Floatplane Destabilization and UFIW (Uncontrolled Flight Into Water)

Last summer while getting the slant tail series of Cessna 172's added to our STC for the Cessna 170 and straight tail 172's for the installation of Aerocet 2200 floats, we made some startling discoveries regarding how minor changes can and do affect aircraft stability. An article was published in the January/February edition of Water Flying. Since that time we have done further flight test on various aircraft and float combinations and made even more remarkable discoveries.

We had completed all company testing and the aircraft was ready for FAA flight testing. While we were waiting for the FAA to be able to test the aircraft/ float combination, the engineering firm I was working with suggested putting spray rails on to help keep water erosion minimized. After a walk around the float pond and noting spray rails on several aircraft of the same make and model, we determined that there was no consistency to size and shape of spray rails. Some were just a few feet long and some all the way from the bow to the front spreader bar ranging from 3-4 inches wide or, below the chine. Pondering this we finally determined that spray rails didn't really need to extend beyond the propeller face as bow wake does not reverse direction.

The first set we made were 3 inches past the propeller face and 3 inches below the chine. Taxi tests were excellent with bow spray deflected downwards as opposed to what it originally was which was inboard and up like the outside bow wake. Flight testing revealed we had created another issue. All the directional stability tests we had passed we now failed.

After several more attempts to regain directional stability without adding a larger ventral or even strakes to the after fuselage, we finally made one that ended at the propeller face and extended down one and a half inches below the chine. Taxi tests were great with bow wake still deflected downwards and we finally had our directional stability back.

We discovered that adding spray rails, or as I now call them, bow strakes, will destabilize the aircraft causing it to fail the directional stability tests all aircraft are required to pass before they can be certified. This is because adding anything forward of the aircraft c.g has a destabilizing effect while adding aft such as ventrals or strakes stabilizes the aircraft. If you look in most float manufacturer parts lists or IPC's, you will see that they do not show spray rails. Most were certified without spray rails/bow strakes and they were added later. If you look at many of the same make and model of aircraft on the same exact floats, you will see widely different approaches to spray rails to prevent erosion of the propeller. Most are 3 inches below the chine and almost all go far past the propeller face some go as far as the front spreader bar, one aircraft had them all the way to the step.

The two tests aircraft must pass that concern us here is the sky brodie and the forward slip. These two tests must be done left and right at various airspeeds and flap settings. The sky brodie is one in which the wings are held level with aileron and full rudder is input. The aircraft starts to "pivot" like a windsock spinning, rudder is released and the aircraft needs to show a tendency to recover. This means the ball has to come off the side of the glass on the slip skid indicator.

The forward slip both left and right in various flap positions is one where you use aileron and rudder fully to slip the aircraft and release the aileron while holding the rudder fully in. The low wing has to recover and lift back to level. This is why many aircraft on floats are limited to less than full flaps.

All aircraft passed this test initially and then when bow strakes were added without testing, all we have tested, fail the two above tests. In addition, the spray rails on most aluminum aircraft are riveted on to an angle attached to the chine. This leaves a small gap between the chine and the rail. Taxi tests and air tests show that not only does the aircraft not meet its directional stability requirements, the water is not directed downwards but instead is forced through that narrow gap into the propeller. Think of putting your thumb over the end of a garden hose. So now the aircraft is destabilized and the attempt to curb erosion has only made it worse. One way to fix the erosion problem is to fill the gap with a sealant like 4200.

Also it does not have to be wide to be effective at deflecting the bow wake. After several tests we had ours on the 170 and 172 just one and a half inches below the chine and we still had a good bow wake deflection and retained our directional stability. In addition, the 3 inch long bow strakes tend to deflect the water straight down sort of like spraying a hose into a mud puddle which we found then deflected back up and into the propeller.

The narrower rails are a more subtle and gentler approach and tend to gently deflect the bow spray and not have the downward force the wider rails have which tends to deflect the water back up.

