VG-21

Maintenance Item # 17

9 May, 1995

FUEL SYSTEM:

FUEL TANKS: There is one 17.5 Gallon welded/riveted aluminum tank in each outboard wing for as you know, a total of 35 gallons. These tanks and all their parts except the filler neck were manufactured by Varga. The most common problems are leaks at the fuel sender flange, a loose bulkhead rivet or a cracked weld seam or, an accumulation of sloshing compound at a quick drain or a fuel outlet.

A leaking sender is usually easy to fix by either tightening the flange screws or replacing the flange gasket which is available at most aircraft supply houses.

If one of your Curtiss CCA-1550 quick drain valves stop working, most likely you will find it plugged with sloshing compound. You will have to drain the tank (remove the quick drain valve) and either clean it or replace it. You should also disconnect the fuel outlet hoses from that tank, remove the finger strainer fittings from the tank outlets and clean the strainers.

Loose rivets and cracked welds are a bigger problem. First make sure that red or blue fuel residue is not just from a leaky scupper drain (see SCUPPER below). If not, you’ll have to disconnect all attaching hardware, remove the outer wing, drill out rivets to remove the center wing rib, loosen the tank support straps and slide out the tank.

The rivets are special jacketed types (P/N 44401ADJ4-5-4) installed before the end was welded on. If you have a leaking rivet you have two options that I know of. (1 ) You can carefully grind off the weld that holds on the end of the tank, drill out the offending rivet and replace it (lf you can find a jacketed rivet to replace it with. We also sealed each rivet with #242 Locktite on later airplanes), clean out the sloshing compound, reinstall and reweld the tank end, check it for leaks again and reslosh it or, (2) Clean out the sloshing compound with a solvent (acetone may work), weld around the rivet, check for leaks and reslosh it. NOTE: We always checked for leaks with the tank in a pressure cage, 3.5 PSI air and soapy water before sloshing the tank because we knew if we had a leak, the sloshing fluid would not always stop it. Unless you have or make a pressure cage to keep from blowing up the tank, you will probably have to check for leaks by filling the tank with fuel.

If you have a cracked weld, clean out the sloshing compound with a solvent, weld the crack, check for leaks and reslosh the tank. Then stick the tank back in the wing, put everything else back together and wish yourself luck.

FUEL VALVES: As you know, there is one shut-off valve for each tank. This Imperial Eastman brass plug valve is still available from many sources including Aircraft Spruce and Wag-Aero. Lee Beery gave an excellent description of maintenance procedures in Maintenance Item #4. A 3/8” Imperial Eastman brass/teflon ball valve works much better (I have one on my airplane) and requires no maintenance but some modifications are required and it is not an FAA certified installation.

GASCOLATOR: Edo-Aire made the G199-105 gascolator that was used on the Varga. It may still be available from the manufacturer who bought the production rights from Edo-Aire, but if you need repair or replacement parts, I don’t know who that is. There are other (much cheaper) gascolators available that will work (l have one on my airplane), but none have the same attachment or are certified on the Varga.

ELECTRIC FUEL PUMP: This is a Facet #476087 fuel pump. I don’t know if this part number is readily available but several aircraft supply houses sell similar Facet fuel pumps and they may be able to get one for you. I’m not sure if another Facet part number will work because, on the Varga, the electric fuel pump is in parallel with the mechanical engine driven pump, not in series, and there may be something different about this fuel pump that is important.

FUEL VENTS: Fuel vents are lines 1/4 aluminum tubing with 1/4 ID flexible hose at each joint. As Lee Beery mentioned in Maintenance Item #4, fuel will siphon out of the vent system if the tanks are filled to the top. As far as I know, Lee’s remedy of running the vent lines out to the wing tips has yet to be tried. Bill Morrisey and I discussed doing this on my airplane on a 337 or getting an STC, but my airplane is experimental and it needs to be done on an airplane with standard certification.

FUEL LINES: All of the fuel line is 3/8” aluminum tube with standard 37 degree AN fittings and standard aircraft flexible fuel hose. CAUTION; Replace your flexible fuel hose within 10 years or 1000 hours, whichever comes first. In 1982, a mechanic removed the gascolator from my 1968 Cessna Cardinal to check out a problem and the rubber fuel hose attached to it was so brittle it broke off in his hand. Don’t let yours get that bad. Otherwise, I’ve heard of no major problems with fuel lines on the Varga.

FUEL CAP/FILLER NECK: The fuel cap is a standard unvented fuel cap similar to those available from most Supply houses. The only problem I’ve ever had with one of these was a bad gasket causing the cap to be loose. The filler neck is also similar to those sold in supply houses.

SCUPPER: The scupper is the depressed area around the recessed filler neck on Vargas. It is there to contain and drain away spilled fuel. If the rubber hose that couples the scupper outlet to the 114’ drain line is old and cracked, it may leak fuel onto the tank and make you think you might have a tank leak. Check this rubber hose first before you pull out the fuel tank.

FUEL GAUGE SYSTEM: The fuel gauge system has apparently been a major source of complaints for many Varga owners. Although I probably shouldn’t take the complaints personally, I was the engineering and quality control manager at Varga Aircraft responsible for this system and I’d to say a few things about the system that might shed some light on the subject.

All airplanes left the factory with a tested and functional fuel gauge system. Every fuel gauge was inspected with a test device provided to us by Rochester Gauges. The fuel gauges were tested at Full, 1/2 and Empty with precision resistors and were considered to be acceptable if indicating within a tolerance of one needle width, except at Empty, which had to indicate a maximum of Empty. All fuel level senders were measured with a regularly calibrated precision ohmmeter at Full and Empty positions and had to read a maximum of 0 ohms at Empty (down) and 15 ohms + or - 1/2 ohm at Full (up). After the airplane was assembled but before fuel was installed, the fuel gauges were read to insure that they indicated empty. The gauges were read again when the aircraft was filled with fuel. If any discrepancies were noted, the system was adjusted or repaired before the airplane was delivered to a customer. Company and FAA flight tests checked the fuel system for operation on every airplane.

