- This topic has 29 replies, 6 voices, and was last updated 13 years, 10 months ago by
.
-
Posts
-
A well known problem of electric motors is that there is no control where the prop will stop. As a result, the prop is moved forward to the point the blades will stay clear of the wings. There are two disadvantages. First, it requires placing the motor forward – thus increasing the longitudinal inertia moment. This can be somewhat mitigated by extending the prop shaft via a nose cone. I happen to do exactly this, but it requires machining and reversing an outrunner’s shaft.
Second, from an aerodynamic point of view, the prop’s disc acts as a forward rudder. Placing the prop further away from the wing actually destabilizes the model, making it more difficult to trim. Obviously, this primarily applies to models with larger props. Although larger props are more efficient, it is well known that such models are much more difficult to trim.
In a conversation with Dave Lacey a few days ago he told me that he found a way to fold the props under the wing! “Wow, how is that done?” was my incredulous response.
Dave needed to move the CG back on his E-36, so the pylon was moved 3/4″ forwards, placing the prop 1.25″ ahead of the wing. He then made a brace from music wire, 1/4″ behind the prop, in the form of a semi hemisphere. (Geometrically, the blades can only hit the wings if they are in the upper hemisphere. With any pylon, the impact sector is actually less than 180 degrees). In fact, if a blade happens to rest upon the brace’s upper arc, it would rotate forwards due to air pressure (the other blade acts like a weathervane – just reducing it’s drag). The brace prevents the blades from hitting the wing and assures that the blades fold beside the fuselage, under the wing, acting as a stop.
Without a stop, the prop’s hub can end in a vertical position, with the lower blade dangling down. Upon landing, the prop’s hub could absorb some of the impact, potentially damaging the motor or bending the shaft.
How to start the motor with a brace behind the props? Pim Ruyter (with a ramped up pylon) first points his electric F1Qs down, letting the blades open. He then starts the motor at a low RPM, enough to spread out the blades, for 3 seconds. He then rotates the model to a launch angle, and launches it under full power. On my models, using a Roger Morrell e-timer, pressing the start button starts the motor at a low RPM. Only when the button is released (when the model is let go) does the motor switch to full power. This way rotating the model upwards has no time constraint.
Having a stop and being able to fold the blades under the wing is a significant development! It favors models with larger props.
This is a wonderful idea by Dave Lacey. I can’t wait to see it in action at the Brooklyn Skycrapers Spring meet …. curious about the wire size and attachment method to the fuselage.
Re: Starting the motor with the nose down and the blades folded foward. Please be very careful with this. There are some blades that will fold far enough so that one tip will go past the motor’s rotation axis. I have seen this happen using Robbe CFK blades with disastrous results. If the tip is past the motor centerline the blade will not open and the resulting unbalance will wreak havoc.
Re:Launch power. This is just my personal opinion, but I don’t think it’s a good idea to launch with anything less than full power. There are too many things that can go wrong, e.g. the motor could cut out when it’s accelerated up to speed because battery power is lower than expected and LVC kicks in. Also, it takes time for the motor to accelerate to full speed. Why lose time on a 5 second motor run?
The question is whether to launch a model under full power (as Dick is suggesting), or to let the motor rev up to full power after the launch as Chuck Groth does. Obviously the former prevents all kinds of mishaps, including a delayed full-power-start that would wreck the model.
Under full (static) power the model is subject to maximum torque as the blades are stalled. As the model accelerates, the blade’s angle of attack drops – increasing their thrust (lift) and decreasing their drag (torque). So, at some point, say one second into the flight, the blades are no longer stalled. The model continues to accelerate until reaching its terminal climb speed (maybe a second later). This shows why tossing a power model is important.