Why is this important? I know of many pilots with lots of experience as I m sure many readers do who were flying their aircraft and landing. Due to winds or clearing an obstacle a slip was initiated. These were done within the parameters of the seaplane flight manual supplement. When they initiated opposite controls to recover, nothing happened. Stick and rudder had no effect. Nearly everyone used full power to no avail and had what we now term UFIW, Uncontrolled Flight Into Water. Perhaps raising flaps in conjunction with adding full power would have helped but at that point you have run out of airspeed, altitude and inspiration, at the exact same moment. UFIW is inevitable.

We added the strakes to the STC for the 172 slant tail series but it has a different ventral than the 170 and straight tail 172 so they could not be added. I did not want to put the 170 back in Experimental and do flight testing again so, I decided to tape them on the 170 and do our own flight tests. We did this and it passed as we thought it would. The next step was to make it legal.

Legal? But it's just a minor modification. Well a quick review of FAR 43 specifically Appendix A shows this is not the case. Here is the FAR reprinted just to item vii:


FAR 43 - Appendix A

Major Alterations, Major Repairs, and Preventive Maintenance

(a) Major alterations

(1) Airframe major alterations. Alterations of the following parts and alterations of the following types, when not listed in the aircraft specifications issued by the FAA, are airframe major alterations:

(i) Wings.

(ii) Tail surfaces.

(iii) Fuselage.

(iv) Engine mounts.

(v) Control system.

(vi) Landing gear.

(vii) Hull or floats.

Hulls and floats are Major alterations and require a 337 to be completed. The FAA office I dealt with was unfamiliar with this as well when I sent in my application for a 337 for altering the floats on that aircraft combination.

There are many STC's for float lockers on various floats. There are none for spray rails/bow strakes. I got into a heated conversation once with an FAA inspector as I installed float lockers and on the 337 listed the floats by serial number. He wanted it for the aircraft. I explained the floats were an appliance and they were being altered not the aircraft. I asked what happens when the floats get sold separate from the aircraft? He was adamant so I filled it out for the aircraft. I'm unsure what ever happened I suppose the buyer of the floats could ask the aircraft owner for a copy of the 337. My 337 lists the aircraft as well as the floats by serial number.

I believe this is a serious issue that needs to be addressed. I'm sure many accidents, injuries maybe even deaths have occurred due to loss of directional stability and control. This issue is no different than the Boeing 737 Max, just fewer people in floatplanes. I'm sure any accidents are "loss of control." The reasons for that may have been attributed to pilot error when in fact the floats being altered and not tested for directional stability could easily have been a factor.

These tests are part of the CAR's which predates the FAA. The CAA was 1938-1958. The last amendment to the CAR's was in 1949. So now that we've been doing the same thing wrong for 81 years and calling it experience, where do we go from here?

The first thing to do is determine if your floats had spray rails when tested. If so, nothing needs to be done. If not, the second step would be to conduct the tests at a safe altitude to see if your current configuration passes the tests. You may want a qualified test pilot to conduct these tests if you don't feel comfortable doing so.

If it passes then you would have your IA submit a form 337 for a field approval. The data package should include your flight test report and findings.

If it does not pass then the next step is to remove them, fly the aircraft and do the tests to determine that is does pass without the strakes installed. Then make some strakes you can tape to the floats. They do not need to be longer than the front face of the propeller. I would do several water taxi tests with these strakes. Remember to use aluminum tape or Shur tape and to tape both sides so they stay attached to the floats for testing. Start with one inch below the chine and see what the bow wake looks like. As we discovered it will degrade if you do not have enough strake below the chine. Once you find an acceptable depth, it is time to test fly and see if it passes the stability tests. If it does then again fill out a 337 and get your major alteration approved.

If it does not pass then you may have to add something behind the center of gravity so it will pass the directional stability tests. On some aircraft it is as simple as a ventral fin below the belly. Some aircraft will only need strakes, angle attached to the belly which may alter it enough to pass the tests. Cessna 180 series have springs attached internally in the tail to center the rudder. There are many ways and as long as the final outcome is a safe and stable aircraft everyone wins.

