Sunday, February 25, 2024
Home » How To's » Glow To Electric Conversion Tips – Part II

Glow To Electric Conversion Tips – Part II



Last month, I discussed a generic 10-step process to convert a glow model to electric. This month we will deal with the specific things you may need to do with any model, but concentrating specifically on the Hobbico Avistar.

This is the original Glow O.S. .46LA engine and muffler installation. The replacement electric motor actually weighs much less than the glow! The difference though is made up in the battery and ESC, but the final e-power conversion actually weighs less than the fully fueled glow model!

From that article, we had decided to use a BP Hobbies BL4120-7 outrunner motor, an APC E 12 x 6 prop, an E-flite 30C 4S 3200mAh LiPo battery pack and a Castle ICE lite 50A ESC. Not explained last month, but relevant now, is that other than swapping out the old 72Mhz receiver for a Spektrum 2.4GHz receiver, I left all three fullsize Futaba servos in the model.

When I received the gift RTF Avistar from a coworker who no longer wanted it, I knew that from seeing this model previously in our club, it was a fine flyer. My original intent was not only to convert it to e-power, but actually bring it to the field to train new pilots.

With the all the original glow related equipment removed, the area below the wing is quite empty. Only the original rudder and elevator servos remain, supplemented with a new 2.4GHz receiver.

The first step was to measure the CG position as received and mark it on the bottom of the wing (fuel tank empty). I checked it against the Avistar instructions and it was right about in the middle of the range specified. Then, I began to pull out all glow related hardware: engine/muffler, engine mount, tank, throttle servo and pushrod. I realized quickly that a nose-gear modification was going to be required because the upper end of the nose gear wire penetrates the now removed engine mount. I found an appropriate size nose gear bearing block in my nylon scrap bin, cut off the top 1/3 and screwed it to the fuselage. That was one step that did not take long!

A battery hatch was cut into the area just forward of the windshield to allow access to the battery without removing the wing. Note the magnet and how the hatch extends past the old firewall for easy removal.

With the selected motor in hand, I now needed to decide how to mount the motor. The supplied hardware (steel X mount) and the length of the motor (shorter than the glow O.S. .46LA engine) required a new firewall about one inch forward of the stock firewall. I could have fabricated one-inch long stand-offs from aluminum or steel tube spacers, but because the sidewalls of the fuselage were exposed wood, I decided just to glue a new firewall across the sides. If your model does not have strong wood sidewalls, then you can make a box extender out of plywood and glue it to the old fuselage as an alternate method. The 3/16-inch firewall was cut out and blind nuts installed for the Outrunner motor. The inside surface of the Avistar, forward of the firewall, was nicely coated with fuelproof epoxy. This should be sanded or filed rough for good adhesion of the new firewall. Approximately oneinch long balsa spacers were glued to the back side of the firewall to properly position and angle the new firewall with the same slight down thrust of the old. It also affords a bit more gluing surface for the new firewall. With the motor installed, the motor wires jump over the new firewall and through the stock hole on the old firewall used for the fuel tank tubing.

With the old glow engine and mount removed, the upper portion of the nose gear is left unsupported. The top third of an old nylon nose gear bearing block was used to support the nose gear wire after removing the glow engine mount.

 As I did not want to have to take the wing on and off each time to swap out batteries, I decided to make a hatch in the upper section of the forward fuselage. A hatch area was cut out as wide as possible, and a new top deck made from 1/16-inch ply and 1/16-inch balsa laminate constructed. A shelf for the new hatch was made by gluing 1/4 x 3/8-inch spruce sticks to the underside of the remaining top deck. A tongue for the aft part of the hatch and a magnet to hold down the front were installed. The hatch was covered with white MonoKote. Note: to make it easy to remove, the hatch protrudes over the old firewall by about 1/4 inch.

The original installation for glow power. Soon to be removed are the foam wrapped flight pack battery and receiver, along with the throttle servo and pushrod. A newer 2.4GHz receiver will be used along with the Castle electronic speed control.

 The LiPo battery needs to be securely mounted inside the old fuel tank area. Access to the ESC/battery connector must be made through the hatch also. I decided to put a false “floor” in the battery bay to raise up the battery so the ESC could be mounted below. This floor was made from 1/8-inch lite ply. The wires from the ESC to the battery can be run up the sidewall, or all the way aft under the floor and forward again, if the supplied leads for your ESC are long enough. A small 1 x 1-inch hole was cut in the lower fuselage to gain access to the ESC wires. One works through this hole to attach the ESC wires to the motor. Not really easy, but it can be done. Later, I made a 1/16-inch sheet balsa scoop (again covered in MonoKote) to cover this hole and provide cooling to the ESC. This scoop was only taped on to allow access to the ESC if later required. The floor for the battery has hook and loop fasteners on it with mates on the battery to keep the battery in place during flight. Lastly a 2 x 2-inch hole was cut in the bottom of the aft fuselage to allow cooling air to enter the front (lower scoop) and travel through the fuselage.

