. Second day about 30 minutes of running and much faster taxing/reaching 25 mph and very brief 4500 rpm spurts climbing the small grade between runway and taxiway. Reached 325F CHT and 900 EGT. Next I will tie plane down and run full throttle static test. I expect 5200 rpm with 35” (10 degree) pitch.
Seat and ergonomics worked great for both helmets! Rudder pedals and ailerons worked well—able to hold a very precise centerline with an 8 mph cross wind.
Received the custom ASI and ALT from UMA mentioned earlier on 9/9, but Aerolite still has not shipped the new panel. They said powder coating was holding things up, but when I asked them to ship it without the coating, it was not shipped—apparently not made yet.
I like the Hall airspeed instrument, but did not intend to fly the plane without the precision UMA ASI and although the EIS shows altitude, again I wanted the precision UMA gauge installed.
So, resigned myself to having to modify existing panel or make a new one myself. Fortunately, Ricardo Trujillo, the electronic guru at airport had a 2 1/4 in manual punch die set so I figured I could do the job, even though I would have liked to have had a shear. Careful use of a jigsaw and some filing will do. May add an overlay to existing panel for strength or just make a whole new panel.
Another possible complication. Checking out what I needed to do to remove the panel, Ricardo noticed that the EIS box was dented and a stand-off broken off, sitting inside but stuck to a piece of clear tape. See pic
Not sure where the stand-off came from, and concerned that it was the attachment point for the wire plug, decided to remove the EIS and inspect.
I also need to fabricate the mounting system for the ASI’s 1/4 inch dia. pitot tube. I decided that rather than having it stick out from the nose, exposed to damage while parked, I would mount it below the nose. Since the nose cone tapers up, by mounting the pitot tube level projecting from the rear nose shroud, it would be 6” below the nose without projecting out past. This would put it outside the boundary layer.
Accurate indication of speed and altitude are especially critical for safe flight in an ultralight, so I will need to complete the install before my first flight. (Many would argue that the Hall speed and EIS altitude are “sufficient” but as I have said, I am “fussy” when it comes to safety.)
I contacted GRT who provided valuable advice and analysis of the broken stand-off. Seems the circuit board is held in place by four screws, two of which are a stand-off/screw combo that provide female threads for two screws to secure the metal cover. The male threaded “stud” is a bit fragile and it appeared assembler at Aerolite pried on the unit or dropped it placing a lateral stress on the cover and the stand-off. To rehabilitate, I fabricated an alternative “stand-off/stud using a 1in long 4-40 screw, with 4 nuts and a washer locked to threads inside the cover, with a fifth nut to engage the cover outside.
I fabricated the overlay for the panel to accept the ASI and ALT out of 0.025” aluminum—two thicknesses added the 0.060 existing panel will give plenty of stiffness. Easier to cut the panel out of thinner material with shears.
I contoured the lower part to provide clearance for feet operating the rudder. The manual 2 1/4 punch worked well—took lots of torque to tighten the clamping bolt and a bit of filing/Dremel to get enough clearance for gauge face and to get an exact alignment of all three aluminum panels. The overlay provides strength in the area where the gauge cutout came within 1/4” of an edge.
Here it is, ready to install with Tiny Tach below the inclinometer.
Spent half the day installing the dash. Without the windscreen, it would be easy. With the windscreen most of the nylon ties need to be engaged with one hand, same for wiring and securing everything. EIS seems to work OK.
Spent the rest of the day figuring out where to mount the pitot tube for the ASI and installing it. Instead of mounting in the nose, I mounted it to the jury strut, well insulated from chafing, ran the tube inside strut fairing. Secured the tubing with nylon ties and Velcro.
The jury strut has an angle of incidence up about 10 degrees. Mounted the pitot tube level.
I tested the Hall device and the new pitot tube with a leaf blower, both reading close to same 30-50 mph readings. The relative accuracy was again confirmed after some fast taxiing.
After making adjustments to the throttle stop (set to maximum throttle position of the carb) and some initial lower RPM engine break-in operation (mostly doing taxiing limiting RPM to 4500 at 75% throttle) I readied the plane for a Static Run In with the plane tied down.
Aerolite shared some info regarding what they called "engine losing RPM" at near full or full throttle. They claimed or inferred that the Tillotson carb impulse drive fuel pump "ran out of capacity". This puzzled me a bit, but sure enough, at about 5100-5200 RPM, the engine did in fact "stumble" almost like it had a RPM Limiting Governor. I was able to reach 5450 RPM with a very gradual application of throttle.