In the second scenario, as soon as the model is air born, the blades begin revving up to full power, say over one second. Initially, at low advance speeds, the RPM is low, so most of the blade is not stalled. As the speed builds up, the RPM increase – matching the model’s speed to the prop’s RPM, which is operating at a higher efficiency. There is also less torque and slip stream rotation against the pylon, fin and stab, mitigating some of the offsetting measures. In addition, the energy of the battery is not wasted turning a stalled prop at full RPM at launch. (In a somewhat related example, if one starts a F1B motor while holding the model, one feels a strong twist – amounting to wasted burst energy). Since some e-timers can control the ramp up rate, a controlled RPM ramp up is one of the differences between electric and conventional power models.
Altitude measurements or mathematical modeling could settle this question. But it definitely would be model specific (weight/motor/battery/prop combinations). In the mean time, launching models under full power is much safer.
Menu for electric model front ends:
Big Chins: mount the motor at the tip of the nose. Works for smallish props.
Long noses: mount the motor just ahead of the wing, and deploy a nose cone to move the prop forward. Works for large props.
Braces: Place the motor further back, or use a shorter nose cone. Add a wire guard (or brace) to prevent the blades from hitting the wing, which also lets them to fold under the wing. Works for any prop size.
These are some thoughts about designing a brace. Imagine a plane parallel to the props plane, between the prop and the wing, which will be called the brace plane. Suppose, just as an example, that the brace plane is placed mid point between the prop and the wing.
Link to larger version of drawing – http://www.freeflight.org/images/prop_brace_2_932.jpgThe critical blade angle is when the blade is moving up, grazing the left wing at its very tip. (The same applies to the symmetrical case, when the blade is moving down on the model’s right side). In the brace plane, measure the prop’s pitch angle when the blade grazes the wing as well as its radius from the center of the motor shaft – referred to as the grazing pitch angle (up or down) and the grazing radius. Note that the grazing pitch is lower for models with large props and/or low pylons.
The angle of the grazing blade, from its flight position (when fully spread out) will be called the graze fold angle. (This is the plane through the prop, parallel to the incoming air flow). Note that a blade’s fold angle when it rests against the wing’s center is larger than the grazing fold angle. Consequently, the radius of the blade in the brace plane when it rests against the wing center is less than the grazing radius. In fact, this is holds for all angles between the up and down grazing pitch angles.
Geometrically, the radius of a brace’s arc from the shaft’s center, spanning from the up to the down graze pich angles, should slightly exceed the grazing radius, assuring that the blades always miss the wing. And when one of the blades rests against the brace’s arc, the fixed brace radius will assure low friction as it slides forwards due to the incoming air pressure.
But we might be less lucky. The up moving blade could hit into the blade’s radial portion on the model’s left side, if it had incrementally folded too much over the last three-quarter turn. To prevent this one should create a rounded entry curve designed to slide the blade upwards, over the brace’s arc. In addition, the radial portion could be padded.
The force of air pressure pushing the blade forward decreases with the blade’s fold angle. (Think about fixed propellors of P-30s freely spinning during glide). With a fixed distance between the prop and the wing, the size of the brace will increase as it is placed further back, with a corresponding increase of the blade’s fold angle. But if the brace is placed too close to the prop, the blades would hit it too soon (while turning too fast) as the propellor is slowing down, causing mutual damage.
In other words, a brace in a plane parallel to the prop’s plane, can only work if the distance between the prop and wing, given a prop’s size and the pylon’s height, is sufficient.[url][/url]
Dan,
This is exactly what we all do now. It also means that the prop’s distance from the wing is at least (prop’s diameter – the hub length)/2, ignoring the hub’s thickness. For example, on my E-box (made by Oldenkamp) the prop is 4” ahead of the wing, compared to the calculated 3.87”=(9” – 32mm/25.4)/2. With an 8″ prop, the distance is 3.37″ and with a 7” prop the distance is 2.87″.Using a prop brace allows you to place the motor further back, reducing the model’s longitudinal moment and allowing lighter strcutres. Unlike gas models, the battery can be used to achieve the desired C.G. position. For example, on some of the Ivers models, because the front mounted motor was so far ahead, the battery (~85 gr.) had to be placed projecting a bit beyond the wing’s trailing edge. So, if the blades could overlap the wing, less material would be used on the nose, and the battery could be moved forwards, reducing the model’s overall weight and its longitudinal moment. Another effect is a more stable climb, as noted above.
