BeechTalk - BT - Raptor Aircraft 5 Seat Pressurized 3,600 NM Range Diesel

Username Protected · **Posted:** 26 Aug 2020, 21:33

Username Protected wrote:

He's said it a few times in his comments, he just prefers to fly with his right hand for the yoke/stick. :shrug:

And yet he built the plane to be left-handed. :scratch:

Username Protected · **Posted:** 26 Aug 2020, 21:40

I just watched the latest video. He's at 90kts on the ground! If I got my plane up to 90kts and was still on the ground, the smallest bounce on the nosewheel would have me airborne. And given how much that thing is bouncing around, I can't see how he's not off the ground. Unless it's so heavy or his angle of incidence of the main wing is wrong.

And he doesn't seem to understand why his elevator is deflecting upward as his airspeed increases??? :crazy:

Velocity's have a sparrow strainer (small, upside down wing stuck on the trailing edge of the elevator) for just that reason. And a pretty hefty elevator trim spring.

But he does love to look at his graphs, doesn't he?

Username Protected · **Posted:** 26 Aug 2020, 21:55

Wow, why doesn’t he get that airspeed fixed? He loves to talk about precise values like it should start flying at 93 knots. Yet, he’s just fine with 93 knots being maybe 104 indicated.

And instead of figuring out why the force from 95 knots is pushing the elevator up. He will just add more trim.

At this point, he just needs to fly it himself. No one needs to risk their lives for this thing. All the gizmos are great..... well, until their stop working.

Username Protected · **Posted:** 26 Aug 2020, 21:58

I just saw today’s video. Wow.

Elevator spring tension is not sufficient to overcome air loads on the takeoff roll so the elevator is streamlining. His solution? Trim it farther up.

The throttle has two potentiometers. The “backup” is compared to the “primary” and if they disagree by more than 10% the ECU throws a fault and pulls the engine to idle. No third pot for a rationality check, you just lose the engine.

Interesting times.

He’s definitely planning to fly it himself.

Username Protected · **Posted:** 26 Aug 2020, 22:07

This is why he needed a couple of guys that did nothing for the several years except work on the engine and electronics. This thing may fly. But, it’s almost guaranteed that some sensor is going to send it to idle and make it a glider.

Username Protected · **Posted:** 26 Aug 2020, 22:08

Username Protected wrote:

The throttle has two potentiometers. The “backup” is compared to the “primary” and if they disagree by more than 10% the ECU throws a fault and pulls the engine to idle. No third pot for a rationality check, you just lose the engine.

He glossed over that didn't he?

He described it like it was a redundancy, but in reality it's a single point of failure. I hope he bought the Harbor Freight extended warranty.

Username Protected · **Posted:** 26 Aug 2020, 22:34

Username Protected wrote:

He described it like it was a redundancy, but in reality it's a single point of failure.

Good way to put it. A mismatch of 10% becomes a worst-case failure. With only one pot, the fault he saw would have perhaps limited the range of available power. Instead, he got zero power.

Username Protected · **Posted:** 26 Aug 2020, 23:50

Username Protected wrote:

He described it like it was a redundancy, but in reality it's a single point of failure.

Good way to put it. A mismatch of 10% becomes a worst-case failure. With only one pot, the fault he saw would have perhaps limited the range of available power. Instead, he got zero power.

Username Protected · **Posted:** 26 Aug 2020, 23:59

Username Protected wrote:

That's pretty much the way electronic throttles work on ground vehicles but it actually makes sense there. Mine will go into "Limp Mode" if the inputs disagree and that allows you to move the car but only very slowly.

For an aircraft, uncommanded high power is a bit less dangerous than an uncommanded shutdown but neither should be acceptable. If this were my design there would be three throttle control sensors with majority voting or a manual override switch to allow the pilot to select between a ground mode where both sensors need to agree to increase power above the lowest output and an airborne mode where both sensors need to agree to decrease power below the highest one's output.

My guess is that designing such a control system is beyond his ability (like a lot of other things).

In any system like this where both data accuracy and verifiability are important even numbers of sources are bad juju. I agree three would be much better. Even then there should be a better failure scenario than idle power. I'd go to a known good low power setting. Enough to just barely fly level with the gear out but not enough to do so with the flaps down. You then descend by extending the flaps and kill the engine with the start/stop control to land.

Username Protected · **Posted:** 27 Aug 2020, 00:08

Username Protected wrote:

In any system like this where both data accuracy and verifiability are important even numbers of sources are bad juju. I agree three would be much better. Even then there should be a better failure scenario than idle power. I'd go to a known good low power setting. Enough to just barely fly level with the gear out but not enough to do so with the flaps down. You then descend by extending the flaps and kill the engine with the start/stop control to land.

My understanding from half a sentence in the video was that he was going to look into whether there was another option in the configuration besides going to idle power during that fault.

Username Protected · **Posted:** 27 Aug 2020, 00:22

Username Protected wrote:

In any system like this where both data accuracy and verifiability are important even numbers of sources are bad juju. I agree three would be much better. Even then there should be a better failure scenario than idle power. I'd go to a known good low power setting. Enough to just barely fly level with the gear out but not enough to do so with the flaps down. You then descend by extending the flaps and kill the engine with the start/stop control to land.

You also need to consider the adverse consequences of letting the engine run at a power level sufficient for continued level flight when the airplane is on the ground.

Username Protected · **Posted:** 27 Aug 2020, 00:56

Username Protected wrote:

In any system like this where both data accuracy and verifiability are important even numbers of sources are bad juju. I agree three would be much better. Even then there should be a better failure scenario than idle power. I'd go to a known good low power setting. Enough to just barely fly level with the gear out but not enough to do so with the flaps down. You then descend by extending the flaps and kill the engine with the start/stop control to land.

You also need to consider the adverse consequences of letting the engine run at a power level sufficient for continued level flight when the airplane is on the ground.

Username Protected · **Posted:** 27 Aug 2020, 02:50

companies with slightly more engineering resources, like airbus and boeing, still manage to have spectacular failures from poorly conceived failure modes of sensor inputs.

If i was planning the initial flight of a brand new piston single airframe design, I'd power it with a carbeurated O-540.

Username Protected · **Posted:** 27 Aug 2020, 08:16

> He described it like it was a redundancy, but in reality it's a single point of failure

It is worse. His goofball design has doubled the chance of a sensor-induced loss of power.

I am very curious about his potentiometer. Is it a simple resistance pot? If so, damn poor choice.

Username Protected · **Posted:** 27 Aug 2020, 08:24

Username Protected wrote:

I'd assume there's a start stop control that you could use to prevent that.

He can ground the mags or pull the mixture lever once on the ground.