Problems Enlarging Aircraft

I have been dabbling lately with creating vary large models in G4.5, using downloaded aircraft and default aircraft. Most times, issues are minor and easily fixed. However, I have run into a couple of aircraft that I'm having some issues with that I cannot figure out.

1. Top Flite P-51
I have successfully got the aircraft enlarged and flying OK when I can get it airborne. The problem lies on the ground as soon as the aircraft is loaded or reset. When on the ground, I get a "break dancing" airplane that usually ends up breaking the prop before I can ever add throttle to it. Adding weight has helped some, but makes the plane too heavy. The problem goes away when I change "Springy Scale Retracts" to "Scale Retracts", but there is no shock absorbtion on landings, making for a very rough landing on anything but a paved surface.

Is there something that I could do to get rid of the "break dancing" without having to go with the stiff "Scale Retracts"?

2. Coast Gaurd Jayhawk
I have gotten the aircraft enlarged and in the air, with plenty of power to spare. However, it is more unstable and out of control than a helicopter should be. With the CG placed towards the rear, it'll take off with level pitch and roll hard left. If I can manage to counter the left roll, the tail drops violently after getting about 5-10 feet off of the ground. The balance at this point is very unpredictable, creating violent changes in direction that often cannot be countered. If I move the CG forward and right, it takes off nose low, but once off the ground will somewhat level out. Once in a hover, the balance is again very unpredictable with the same violent changes in direction that cannot be countered. I've tried making the head a bit softer to absorb some of the movement, but that generally causes excessive blade flex and reducing my control response when trying to counter it's movements. Making the head stiffer does not help any.

What can I do to make the larger size Jayhawk fly more controlably?

If you need the AV files that I'm working on to get a feel for what I'm talking about, just let me know and I'll try to post them here.

Thanks in advance for any assistance you may be able to provide.

1. This issue has to do with a flaw in the 3D model. Other than what you did by disabling the springy, there's nothing an AV can do to fix the problem.

Hey Mydartswinger


That's ok we don't need to see the AV. We have all seen this before. The swaps are full of Large sized AV that don't behave well. Remember that this is a RC Aircraft sim and not a flight sim for full sized planes. The option to scale Aircraft is there for people that might want to take a .40cc Size RC and scale it up a bit to get a .60cc size. All calculations are based on know RC engine, weights, and flight behaviors. It was never intended to make every plane out there a 99% scale version of the real plane. Your not going to get every plane scaled up that 400% You win some you lose some.

Like Junkboy said...

Realflight can scale things up to 200% and produce fairly realistic flight - IF - you adjust the weights, power, components, etc. etc. etc. in the physics editor.

There is a nice LONG thread called something like "Physics for the Real World" here that details all of this.

Scaling a model up is somewhat involved requiring far more than merely changing the scaling values.

Beyond 200% things start to degrade and by 400% even with the required adjustments things go really wrong.

It is an RC sim after all.

You can try adding weight to the model and increasing or decreasing the 'stiffness' setting of the springy retracts. You can also try a different tire type under the 'wheel' setting in the editor. Also, you can play with the friction setting for the wheels. By adjusting these parameters, I have had some success controlling the dancing on larger AVs with springy retracts.

Griff
Twoeleven

soften the struts

drop the strut back to around 15% or so. It has helped on a couple that I have done.

Deewng

Thanks for the replies. Sorry it's taken so long for me to respond.

Understood, it is an RC simulator not particularly designed for full scale recreations. My intent was not to make it fly exactly like the full scale aircraft, just a very large, nearly full size RC that flies at decent speeds, not too floaty (for the faster aircraft, floaty's fine on aircraft meant to float, but neither one of the 2 I'm having these issues with).

twoeleven,
I've tried adding weight (lots of weight, including to the LG themselves) to the model and adjusting the stiffnes, both up and down to no avail.

opjose,
First off, thank you for letting me know that my files need improvement. I actually thought they flew OK (maybe not great, but OK). Now I can look forward and improve on what I have. I've only done a couple of AV's before, and they were scaling down to get a plane to fly like one that I had at the time. I'm not very experienced at it and still have a lot to learn. I guess that's part of what the swap pages are about.

I found that thread about the realistic physics, I will read it, and make improvements to the models based on instructions provided in that thread. I have, however, added weight, changed power sources to move the added weight (including going into the torque generator editor and adding power that way as well as using the "torque from engine modification" setting), changed prop diameters/pitch, adjusted CG's, and increased the strength to allow for a more positive flight experience than scaling alone. I did use the scaling percentages to get the desired graphics and physics size without having to manually re-do every dimension on the aircraft.


Thanks agian to all for the replies and advice.

Do NOT rely on "Torque from engine modification" to get the desired power output.

Instead, grab all of the specs for the engine from the manufacturer that you can.

e.g. They will say things like 2.2HP @ 7400 RPM using a 18x6 APC prop.

Since we KNOW that manufacturers go for optimal numbers, start plugging values into the engine curves until the editor shows your new "engine" producing about that ( but not more than that ) AT FULL THROTTLE.

This will be the high point of your engine response curve.

Note that the Nav Guide will tend to show a lower HP or watt output value than the physics editor... this is exactly what we want.....

Most specs also provide an RPM range.

Adjust the curve so that the engine hits ZERO torque beyond the high RPM range given, and then smoothly taper from the high point of the curve to this point ( usually quickly ).

Double check the HP rating in the editor as you do so.

Once done you can play with the left side of the curve, and you've ended up with a fairly representative "engine" for your plane.


Hint:

To avoid having your work overwritten by a subsequent model adjust your naming of all inter-related files including Engines, Colorschemes, etc.

Let's say that "John" creates a P-51 using a 50CC DLE engine.

Someone else may also use a 50CC DLE engine and if their model uses the same nomenclature, theirs will overwrite yours.

Later someone runs your plane or variant and discovers that it does not behave "like the real thing" as a result...

So how to solve?

John names his files like this...

Plane: P51_John
Engine: P51_John_DLE50CC
Colorscheme: P51_John_Default

etc.
---

Also remember that as you scale up, different airfoils are often used.

Typically a .40 size model has a thicker airfoil than say a 100cc version of the same plane. Of course there are purposeful exceptions to this, but don't overlook this.

It plays a large role in determining the overall response of the plane.

---

If you have multiple battery packs in your real plane, add "mass" components to reflect this, approximating where you have your real packs.

---

I find that G3/G4/G5's C.G. values to be off.

That is, if I set the C.G. on the virtual model at exactly the same distance as the real world model, I end with a more nose heavy plane.

Typically to get identical performance I find that I must take the C.G. BACK in Realflight by a significant amount. This amount depends upon the size of the modeled plane.

---

The "trimming" guides used for pattern planes are of fantastic help when adjusting the physics of a Realflight model, particularly if you use the same techniques in a real world model to compare against.

These guides step you through most of the nuances required to properly adjust a plane for FAI and pattern competitions.

If you apply the same methodology to Realflight's physics modeling, you can get surprisingly accurate results.

If you do so you'll also see the C.G. issue that I take about above...

I find that if you use the trimming guides properly you can often end up with a Realflight model that exhibits EXACTLY the same quirks and idiosyncracies as it's real world counterpart.

When you get to this, you've effectively nailed the physics perfectly.