Thread Tools
This thread is privately moderated by BAF23, who may elect to delete unwanted replies.
Jul 27, 2015, 05:48 PM
The sky is the limit
BAF23's Avatar
Thread OP
Build Log

Secondhand unflown Aermacchi MB339 PAN

Having had a ride in the back seat of the leader of the Frecce Tricolori during the 1984 Sanicole airshow, the FlyFly/TopGun 90mm EDF Aeromacchi 339 model had been on my wish list for some time. I saw somebody else in the club fly this relatively large but seemingly docile model , and when I decided to get rid of my smaller EDF’s and standardize on 90mm fan models, I immediately thought of this as a proficiency trainer. End of June 2015 I saw a completed yet unflown one for sale on the secondhand market and purchased it for 200 euro.

The assembly and finish quality was atrocious, but the value was the Typhoon HET EDF 650-58-1760 Motor with an 11 blade TAFT fan, the rest was for the model with servos and replacement electric retracts and battery. I figured all those items later could be used on my still dormant FlyFly F100 Super Sabre as well.

The first detailed inspection at home did not make me happy. Servos seemed to be the cheapest HK variety, summarily glued into the recesses and covered by painted-over sticky tape. The reliability didn’t bother me too much because there were separate servos for each elevator and aileron, thereby allowing for half the control in pitch or roll even after a servo failure. The drag produced by the thick Christmas decoration letters with rough sparkles would have been enormous, and these were quickly peeled off.

Some of the paint also came off and because it all had been applied with a brush, the model looked even more battered. I then removed the wing and discovered to my horror that the Turnigy DLUX 100A ESC with 5,5V 3A SBec was only kept suspended from the top of the intake by a twisted single piece of wire normally used to keep power cords together during transport. The Velcro was only at the back of the ESC and not on the fuselage, suggesting the ESC had been used in another aircraft before. The 11 blade fan had no nosecone installed, and the blades showed serious FOD ingestion damage rendering them useless. These also seemed to have seriously rubbed against the Haoye casing wall. During fan removal I was astounded that isolation plastic had been used between the core and a plastic insert to make the 8mm fit over the 5mm motor shaft. No wonder it felt wobbly and had rubbed the sides. The motor seemed brand-new.

The seller sent me some missing items including a bunch of various nosecones. Another proof it hadn’t flown was that the Zippy 6S4500 35c battery was connected to the 100Amp ESC by green Multiplex plugs (which had started to melt either during soldering or ground runs). I replaced these by EC5 connectors.

A few days later I took more time to investigate and found out that the 11-blade ruined fans only were compatible with the 1760 high rev motor when using a 150amp ESC. The 100 euro brand-new motor was worth keeping, so I ordered a new Wemotec Midi Pro 6-blade fan EDF unit for 5mm axles to create a compatible 6S power source. My to do list quickly grew longer, this bird wasn’t ready to fly and needed essential work to get ready.

To do:
Replace schroud/fan combo
Ensure 3 engine wires fixed in trust tube
Fan to trust tube transition
Wing-fuselage plugs
Replace green Mpx by EC5 plugs
Fasten ESC to top of fuselage (temperature!)
7,6mm engine axle?? Excessive play with 8mm fans
Install nosecone on Fan
Remove rough decoration letters
Tighten gearlegs
Reinforce aft canopy pin receptacle
What to do with tips and pylon tanks?
Solidify battery tray floor
Purchase Motip green + red touch-up paint

Every item I inspected either had to be fastened, strengthened, altered or replaced. Gearlegs had been surrounded by white tape to cover the up to 4 holes that had been made to try to tighten thr screws in aluminum legs. Even the canopy assembly had to be corrected, he had used Velcro bits between the body and clear canopy, which left gaping gaps between both with the risk of them coming apart in the air. It took much more time to correct the model than to assemble a new one (not available on the market anymore), and both looks and assembly will remain well below even beginner’s standards. In a week’s time it became technically airworthy, whilst for safety I altered the electronics to have both elevator servos on a dedicated channel (thereby eliminating an unreliable HK servo reverser), but was unable to modify the glued-in aileron y-cable to obtain differential ailerons on separate channels. I would have preferred that because the model has no rudder, the rudder channel only controls the nose wheel steering. I got around the problem by adjusting the aileron servo arms to a 45 forward angle in neutral position. I had to use longer actuators, but with that trick I had been able to create about 50% aileron differential.

I didn’t trust the 3Amp ESC-BEC to handle 4 servo’s during the gear retraction, and augmented the power by connecting a 2S950mah battery through a 5v5Amp Ubec and Schottky diode as a boost and backup. 90mm EDF’s being notorious power guzzlers, I also used a FAS100 power consumed/amp sensor between the battery and the ESC. This was in turn connected via a vario/data converter to a FrSky X8R receiver. Because I don’t intend to fly that model for years, I designed both the dual power feed and the telemetry sensors in a modular way, easy to disconnect with standard EC5/JST/JT plugs to quickly reinstall in a different 90mm EDF model. With a model copy/paste in the transmitter, I thus will be able to quickly swap the whole nine yards (the cables are not really that long:-) very rapidly from the MB339 to any other 90mm EDF, keeping most of the programming/connectors intact and thus saving enormously on converting time.

