J7W Shinden is available

Feb. 01, 2007

I've started J7W Shinden, an aircraft that has a canard configuration with a pusher propeller. At the time I was looking for the next airplane to make, my son said "Dad, I like this one. I want fly this." when he was seeing (not reading yet) a book of warbirds. "Oh dear, this airplane is very different from others, and probebly few data are available so it's a bit hard to make. What else do you like next to this?", I said. "I like it. This one I like. I LIKE Shinden!!" he claimed, pointing out the picture of Shinden. Okay, here's the story of making Shinden.

From the next day, I'd collected three-view drawings and basic characteristics of Shinden. There are some three-view drawings. great so I made a 3d model. There are some websites that has engine power and some basic characteristics though some data vary per article. hmm it's OK since there's only a couple of prototypes were made and only one flew. What about aerodynamical characteristics? coefficients? No data available.

So this is the beginning of guesstimating parameters. First I'd tried JSBSim but I didn't know how to make a flight model for a canard-type airplane. Plus, the configuration files for engine and propeller didn't seem working as I'd expected. I'd read both documents and source code to find out how to make an acceptable thruest/power table, but I couldn't find a good parameters. It was getting better, but better is not good. it's not even closer to the specs or the data described in the flight record of Shinden in a website. Moreover, The engine torque made Shinden roll into the opposite direction but I didn't know how to invert the direction of the moment. I've tried to change some parameters but I couldn't make it. Okay, use yasim for now.

It is relatively easier to make a flight model with yasim since it doesn't require pretty much coefficients data. Plus, I'm a bit familiar with it since I've already made Ki-84 with yasim. So I wrote an xml file for J7W, guessing parameters referring to Shinden's flight records. I also placed a huge ballast weight to adjust the CG of this airplane.

first flight

Okay, so this is the first flight. See how does it fly. My son was a bit excited as this was the first time he saw Shinden in FlightGear. so did I, of course. Shinden slowly starts to taxi, engine revs up about 2500 rpm and still gradually revs up, looks good. At about 110 kt, it slowly takes off as I expect.

"Wow, I made it!" I thought, but the very next moment, Shinden got spinned like a gyro and crashed onto the ground. "Oh... what happened? I just gently pulled back the stick a little bit to take off." My son was saying "Daddy, Shinden crashed?" "Yeah, it looks like that.."

The reason I crashed it was simple, too much sinsitivity in elevator controls. I'd read some articles about canard configuration so I limited the angle of elevators like 0.8 but it was not enough. When I pulled the stick up, wing suddenly stalled and Shinden lost its control. so I limited the angle of elevators to 0.6 and tried again.

Second flight

Shinden took off a bit later but it was OK. My big question was 'why this aircraft takes off from the tail instead of the nose? I know the flaps bring the nose down but isn't that too much?'

The answer was in the flight record of Shinden's test flight at a website. It has flaps at the bottom side of the horizontal stabilizer too!! Yeah, I saw some weird lines at the bottom of the stabilizer but I didn't even think it could be a flap. So I changed the wing parameters to make flaps under the horizontal stabilier. However, it didn't help putting the nose down at the take off, so I changed the incidence and idrag parameters to stabilize the nose.

Third flight

It took off with the nose up. Changing the parameter seemed working. good. It rolled clockwise very much but it was supposed to. I make a very big sweeping turn very carefully manipulating elevators and ailerons, but I got spinned again.

The reason for this was a bit complicated. The nose wing may stall too soon. The rear can be too heavy. The parameters of the wings could be wrong, or all of these.

Nth flight

I don't know how many times I modified the configuration for stabilizing this aircraft. but it seems working way much better than the first model. Now I'm going to publish this aircraft.