A 3/16-inch 5-ply firewall was fabricated to accept the BP BL-4120-7 outrunner motor.

With the front portion of the fuselage under the wing now vacant, I installed the Spektrum AR7000 receiver and its satellite on the floor and wall as per Spektrums recommendations. The Futaba servos were plugged in and the model bench tested. All seemed well, and to my surprise, the CG of the model was exactly where it was before the deconstruction began! I weighed the model and as I expected the ready-to-fly aircraft weighed a mere five ounces more than an empty tank glow powered Avistar. With a full tank, a glow Avistar would actually weigh a bit more! All that is left to do is charge the battery and go fly.

The inside wall of the model was roughedup where the new firewall will be glued.


 Castle Ice controllers have data logging capability. Since these became available, all my new models have one installed. Knowledge is power! From the data collected, I can see exactly how much power I am consuming, watch for over-temp conditions and record the motors rpm. I use a stand-alone airspeed logger from Eagle Tree Systems to record this measurement separately. The airspeed measurement device, without using the full Eagle Tree data logger, only displays the highest airspeed reached. That is good enough for my testing.

Motor mount was trial-fit before gluing it with epoxy. Note the one-inch balsa spacers to properly position the new firewall for the chosen motor.

Static (bench) data was taken at home, then again at the field before the first flight. During the first familiarization flight data was recorded, but I was not interested in it. The model was accelerated to takeoff speed (less than full throttle) and quickly became airborne. I flew the model up to altitude to trim it out. After being pleased that the model flew well, I landed the model, dumped the data onto my laptop and cleared the log file on the ESC. I recharged the battery pack and flew the model again looking for two data points: max power and rpm at maximum throttle and then again at a “comfortable” cruise.

If your particular model will not accept a new firewall between the sides, build a plywood box extender and glue that to the old firewall.

For the full-throttle run, I carefully took the model off at less than full power and climbed only to about 100 feet. I proceeded well away from the field to set up for a long high-speed pass at as constant an altitude as possible. I made two passes, one right after the other and landed. I was probably in the air less than one minute. I offloaded the new data and cleared the log file again. I recorded the maximum airspeed off the Eagle Tree unit. Now, without changing or charging the battery, I took off again, at just barely enough power to climb. I wanted to make sure that for this next test, cruise power/speed, the maximum airspeed reached would be at cruise and not at takeoff. I flew the model up to about 200-300 feet and began to slowly increase the speed until I reached one where I would be comfortable training a new pilot. Once I visually found that airspeed, I held it for about 20-30 seconds to be able to get a good average in the data logger. I landed the model and again offloaded the data and recorded the highest speed reached (this time cruise, not top speed). In the chart below is the data collected and averaged.

Motor/prop installed with about 1/4-inch gap between the prop and the “cheeks.”

If you remember from last month, I was actually looking for around 35 amps (static). The difference between the advertised Kv of the BP motor and what was actually manufactured was enough to drop the current 10 percent from the calculated. However, I was pleasantly surprised to see that the model flies extremely well at only 460 watts at take off, 343 watts at full throttle with a 60 mph top speed and a mere 170 watts during cruise speed of 45 mph. With a 3200mAh pack on board, even if I only fly it to 1/2 capacity, I can expect almost nine minutes of cruise flight, more than enough to saturate a new pilot!

Below left: I installed the Castle Phoenix Ice Lite 50 ESC below the battery pack in the forward, lower fuselage. A hole had to be cut in the lower skin to provide air for cooling and to mate the ESC wires to the motor. It is secured to the bottom of the battery deck with velcro. Right: A sheet balsa scoop covers the hole and is just taped in place to retain ESC access as required.

The model is not just a good basic trainer with electric power, but is also rather aerobatic at slightly less than 100 watts/lb. The model has a good speed range, excellent climb at full throttle, makes big round loops and rolls like a warbird! It can stall turn and fly inverted. This project has been very rewarding, and I now have a model that I feel I can comfortably train a new pilot on, as well as let them graduate to some basic aerobatics when ready. I hope you too can find some model out there you can successfully convert to e-power and enjoy it as much as I do this one!


A 2 x 2-inch hole was cut in the lower aft fuselage to allow cooling air to exit.

Until next time, fly clean, fly quiet, charge safely!










BP Hobbies, (732) 287-3933

Castle Creations, (913) 390-6939

Eagle Tree Systems, (425) 614-0450

E-flite, distributed exclusively by Horizon Hobby Distributors, (800) 338-4639

Futaba, distributed exclusively by Great Planes Model Distributors, (800) 682-8948

Spektrum, distributed by Horizon Hobby, (800) 338-4639