Float Drag

[Chris Finlay 0]

When the floats are down it seems that there is hardly any drag i.e. You need to close the throttles in order to descend but the speed remains fairly constant and the decent rate is 200 fps. I believe in the real Catalina that they use 13-15 inches of MP which give them about 200-300 fps decent rate. But in the FSX model this is not able to be done.

Chris

[Aerosoft Team [Inactive Account] 51558]

When the floats are down it seems that there is hardly any drag i.e. You need to close the throttles in order to descend but the speed remains fairly constant and the decent rate is 200 fps. I believe in the real Catalina that they use 13-15 inches of MP which give them about 200-300 fps decent rate. But in the FSX model this is not able to be done.

Chris

Lol, other say there is too much drag with the floats down.

Are you sure you are on 100% realism? I agree that it's a tad light, but that's a side effect from some other parameters, not really a lot that can be done there. Keep in mind the 'modern' models are normally a bit lighter then the originals tended to be. The Checklist, page power settings advises 17 inch, 2050 rpm and almost always with the cowl flaps open. I get close to what I expect then.

[ChrisB 11]

Lol, other say there is too much drag with the floats down.

hahaha... that was me!

Strange how perceptions vary... I found that with the tip floats down, the Cat sunk rather faster than I expected. Not that I am right of course, but...

[Aerosoft Team [Inactive Account] 51558]

Just an example of the simple fact I can never win.

But still we are looking at this issue and I will keep it on the front burner. Any more comments are highly welcome.

[Snave 466]

NO hands on experience of a Cat, but mathematically speaking the only `additional` drag from the floats comes from the mounting gear exposed to the elements, not the floats. They never retract, only swing through 90 degrees, so their drag remains the same - or very nearly if one pedantically recognises that the parasitic drag of the upper surface of each float body is only exposed when the floats swing down...

[Paul K 93]

I would agree with Snave on that. There isn't much more out in the airflow with the floats dangling than when they are tucked up; just the struts and bracing.

[Snave 466]

Thinking more upon this, with such a low aspect ratio wing the floats in the retracted position effectively act as small `endplates` to the wingtip and so might actually provide more lift for the aircraft through recovered vortex efficiency and effective wingspan increase, with the cavities exposed by the deployment of the floats leading to an appreciable loss of low speed lift in the high AOA or Low IAS regime.

Interestingly, the modern CL415 amphib also features these endplates, but not for lift, but to improve lateral stability - not sure if the same result was observed in the Catalina as it was never criticised for poor stability.

As the Cat, bless `er, spent practically its whole life in the low IAS regime the effect may be substantive after all - but not for the drag reason...

Anyone know the airfoil used on the Cat wing..?

[Finn 873]

First the aerodymics of the main wing itself is changed when the floats comes down.

Not alot, but should still have an effect.

The real drag numbers of the floats are not known, but I know that pilots sometimes used to lower the floats during a landing on land to hel bring the airspeed down, so offcourse some kind of drag increase must be there.

Finn

[Snave 466]

Yes, but as mentioned above the change could be decreased lift or increased drag, or both. I think it reasonable to assume drag increases when the struts are exposed, but the change of fineness ratio of what is already a very low-aspect ratio wing, and the absence of any washout on the outer wing panels would tend to indicate that the more important loss of lift component is at play. Drag increases as the square of the speed, so at the speeds the old Cat is capable of, marginal extra induced drag is probably hardly significant!

Possibly the landing argument was a placebo - you'd probably get better results by holding the nose up to present the great slab of wing to the prevailing airflow!

The included flight manual refers to a change in stall speed of 63.5 knots `clean`, 55 kts with gear down, and 57 kts with floats, but actually doesn't make a like-for-like comparison as in the clean case the throttles are closed, whereas the latter two are for 12in/2300rpm. For landing, 2 kts difference would hardly be noticeable in terms of the landing run but I s'pose every little helps...