In my opinion, the "limiting stumble" at or near max RPM is not due to the fuel pump but rather by the unique "dual venturi" design the the Tillotson HR carb. The high speed jet is controlled by a small secondary venturi. Below 75% throttle, air and fuel from this venturi flows smoothly on both sides of the throttle plate and joins the air and fuel from the low speed jet. Somewhere in the 80-85% throttle position, the throttle plate splits the air and fuel from the secondary venturi with an "angle of attack" that induces a "stall", creating a momentary turbulence. The max RPM of well designed gasoline engines is controlled by the size of the carburetor. At some point, air flow is limited at wide open throttle--even with no load. In my case, the max RPM of the engine with the 35" pitch propeller is 5450. And, to some extent may experience a "stumble" if throttle is moved from 75% to max too rapidly. For this reason, after the static run in, I readjusted the throttle stop to around 90% of max throttle plate movement and mounted a pliable material to cushion the stop. The plane will fly very well if I limit RPM to 5000-5100.
Yikes! During the run in I discovered a fuel leak at the fitting from the fuel tank to the "sight glass" system. More disturbing was that the clear "Saint Gobain C-210-A" clear polyurethane tubing crumbled like it was 10 years old. Not just at the fitting but at every location where it was compressed to mount it to the frame. C-210-A is claimed to be tolerant to fuel. The plane has a different brand of the same type, but tinted blue instead of clear, for the entire fuel supply to the carb. The tinted blue tubing appeared to be OK.
I removed the sight glass tubing and plugged the fitting from the tank. I also ordered an old fashioned fuel cap with a mechanical float controlled gauge to see if that would be satisfactory. In addition, I researched the possibility of replacing the poly tubing with race car quality steel braided, teflon PFTE lined fuel lines. Rule of thumb is that the clear poly tubing should be replaced every two years--it deteriorates over time from UV sunlight and high humidity exposure. There is no real need for "see thru" fuel lines. Some like to see the fuel move up when squeezing the primer bulb, but in reality, the best use of the primer bulb is to simply lightly squeeze until it is firm.
When the new fuel cap arrived, it fit except that the tabs needed to be bent .040” to allow for the thick 1/4” thick bung. In addition, the vent hole was only 0.060" so I added an additional 0.060" hole which should be more than adequate for the fuel flow on this engine.
Above pic is the fuel system without the sight glass using the clear blue polyurethane all the way to carb.
The new fuel cap gauge will read empty when there is still 1/2 gallon or about 10 minutes (4-5 miles) of fuel remaining. The gauge reading slightly above empty is with one gallon of fuel. (Because the gauge is a little more than one inch shorter than tank diameter, essential the gauge is reading the top 4 gallons. So when gauge reads 1/4, there should be two gallons left or at least one half hour or 20 miles of flight available.
To see the gauge, Pilot could loosen the left shoulder harness and turn head and body to see directly, or as seen thru the mirror, the pilot can glance at the rear view mirror mounted on the bar.
This eliminates the need for the several feet of clear tubing wire tied to the frame for a sight tube and reduces one source of fuel leakage.
I analyzed the venting for this cap, and determined it to be smaller than the cap that came with the plane. The gauge cap had only a 0.060" vent hole. I enlarged this to 0.080" and added a 0.125: hole in the center. i calculated this to be more than sufficient for 5-6 GPH fuel consumption. I became concerned about water intrusion and suction from air moving over the domed shape of the cap, so I fabricated a "ram air" cap cover and attached it with adhesive.
During the static run-in, I noticed that the CHT tended to be higher than I thought it should be. It rose faster than expected. I readjusted the high speed and low speed mixture setting about 10 degrees more rich and tested with more taxiing. I believe I am now running just a few points richer than ROP EGT which has lowered CHT's some and is where it should be. (This conclusion turned out to be very wrong-adjusting carb more rich fouled the plugs big time—the high CHT during static running was simply due to lower air flow over the engine.)
Cross winds and gusts have been high, so crow hopping opportunity has been limited. I have set a personal limit for maximum of a 5 mph crosswind gust component to insure that I am not distracted while the plane is off the ground during these crow hopping tests. (5 mph crosswind component is equivalent to 10 mph at 30 degrees) As the “test pilot” I want to evaluate the elevator sensitivity in ground effect.
Have now burned thru 4 gallons of Sunoco 260GTX (93/103=98 octane) There are 1.6 hours on the engine. I am mixing 3 oz of oil per gallon for a 42:1 mix. Slight dry carbon on muffler but no smoke. Oil seems to mix with green gasoline well. 260GTX is unleaded with no ethanol or oxegenates.
I found a good wind day and was able to perform five very nice take off and landings—i.e. crow hopping. But, most days the crosswind was too high.
While waiting for good conditions, I mounted my Virb Ultra 30 camera on the yoke so that I can analyze my inputs and the plane's performance during testing.
I decided to try my crow hopping after 6PM. On my first try, the engine would not run more than 3500 RPM—sputtering. Pulled plugs-they were soaked in raw fuel—way too rich mixture, made worse by low barometric pressure and a rise in density altitude. Changed plugs/ran better, two crow hops but still stuttering or what I would later learn the Aussies call “four stroking”-not firing on every stroke.
The next day I pulled plugs and they looked OK, but a tad too rich. Leaned to 1 1/8 turn on high speed jet and 1 1/4 turn on idle/intermediate. Ran two crow hops with excellent results. Pulled one plug-still a bit rich for 2500 ft density altitude, but probably just right for cooler days.
