I am just finishing my Chrysalis and am having balance problems.

I have read all the tips about keeping the tail light, and I tried to do so - very light coats of water-based poly, etc. My gross wt is right at 9 oz with Tetra Rx, HS-50's and a 150 mah pack - all as far forward as possible. My cg is still about .75 in behind the mid-point of the range on the plans. Here's my question: Since the Chrysalis is already almost 6 inches longer than a Monarch, could I shorten the tail by, say 3 inches? Is the tail volume less than a Monarch's or would I still have as much or more control moment? I'm a novice to HLG's and appreciate the extra stability designed into the Chrysalis, but could I compromise it some and still be ok? Thanks, John

From : Don Stackhouse

John, my first question is did you build it with a conventional or a V-tail? If it's the conventional tail, did you cut down the horizontal stabilizer as shown on the plans, or leave it full size?

If you used the oversize conventional tail, you can cut it down. No need to remove it from the fuselage. Make a cardboard template traced from the drawing on the plans, trace it onto the stabilizer, the cut off the leading edges and tips to the new shape. Sand the radius into the new leading edges, rub on some Varathane to seal the edges and you're done.

Of course the V-tail is the lightest option of all, but the small-stab conventional tail isn't too much heavier. Most of the weight difference between those two is in the joint where the tail attaches to the fuselage.

So much for the options shown on the plans, now let's get back to the heart of your question. First let's do a little reality check.

The Chrysalis was originally designed at a time when the smallest commonly available servo on the market was the Hitec HS-60. At that time, I intended for the Chrysalis to be a good quality SPORT hand-launch, maybe good enough for club contests and such, with emphasis on non-exotic materials (that's why there's no carbon fiber in it), easy construction that avoided features that beginners usually have problems with (which is precisely why there is no leading edge sheeting or carving required), with good performance, stability and handling for a reasonable price. It was NOT intended to have "world-beater" performance, nor did I expect to ever see one at the Nationals. As it turns out, its performance is very competitive with our old flagship hlg, the Monarch 'C' (which is the main reason why we immediately started working on the Monarch 'D'!), and I've seen a few Chrysali at each Nats since we first started making them. Still, you should not expect it to be more than it is. If that last few percent is THAT important to you, you really should be getting a 'D-lite' or a Wizard 'Lite'.

A really light building job in the tail section will help, and there are some "heroic measures" some folks have used for that. The finishing techniques you use back there can have a HUGE influence on tail weight. Even so, if you are using radio equipment that's far lighter than the lightest that was available when the model was designed, you can expect to have some C/G problems.

As far as shortening the tail, that can be a real "Pandora's box". Yes, I have heard of folks doing that. I haven't personally test flown their modified models, and I can't vouch for their abilities for quantitative handling evaluations. From what I know of stability and control, I'd say it's a very risky move.

Let's look at three major factors:

First there's static stability in pitch and yaw. This is linear with tail moment, and also linear with tail area. If you cut the tail moment in half, you need twice as much tail area to keep the static stability the same. One other issue: if you reduce tail moment without changing tail area, then you've reduced static stability, and in the process you also move the aft C/G limit further forward. This may require MORE noseweight.

Second there's dynamic stability, the ability to damp out oscillations. This is linear with tail area, but proportional to the SQUARE of the tail moment arm. That's right, if you cut the tail moment in half, you need FOUR TIMES the tail area to keep dynamic stability the same. In a piloted, close-in aircraft (like an HLG), this can also influence the aft limit of the C/G. If the airplane has very high damping in pitch and yaw (like the Chrysalis), it responds to disturbances very slowly. It may be possible (and in the case of the Chrysalis it definitely is, because I've done it) to set the C/G so far aft that the model is statically unstable by a small amount, and still fly the model successfully. As long as you pay close attention, you can correct a disturbance faster than the model can get itself into trouble. More mental workload for you, and I don't really recommend doing this, but with enough dynamic stability, a shortage of static stability isn't as much of a problem.

Note that the relationship between tail moment and the two basic types of stability do not follow the same mathematical relationships. This means that if you change something (like tail moment), it will be mathematically impossible to adjust something else (like tail area) to get BOTH parameters back to their original values.

Then there's the third factor, dutch roll/spiral stability balance. As you add tail area or moment arm (or reduce dihedral), the model becomes more resistant to dutch roll, but spirally becomes more unstable (it has a greater tendency during a turn to wind itself up into a "graveyard spiral"). More dihedral or less tail area or moment arm adds spiral stability, but also increases the tendency for dutch roll. The relationships here can be very non-linear.

A big part of development work is figuring out how to balance the design to get the desired behavior in each of these categories. Roll and pitch response are also intimately tied to this balance. It's one of the more dramatic examples of how a small change in one feature of an aircraft's design can have big "ripple effects" throughout the rest of the design.

So does this mean that shortening the tail will make the model unflyable? Certainly not. Will it behave the same as a stock Chrysalis? Also, most certainly not. In making this mod, you elevate yourself to the category of "aircraft designer and test pilot", and you should expect to have to either settle for whatever handling you get, or else go through what could be a long series of modifications and re-tests to get everything back in balance again. Of course there's no guarantee that you won't have to add even more noseweight before you're done.

The Chrysalis has a longer tail moment than the Monarch partly because of the larger wing area and chord. The volume coefficients really aren't that much different between the two designs.

Regarding performance, there's one other issue. The Chrysalis was originally designed when radios were a bit heavier. It was also designed to allow for folks on a budget, who might not want to spend mega-$$ for the latest in ultra-micro radio gear. This influenced the wing area. If you make the model drastically lighter, it will now have more wing area than it needs. This will reduce airspeed in flight, which hurts both penetration and Reynolds numbers. Meanwhile, nobody changed the amount of wind it has to deal with in flight. If we really want to make it tremendously lighter, then it should have a new wing design, with less area, which also changes the Reynolds numbers, which means we also need new airfoils. In addition, the reduced area and chord means a new tail design, which further alters structural characteristics and mass distribution, which causes further changes in wing and tail design, which causes.... See what I mean about ripple effects?

It's your model, and if you want to experiment with new tail designs, different moments, etc., that's fine with me. One of the advantages of the Chrysalis is that it's cheap and simple enough to allow for such experiments. We do recommend building one completely stock, so you have a control sample to compare with the results of your experiments. However, please recognize that when you start changing ANYTHING in the design of a model, you should expect some surprises. You should also plan on going through more than one design iteration before getting a result you're happy with. Even with all the experience and design tools we've accumulated over the years, Joe and I still usually go through several prototypes on any new design before we're entirely satisfied with the results.

However, if you just want to have a good model to fly, which someone else has already gone through all the blood, sweat and tears required to get all the parameters just right, then I recommend keeping your Chrysalis aerodynamically stock. Save all the weight in the tail that you can, then if you still need some nose ballast, put in a larger battery pack. At least then the extra weight is doing something useful. Joe put a 200 mah pack in one of his Chrysali, and still had a very light ship. Don't worry, it will fly just fine.

Don Stackhouse
DJ Aerotech