Realizing it would never be a nice model of the aircraft I had a ride in, I decided to keep the thick wingtanks that the builder had installed on the tip of the wings, but removed the thin ones under the wing. I know that configuration is just the reverse of how the Frecce Tricolori fly, but I read about the model’s reduced tip stall tendency with tiptanks, and additional drag of the pylon tanks, and opted for a practical EDF trainer instead of a representative model. Just keeping a steady pull on the pylontanks made them slowly come out of their wing cutouts, starting me worrying about how the rest of the glue will stand the test of time and forces in the air. The wing/fuselage junction on the bottom allowed too much play when the nose was bent up or down, easily achieved in the air and on touchdown with the 700gr battery in the nose. I thus fashioned an ungainly plywood plaque to make the junction rigid by way of 4mm nylon screws in fuselage plugs just forward of the cheater intake. It still feels a bit spongy due to the soft foam, but at least the complete nose probably will note separate anymore in case of heavy maneuvering or landing. A plywood plaque also was glued to the back of the canopy so the 5mm! pin had something solid to grab on to latch, whilst minimizing the fore/aft play. The latch pin assemble was very wobbly and had to be glued properly.

After a rough finger balancing, I noted where the main battery had to come and installed a foam support forward of the cradle and Velcro fastener. The rest of the electronics were distributed lengthwise and wiring looms grouped by use of nylon straps. There was ample space to get rid of all the electronics, and after labeling and tidying the wires, I set the receivers’ failsafe positions (engine idle, slight left aileron and gear up), and performed a weight and balance. By placing the electronic accessories and receiver boost battery forward, I was able to move the engine battery much back and have it closer to the CG thus minimizing fuselage load and inertia coupling. Without batteries the weight is 2420gr, and with the 2 batteries weighing 740 gr, this means a flyaway total of 3150gr. Measuring the wing I calculated about 31,2 square decimeters which result in a wingload of roughly 100gr/dm2.

The large wing intakes prevented me from using my Sig balancer so I had to use my fingers just outboard of them to balance the model on a 90mm back mark by moving the main battery along the tray. Once I got the ideal battery position pinpointed, I glued front and rear retainers to keep it there, and used a strap to hold it against the frame. To get this advertised ready to fly model airworthy took me much longer than anticipated, and I still have very mixed feelings over the obtained results. Flight testing will make it clear if it will be a usable EDF trainer, or a lemon I better had not purchased. After a couple of years mainly flying lightweight foamies, this model’s wingload comes closer to my 2015 fleet average and will allow me to keep model flying proficiency in average wind conditions. I really don’t care much about that model but hope if anything goes wrong that I will be able to recuperate most of the non-foam hardware for other projects.

The maiden flight

At the end of July 2015 we finally had a few flying days where the temperatures were not too hot, the wind was not blowing too hard, and the raindrops were minimal. After a friend performed a buddy check on my model, helped me with the range check under various angles, and confirmed the Amp check being the predicted 56 Amp, we declared it fit for the maiden and lost no time in taxiing it to takeoff position. The nosewheel responded crisply to my inputs and the 50% expo allowed precise directional control at high and low speeds.

The six-blade fan is a bit of a screamer at full throttle, but acceleration was good and after rolling for about 50meters, I slowly started applying backpressure and the Macchi climbed out nicely and very stable in pitch. I continued the shallow climb to altitude and the aileron response also seemed very docile, the 25% expo for both elevator and ailerons seemed on the mark. After reducing the power at level-off, a few trim clicks were sufficient for hands off straight and level. General handling felt remarkably solid and my differential aileron setup allowed turns to be initiated without any adverse yaw, despite the lack of active rudder. The airplane showing no vices, I retracted the landing gear and further explored the flight envelope. This airframe is so draggy that power has little effect on maximum airspeed, it just adds noise. A barrel roll, loop and aileron roll passed very well so I decided to reduce the power for some stalls. My CG being just a tad forward of the recommended 90mm (that seemed close to the 33% MAC and worried me), seemed a bit too much forward because even with full back stick it didn’t want to stall but slowly mushed down with the nose almost level. There was no sign of any wingrock, probably thanks to the tiptanks. Having no rudder, I could not perform spins, but was prepared to use aileron into the spin to get out of it if it incidentally would get into one.

I then came back down to pattern altitude and lowered the gear, observing the correct deployment in the process. The first low approach demonstrated that the draggy airframe just prevents high speed flight, but in no way makes it a brick for landing. Even with the engine out it glided much better than anticipated. The relatively slow speed seems to favor a larger square pattern to lose the altitude, so my military continuous final turns brought me in too high and too hot, with me completely cutting the engine ahead of the threshold. Even at that moderate speed and forward CG, I still had sufficient elevator control to grease it in after a flare in ground effect. Rollout again took only about 40meters, and during taxi in I was happy to see that the main gear still tracked straight (remember the loose screws). That had been one of the easiest maiden flights I made, postflight inspection revealed no flaws except that the battery consumption figures on the telemetry didn’t match the figures of the battery tester.