[VulcanB2 5]

The floats may simply move through 90 degrees, but the physics isn't that simple.

RW Example: the speed brakes on the Vulcan, for their relative size, increases the drag of the aircraft 400%.

Whilst the floats are on the wingtip, they're relatively streamlined, but once they're hanging down in the airflow, below the wing, I'm betting they have quite an effect (not to mention causing a slight pitch down moment).

Best regards,

Robin.

[Snave 466]

The floats may simply move through 90 degrees, but the physics isn't that simple.

RW Example: the speed brakes on the Vulcan, for their relative size, increases the drag of the aircraft 400%.

Whilst the floats are on the wingtip, they're relatively streamlined, but once they're hanging down in the airflow, below the wing, I'm betting they have quite an effect (not to mention causing a slight pitch down moment).

Best regards,

Robin.

I'm not sure I follow the logic? The floats are subject to Form Drag wherever they are located, as their shape does not change, nor does their volume, (with the possible exception of the upper surface of the float exposed when rotated into the landing configuration, because when retracted this combines with the wing to eliminate this aspect of form drag).

Parasitic drag would necessarily increase as a result of the float empennage being exposed, and this would also impact on the efficiency of the underside of the outer wing panels, which are no longer contiguous. But with a low-aspect-ratio wing at low relative IAS the lift loss of the outer panels is a greater contributor to reduced aerodynamic efficiency than drag (drag increases as a square of the speed).

The original Cat manual clearly indicates the effect is actually minimal, and the later flying boats and amphibians presumably re-visited the drag assumptions and calculations and decided retracting floats was an unnecessary complication - the Martin PBM Mariner (which followed the PBY by just a few years) had retractable floats in the early versions, but fixed floats in the later variants... and post war flying-boats deemed them unnecessary, producing the suspicion of a pre-war `fad` rather than factual design proof.

As for the pitch down moment, there is no mention of it in the manual, yet the pitch reaction of full power application is covered on a multiplicity of occasions. Do you have a source for the conclusion?

[VulcanB2 5]

Hi,

Regarding the pitch down: simply the floats are drag, and when extended they are below the wing. This results in a turning moment which will result in a slight pitch down motion as there is nothing extended above the wing to counteract the force. When they're retracted, they're mostly out of the way.

I'm aware of later seaplanes having the fixed floats. I thought this was simply a weight/complexity thing as retracting floats are something else to go wrong. In the case of the Catalina however, given it can land on land too, maybe it was a ground clearance issue that was being resolved above an aerodynamic one?

Best regards,

Robin.

[Snave 466]

Hi,

Regarding the pitch down: simply the floats are drag, and when extended they are below the wing. This results in a turning moment which will result in a slight pitch down motion as there is nothing extended above the wing to counteract the force. When they're retracted, they're mostly out of the way.

I'm aware of later seaplanes having the fixed floats. I thought this was simply a weight/complexity thing as retracting floats are something else to go wrong. In the case of the Catalina however, given it can land on land too, maybe it was a ground clearance issue that was being resolved above an aerodynamic one?

Best regards,

Robin.

I don't know about the turning moment - it would depend on the precise position of the arc of rotation of the aircraft, and a parasol wing changes things substantially by moving the mean effective centre of lift considerably in the vertical axis, so dampening the pitch response (the aircraft body is now acting more as a pendulum). In effect, the floats moment is an almost-irrelevant addition to the already-considerable pitch effect.

Technically, the floats do not retract, they rotate, so their parasitic drag effect is almost constant, so really we would be considering the effect solely of the float empennage, so I doubt the effect is substantial. Which is presumably why it isn't mentioned in the manual.

When designed, the Catalina was a pure flying boat. Therefore whatever reasons for justification of the float retraction, landing on land isn't one of them. My guess is that in the 1930's when retractable gear was a new idea, the designers at Consolidated tried to modernise the design of the flying boat and this was one of the modern features deemed necessary for styling rather than outright functionality. Clearly the blisters, gun turret and profusion of aerials shows that they soon learned to think differently!