I decided that the mixture setting for the Tillotson HR197A carb was so sensitive that I needed a precision tool for adjustments.
After first more lean adjustment
General thought by many is the color or darkness of electrode indicates rich/lean mixture. Darker is richer; lighter is leaner. The insulator in the final pic is lighter, but still some signs of unburned oil and fuel.
Decent plug condition after last adjustment
The video shows a good takeoff blast-engine ran beautifully.
But, the next day, however was cold and high pressure, so density altitude was 750 feet vs 2000+ in previous runs. The engine cut out during takeoff acceleration--then able to be re- started immediately. Intuition indicated perhaps a richer mixture, but that did not solve the problem.
So, I researched the issue of propeller pitch and concluded that the suggested 10.5 degrees was not the same measurement referred to in IVO Propeller's instructions. (And that my mixture issues was related to the prop pitch being too low allowing the engine to be governed by air starvation rather than prop loading.) The 10.5 suggestion was measured at the tip. The IVO figure was the "geometric" pitch angle, usually measured at 75% of the tip diameter or 23.25" from the center of hub. I couple of iterations ended with a substantial increase in pitch with the engine still exceeding 5100 rpm during the takeoff roll. Final setting was 2.5 turns out from IVO's "neutral" measuring at 23.25" was 13.8 degrees. (Using the formula: Prop Circumference/Cotangent pitch in degrees=geometric pitch in inches) Calculated this is 47 inches. Using IVO's information, for 2.5 turns x 3.10 per turn, you get 43 inches.
The difference is most likely due to the change in propeller blade shape when adjusting pitch using IVO's design. The "average" pitch is no longer calculated at 75%, but closer to the tip. So we will use the 43 inch figure as it better matches what we saw on the engine tach. This is equivalent to a "geometric" pitch of 43 inches and 12.5 degrees.
102.2-88.4 = 13.8 at 23.25"
100-88.4 = 11.6 at tip
1 7/16 " + 2.5 turns on 1x12 threads= 1.60"
With 43 inch pitch and 35% slip, full throttle of 4700 RPM max hp should result in 48 mph climb and 70 mph top speed with 5% slip. (Flight tests indicate 5200 RPM 50 mph climb and 4700 RPM--75% throttle cruise at 55 mph. This indicates that actual "average" pitch is less than 43 inches and slip at cruise is higher than estimated. My guess is pitch is actually 40 and slip at cruise is 20%)
Running with the 43 inch pitch produced great acceleration and max rpm of around 5100 but the engine still showed signs of excessively rich mixture. After making the pitch adjustment, I used the Tillotson precision adjuster to go back to my earlier 1 1/4 (1 rev +7 marks) out idle setting and 1 1/8 (1 rev +4 marks) out high speed. (essentially this was a reduction of 1/8 turn for the high speed jet.) The Tillotson tool works great and allows adjustments as fine as 1/30th of a turn. This still did not produce desired results.
I then proceeded to estimate ZERO ROP be leaning the HS jet. This appeared to be at 1 +1/30th. I then richened the setting 5/30th to 1+6/30. (Slightly leaner than the "original" setting and most probably near the ideal 150 ROP setting.) To eliminate the the "four stroking" in the 3000-4000 range, I leaned the idle setting from1+7/30 to 1+3/30. Later flight testing indicated that a 1000 ft change in density altitude requires a 1/30 change in the high speed jet.
At this point, mixture adjustment of the high speed jet will probably need to be governed by CHT/EGT readings as well as spark plug condition.
Earlier Baseline
Final settings:
The plane is FINALLY “ready to fly” after 4.5 hours of engine run-in, dialing in and crow hopping for the pilot to be ready to fly this particular plane. Crow hopping is really “low altitude controlled flight close to ground”. This insures pilot is familiar with plane’s characteristics while in the critical ground effect during landings…controlling descent rate with throttle as well as elevator.
Some “small” changes in preparation for first real flight.
First, my Garmin VIRB Ultra required a special housing to mount it in a way that it’s view is adjustable. Somehow, I lost the factory version and Garmin discontinued accessories when the stopped production of the VIRB a couple years ago. Fit my crow hopping, I mounted the camera on the yoke but I had to orient it vertically, missing out on the view of the Hall air speed indicator. I recd a new housing from a Chinese producer and made a bracket so that the camera could be mounted horizontally. The added height allowed the camera to be a sort of yoke position indicator as the “sight picture” changes—nose level puts top of camera in line with bottom of tach—yoke back too far brings nose wheel arm to come into view.
Yoke level flight
Yoke back (nose up)
I also added a “yaw string” as a redundant yaw indicator. Yaw strings are more sensitive than inclinometers and are viewed “heads up” without having to look down at the panel. The yaw string points in the direction from back to front to the rudder needed to reduce yaw--same rudder that "step on the ball" indicates on the inclinometer. I later shortened the string so that it hung from the mast without dragging on windscreen,