In the interim simply reducing the size of your rails would be an excellent option. In taxi position drop a plumb bob from the face of the propeller and see where it ends on the rail. Cut the rail that far aft and then using a cutoff wheel make it no more than one and one quarter to one and one half inch below the chine. It you have a gap between the chine and the rail seal it with 3M 4200. I was going to initially drill out the angle on the floats I have helped modify and re rivet them but it was easier to simply use a cutoff wheel and cut the leg angle sticking down.

If anyone has further thoughts or questions on this subject feel free to contact me.

This is a fantastic article, thank you

"This is because adding anything forward of the aircraft c.g has a destabilizing effect while adding aft such as ventrals or strakes stabilizes the aircraft."

I don't think is categorically true with respect to pitch stability, although it is generally true with yaw.

The likely problem with float strakes is that they create a positive lift vector ahead of the CG, which moves the combined center of pressure forward, closer to the CG. As noted, this does decrease stability. All the strakes I've seen would create lift.

However, if the strakes created a negative lift vector they would move the combined center of pressure further rearward, and increase static stability. This creates extra form drag from the strake, induced drag at the tail which now has to compensate for it, and increases the bending load on the tail boom.

A very interesting story, and I enjoyed the read.

Awesome write up!! Well worded and with excellent information. Your experience is a great example of unintended consequences and how they can affect us all. And just when I was thinking the internet totally sucked.

jcadwell wrote:

"This is because adding anything forward of the aircraft c.g has a destabilizing effect while adding aft such as ventrals or strakes stabilizes the aircraft."

I don't think is categorically true with respect to pitch stability, although it is generally true with yaw.

The likely problem with float strakes is that they create a positive lift vector ahead of the CG, which moves the combined center of pressure forward, closer to the CG. As noted, this does decrease stability. All the strakes I've seen would create lift.

However, if the strakes created a negative lift vector they would move the combined center of pressure further rearward, and increase static stability. This creates extra form drag from the strake, induced drag at the tail which now has to compensate for it, and increases the bending load on the tail boom.

A very interesting story, and I enjoyed the read.

There are strakes and then there are strakes. I BELIEVE what he's talking about are strakes that are mounted vertically, down from the chine. In fact, many strakes on floats are, as you're referring to, mounted horizontally, protruding inward from the chine. The former is essentially an extension of the inboard vertical side of the float, while the latter is effectively an extension of the float bottom, only more horizontal.

The horizontal strakes (for lack of a better term) have little if any effect on yaw stability in my experience. In my experience with vertical strakes, which are in fact installed on some production, certificated floats, they are generally quite small, and I've never noticed the tendency he described.

My Cessna 170 was for years mounted on PeeKay B-2300 straight floats. Those are quite large floats, originally designed for the Hawk XP, a significantly heavier airplane. On those floats (and there were no strakes installed), the 170 was significantly yaw unstable. Shove a rudder all the way to floor and the airplane would yaw dramatically, and pretty much stay there. I've slipped that airplane with all flap settings and never encountered any tendency to NOT recover from the yaw, with proper control inputs.

That float installation was approved (via STC) in Alaska, and I knew the FAA test pilot....and he was a very practical gent. He approved some stuff that I've seen that probably wouldn't get by another FAA type.

In fact, that airplane was sufficiently unstable in yaw that PeeKay offered a ventral fin as an option. Even though it wasn't required, I bought one after the first season flying the plane. That fin did help, but the plane was still very yaw unstable. It taught everyone who flew it what rudder pedals are for. That said, the 170 is slightly yaw unstable to begin with.

As to the reason some float planes are prohibited from using full flaps (like the Cessna 206 on Wipaire floats), that is a limitation because that airplane is legal at 3800 pounds and can't meet initial climb criteria at max weight and worst case scenario. And, yes, those floats have small vertical strakes.