The second flight was made with a different battery and confirmed the erroneous power consumption figures, at least that eliminated a battery problem. During that second flight I performed a couple of CG dive tests and those indicated that I can move the CG just a tad more back. This time I also flew a couple of Cuban eights and was surprised how well it kept its direction without the benefit of rudder inputs. This time I touched down about a meter short of the runway but again the gear held well despite the touch in the grass. This airplane is very docile and seems very well suited to resume flying after a period of inactivity. I wonder how stable it will be in gusty winds (without rudder) but anticipate this model to allow met to fly on moderate windy days. Despite its atrocious finish, it flies superbly and that is why I purchased it on the second-hand market, so far it has filled most of my expectations as a cheap no-nonsense practice ship.

Back home all controls and VWS were “nullified” to be correct at zero trims, a field solution regarding engine power cables was solidified, the position of the RX battery was changed to the opposite side and more to the back, and another FAS100 sensor installed to check what the problem is regarding power consumption telemetry. I attempted to reduce some drag by removing all the letters and taping up the wing cutouts for the 4 pylon tanks, but I don’t expect wonders. A lot of the drag comes from the model design itself, but I do not rule out that one day I might drastically clean this airframe up and then vinyl it, or find another yet to be assembled airframe to start from scratch and make a beauty of it.
Last edited by BAF23; Nov 07, 2015 at 01:03 PM.
Sign up now
to remove ads between posts
Oct 03, 2015, 05:35 PM
The sky is the limit
BAF23's Avatar
Thread OP

Getting along better with the ugly duck

The following couple of flights I further fine-tuned the Macchi and after a few aerobatic maneuvers perfected my approaches by flying at least 5 at the end of each flight because my Boeing 737 was in his test program and I wanted practice. Not only the Boeing had its share of development troubles, the Macchi unexpectedly also got me in the repair shop. When at altitude and performing a cuban-eight the engine died on me. I called an emergency landing and setup for finals. Thinking I had ample altitude, I dropped the gear and aimed for the threshold wit a 30 angle-in. All the guys around thought this would be a piece of cake, but the drag of the airplane again showed itself and and I was ending up about 10 meters short with no more airspeed to trade for altitude. I knew this was the well- groomed under-run but trusted the surface too much for this long nose-geared retract and it tore away part of the poorly designed delicate aluminum plate to which it was bolted, shearing all three servo cables with a clean gut where they exited the retract. The main gear had no damage neither the rest of the airframe.

Back home after removing the wing and removing the EDF unit and the ESC from the fuselage, I quickly discovered the culprit for the sudden engine failure. One of the bullet connectors between the ESC and the motor had been so poorly welded that it separated leaving a very rough uneven flat colored end the wire soldering dangling in between the other two. After cutting the isolation from the other connectors to check, the other wire welds proved solid enough so that was a quick repair. The nose-gear required more work and I decided to mount a new assembly that had a longer steering arm, so I had to fabricate a totally different 20cm long baseplate which I made out of plywood that was glued to the existing battery wooden plate that ran along a good portion of the nose. I also reinforced the attachment points of the main gear.

Following flights became no-brainers because after literally replacing almost everything, the model became trustful and behaved in a very predictable manner. It quickly became my favorite model flying-wise and it is a joy to play with. Early October 2015 a friend of mine had her camera and made following short video

Macchi (1 min 49 sec)

The Cg is now nailed but difficult to measure (due to the intakes) so I put the mark at the rear flat lip of the auxiliary bottom air intake. It flies like a charm but even with the lowcount fanblades it needs power and the 25c batteries gets a bit warm. Using the 35c battery offers better results. I made many more flights mixing mild aerobatics, low passes and touch and go's with a throttle percentage timer set at 5 minutes, but actually landing after about 8 minutes real flying time with 30% power remaining in the 4500 and 5000 mah 6S batteries.
Last edited by BAF23; Nov 07, 2015 at 01:19 PM.
Aug 05, 2018, 05:22 PM
The sky is the limit
BAF23's Avatar
Thread OP


After 3 years of faithful service I decided to sell my MB339 after removal of the power train (which I intend to use on a FlyFly Hunter) but nobody seemed interested and I also didn't want anybody to tell others that their newly acquired ugly duck came from me because this was not the standard I like to be associated with. My ex is fond of Italy and dreams about the Frecce Tricolori so I dismantled the model and swapped it in static condition for a bottle of Limoncello during the summer of 2018. It had served it's purpose (as a 6S EDF trainer) well, but due to the straight wing being wide during transport and hangarage (I never dismantled it before), I opted for a narrower swept wing foamy to replace it. Time to say goodbye to that cheap but ungainly finished foamy that flew very well and gave me much pleasure and training. Here is a video of it's last flight before dismantling
Laurence (4 min 46 sec)
Last edited by BAF23; Oct 09, 2018 at 06:18 PM.

Quick Reply
Thread Tools