But using a brace for small props might not be worth the effort.
The following is Pin Ruyter’s response (including a nice picture of his model).
“Aram,
How to make easy solutions difficult.What I do; use a Brushless Controller which can be programmed with a computer e.g. Castle link or Hacker Pro. Use your computer to program the controller with a soft brake ( 3 options, no brake, soft brake or hard brake). When the motor stops the brake forces the prop. to stop and folds. Due to the shape of my front fuselage the prop. always fold under the wing against the fuselage (see attached picture of my model).
Pim,”My response:
Agree that it is another workable solution. As the propeller is slowing down, the blades will start grazing and bounce off (and up) from the top of the forward curved pylon, thus slowing the prop. Eventually a blade will bounce back, thus stopping the prop completely. Since the blade is elastic, the prop actually turns backwards (counter clockwise), a partial turn. The left blade will end up resting alongside the fuselage, not against the top of the wing. Judging from the picture, the blade folding angle when it grazes the pylon might be within 50-60 degrees, indication that the propellor is turning slowly at that point.The effect of your forward curved pylon can be replicated with a wire brace; with the same cross section at about the point the blade grazes the pylon. Although such a brace is less bullet proof than the example above, it evidently works in practice.
Seems to me that programing your controller for a soft brake robs you of critical run time. It is interesting to see the many different ways people approach Electric Freeflight.
Hmmm!
It does look a little like a submarine doesn’t it?
And you are widely known for your aversion to water.
Or is it the minimal use of balsa?
Cheers!Props that can hit the wing, a soft brake stop, climbing at 37 degrees with a multi-position stab. It just doesn’t seem the way to go.
Pim Ruyter uses a soft break, which allows the blades to gradually lean backwards, graze the humped pylon and then fold under the wing. A hard stop is generally avoided with large props, not to overstain the motor or the fire wall.
Should the ramping down period be considered as part of the motor run? As soon as a model starts decelerating, its advance rate is falling, its RPM is falling and the blades start pivoting backwards into a partial fold. The blades’ angle of attack quickly falls, becoming negative on their outer portions, accounting for the side slip due to their partially folded angle. The centrifugal force continues to spread the blades out as the propeller running above a minimal RPM. Although ramping down can be observed, the propellor is really acting as a weather vane, definitely not generating any thrust.
A brace will work as long as the motor does not lock immediately. After the motor is stopped, if the prop happens to turn the motor, the electricity (EMF) that is generated (the motor is now a generator) is sensed instantaneously by the controller, which in turn uses it to stop the motor. As an aside, turning a disconnected motor requires a fair amount of torque due to the interactions of magnets with the electromagnet’s wire coils.
Based on a conversation with technical support at Castel Creations, the Phenix controllers have three stopping parameters. (1) stop delay after unpowering the motor (2) breaking percentage and (3) break ramping. For large props they can be set at (1) zero delay, (2) 100% break which means that all the electric power generated by the motor is used by the controller to stop it and (3) medium break ramping which takes about .5 seconds (from a menu of (a) zero (ramping), (b) fast, (c) medium, (d) slow and (e) very slow). If the break ramping is too abrupt, the breaking torque would damage the motor or the fire wall. (Their simpler controller, called Thunderbird has (1) no delay, (2) 50% break and (3) no break ramping).
So, if breaking is not instanteous, a prop brace will allow the blades to fold under the wing. This also applies to rubber models, and to power models when the break does not apply instantaneously.
“Should the ramping down period be considered as part of the motor run? “
The rules at this point are a bit vague. If I am timing your model my watch will run as long as there is energy flowing to the motor. The motor run will be more clearly defined for the AMA classes in the next rules cycle. As for the big prop approach, we can let the flight line determine it’s merit.
- You must be logged in to reply to this topic.