Take a look at the Grumman Goose. BIG horizontal strakes, all the way from the bow around to behind the engines. Those strakes, like most of the horizontal strakes, are curved downward, to deflect water away from the prop.

I've flown a few floatplanes without strakes, and never had any problem keeping water out of the prop.....Except the deHavilland Beaver I flew in Kodiak. It was mounted on Bristol Aero 4580 floats (license built in Canada to EDO specs). But, that airplane was equipped with a VERY long propeller (one foot longer than approved). Once we put a legal prop on it, keeping the prop out of the water using good technique was no problem.

As to the concept of "Controlled Flight Into Water", I've never heard of such a thing, and while I've seen a number of seaplane accidents, I've never seen nor heard of an accident that would've been caused by simply flying the plane into the water, sans flare.

Which, of course, does not imply for a moment that such a phenomenon does not exist.

As I noted, horizontal chines, curved downward, work fine. Small vertical chines existed on several floats I've flown a lot, and I never got any suggestion that they destabilized the plane, but these were all factory installed, which implies they completed certification flight test. Maybe.

I've never hated ventral fins, like some seaplane guys do. They really make many seaplanes much nicer to fly in cruise, and for that I'm willing to deal with the few downsides of a ventral fin.

MTV

Hey MTV I’m pretty sure we know each other from Kodiak days the 3 cases of UFIW I refer to all occurred in Kodiak with pilots I’m sure you know, two in Beavers and one in a Super Cub.

While rare it is possible in the right circumstances to have it occur. All 3 quit doing slips with flaps and 2 are air taxi pilots.

You are correct the strakes are called spray rails and are vertical off the bow. Also you have to consider the arm outboard from the center of the aircraft as well as the forward and aft arm.

Lots of other things to consider too

Hey MTV I’m pretty sure we know each other from Kodiak days the 3 cases of UFIW I refer to all occurred in Kodiak with pilots I’m sure you know, two in Beavers and one in a Super Cub.

While rare it is possible in the right circumstances to have it occur. All 3 quit doing slips with flaps and 2 are air taxi pilots.

You are correct the strakes are called spray rails and are vertical off the bow. Also you have to consider the arm outboard from the center of the aircraft as well as the forward and aft arm.

Lots of other things to consider too

Yes, there is a lot going on there, and as you’ve noted, modifications to floats seem to be treated pretty casually by pilots, mechanics and the FAA.

As I noted, I’ve never been bothered much by Aircraft with ventral fins. All our Beavers had them as did the 206s....just something else you have to pay attention to around beaches and docks. But, flying yaw unstable airplanes is not much fun.

MTV

Very interesting discussion going on here. From an engineering perspective, it makes sense. Any surface area forward of CG/CL destabilizes the aircraft. Surface area behind stabilizes. This is the idea behind spades on control surfaces. They are ahead of the hinge line to "destablize" the control surface, making it much lighter feeling by pilot. Crazy to imagine that small things can make a big difference but a floatplane is a very aerodynamically complex system I guess.

This is a huge reason why normal category is the only thing available on a 172 on floats, and why the 182 is a normal category only as well.

My SES training in in the 172 included a brief but deeply convincing demonstration of the lack of lateral stability in a departure stall, something that is ho hum in a land version. And while a 182 is not certified for spins (for a good reason....cg range), it behaves well with forward cg, and recovers dramatically more slowly as the cg moves rearward towards the CP.

This is a huge reason why normal category is the only thing available on a 172 on floats, and why the 182 is a normal category only as well.

My SES training in in the 172 included a brief but deeply convincing demonstration of the lack of lateral stability in a departure stall, something that is ho hum in a land version. And while a 182 is not certified for spins (for a good reason....cg range), it behaves well with forward cg, and recovers dramatically more slowly as the cg moves rearward towards the CP.

Wipaire did full certification flight test on a Cessna 206 on floats, including spins, and with a pretty large lake boat strapped to the float struts.

They decided not to go forward with external load certification for a lot of very good reason, but I was told it spun okay. And recovered okay.

MTV