During the certification of the Model 2180, we had to recertify the fuel system. This included fuel tank capacity, acceptable fuel flow for the larger engine, the amount of unusable fuel and the accuracy of the fuel gauge system. There were a few interesting things that we discovered during this testing. One was that the forward fuel outlet and line on fuel tank was unnecessary and perhaps undesirable. In a steep glide angle with very low fuel, the aft fuel port would uncover and draw air. We removed the forward fuel line on the Model 2180 but went out of business before we made the change to the Model 2150. Another item of interest was the fuel quantity test. When a tank was filled with fuel by quantity, less than 17 gallons were installed, but when weighed and its weight adjusted for fuel temperature (it was a hot spring day in Phoenix), it was calculated to be exactly 17.5 gallons, the approved amount. Also, although the manual says there is unusable fuel, in reality, when the engine stops, there is NO fuel left in the system. Lastly, a test of the fuel quantity system was done by filling a tank on the flight test aircraft in incremental calibrated amounts and reading the gauges. This test showed that the gauges could be off by more than 10% at 1/4 and 3/4 levels but were accurate at Empty, Half and Full positions. At the time, this was acceptable to the FAA, I assume, based on their experience with other fuel gauge systems.

One thing of concern to the FAA was that when the gauges finally said empty, fuel was exhausted within only minutes. They thought (and we agreed) that the pilot should have more time than that. On fuel systems manufactured for Varga SIN VAC16O-80 and later, two changes were made to increase the amount of fuel left when the gauges read empty. The first was to shorten the arm on the fuel level sender to give it less travel and more clearance for the float at the top and bottom of its travel. In other words, the sender float bottoms out before it gets to the bottom of the tank. The second thing we did was rotate the tank fuel level sender mounting flange inboard(away from each end) so that we no longer needed right hand and left hand sending units. The effect of these changes was to reduce the accuracy of the fuel gauge system at Empty, but give an Empty indication well before the tanks were dry.

Early Vargas had the same fuel gauge system that was installed on the Shinn, including some fuel system parts built by Shinn or its subcontractors. I don’t remember when we started to run out of Shinn engine clusters (fuel gauges) and senders, but I believe Rochester Gauges (RG) was the only vendor that was willing or able to make replacement parts for us. Although its been said that there were no problems with Morriseys or Shinn’s, it may be interesting to note that one reason we were running out of original inventory fuel gauges and fuel level senders is because we were selling some of them to Morrisey or Shinn owners as replacement or repair parts.

This continued to be a problem for Morrisey/Shinn owners even after we switched to the RG parts because the RG fuel gauges wont fit in the AC Spark Plug Division (AC) engine clusters, it was hard to find someone who could repair the AC fuel gauges and a replacement RG engine cluster cost more than $100 (the AC cluster only cost Shinn about $17), which some owners thought was way too expensive.

GAUGES: Fuel gauges are a part of a RG engine gauge cluster which is similar to those used in Beechcraft airplanes and modified for use on the Varga. They are really just 0-14 VDC voltmeters. Early Vargas used AC Spark Plug Division engine clusters that were inherited from Shinn inventory.

SENDERS: Rochester Gauges also made the fuel level senders used on most Vargas. Early Vargas used Stewart Warner (SW) units from Shinn inventory. The fuel level senders are just potentiometers that vary resistance from 0 to 15 ohms. When we ran out of Shinn inventory, they had to be made special for the Varga because most standard senders were made for one sender per gauge systems and have a higher resistance range. The Varga has two senders connected in series for each gauge.

The RG and SW units are electrically identical but have some mechanical differences. The RG senders are aluminum and the SW senders are steel. One problem we had with the RG senders is that the floats and float arms are also different. Both types were bought as one part number and the float arms were bent to make them right and left hand parts. Although the RC senders were made within the tolerances that were given, the RG floats were a little thicker and the arms had to be adjusted to within a closer tolerance range. This meant that if the fuel tank warped or canned too much from the welding process, the floats would sometimes stick to the sloshing compound at the top or bottom of the tank. When this was discovered on an airplane during assembly and inspection, the sender was readjusted to fit. This problem was reported by Varga owners as a tank that would never read MUCH LESS THAN HALF FULL (a float stuck to the top) or never read MUCH MORE THAN HALF FULL (a float stuck to the bottom).

As designed, both fuel level senders are insulated from the fuel tank with RG #722 Nylon Grommets and a rubber gasket and are wired in series to give a total resistance of 0 to 30 ohms. The outboard sender is grounded at its flange with a wire running to a ground stud attached to a wing rib.

Another problem reported by Varga owners as a NO MORE THAN HALF FULL INDICATION (or always empty, if the wires are crossed) is caused by a grounded inboard sending unit. The screw attaching the ground lead to the inboard sender is much shorter than the flange attaching screws so that it won’t touch the fuel tank when tightened all the way. If a longer screw is accidentally installed, it will go through the gasket, touch the tank and ground out that sending unit so that it is the only one being read (the gauge sees 0 to 7 volts instead of 0 to14 volts).

An ALWAYS EMPTY indication is probably a bad gauge, shorted wire to the sending units, a shorted inboard sending unit or both sender floats stuck to the bottom of the tank.

An ALWAYS FULL indication can be caused by a break (open) in the electrical system (wire, sender or gauge) or, more unlikely, by both sender floats being stuck to the top of the tank.

Another problem is that the senders can just wear out and be faulty by being internally grounded shorted, giving a less than full range reading or internally open, giving an always full reading.