[Niklas 0]

Snave you mention both form and parasitic drag, but what about the changed airflow around the edge of the wing? The shape of the wing changes as the floats change position. So the induced drag created by the wingtip will change, how much I do not know. Which position that will create the most induced drag I do not know either, but there must be a change. The CG will change a little as well, but will be hardly noticable and with no effect towards the drag. The struts that hold the floats down will create some noticable drag. Anything else that could change in regards to the floats position?

Happy Flying

Niklas

[JorisVdB 0]

I have to admit, that if you're flying in standard mode, with the cowl flaps open, the Cat is very hard to slow down, she doesn't slow down a lot, it appears...not for such a big flying boat. I'll check it with the cowl flaps fully opened...

[Snave 466]

Snave you mention both form and parasitic drag, but what about the changed airflow around the edge of the wing? The shape of the wing changes as the floats change position. So the induced drag created by the wingtip will change, how much I do not know. Which position that will create the most induced drag I do not know either, but there must be a change. The CG will change a little as well, but will be hardly noticable and with no effect towards the drag. The struts that hold the floats down will create some noticable drag. Anything else that could change in regards to the floats position?

Happy Flying

Niklas

This was the point I made earlier about the endplate effect. But I don't know how relevant the vortex reduction can be for such a low-aspect ratio wing which already possesses high induced drag? Even in airliners with modern computer-designed wingtip finning the order of around 6-9% is the best efficiency increase one can expect - vital to an airliner with high-aspect ratio wings and a high cruise speed, less relevant to a thirties waterborne flying machine which actually is slower than most modern cars!

But induced drag remains as a square of the speed, and the endplate effect must be there, but it isn't a true endplate and the apparent spanwise increase is less significant as the floats are not true wing profile, so if there is any gain I surmise that it is probably from the struts and bracing being out of the breeze, rather than the floats. There must be some compromise from having no doors to shut and restore the underside of the outer wing profile after the floats and struts are deployed, but if the designers didn't figure that as significant I rather suspect the rest of it probably amounts to a few percentage points everywhere except for bragging rights - apart from the Dornier 26 I can't recall any other flying boats before the PBY having retractable floats and most were biplanes or sesquiplanes with wires everywhere!

If all designers were using the same slide rules, and the development of speed in the thirties was primarily the result of developments of waterborne craft - think Schneider Trophy - one wonders if it was such a good idea why they weren't all doing it!?!

[Niklas 0]

I can only agree with you Simon. The amount of the increase in drag, and how it should be simulated, is all guesswork. Probably even a FSX limitation

Great that you mention the Schneider Trophy, it was elemental in the development of airplanes in the 30'es. Just think that back then seaplanes were faster than planes with wheels

[VulcanB2 5]

Hi,

Good points.

In that case, may I suggest the drag with the floats extended seems extreme compared to with them retracted?

I've found that with the floats extended and the cowl flaps fully open, even at full power she is struggling to climb away (if I can get a positive rate of climb at all).

It seems there is little margin between the stall (~80 kts) and straight and level maximum speed in the water landing configuration.

I'll re-fly and provide some figures. A couple of times it has resulted in a forced landing.

Best regards,

Robin.

[fliger747 0]

Opinions like mileage varys. My general opnion is that the float drag is noticable and probably within the ballpark. Not having any data to reference, I can only reference a very experienced feel of airplanes. Several factors change here, the floats when extended change the endform of the wing, increasing both form and induced drag at that juncture, and take a faired structure and make a very unfaired structure out of it. Struts, especially dependent on their shape, can have a large drag factor. A round structure such as a bracing wire or tube will have about the same drag as a streamlined form perhaps seven or eight times it's diameter. A lot of "junk" hanging out in the slipstream.

For a long range aircraft, reduction in weight and simplicity probably outweighed any ideas about trying to streamline this area. In fact without any flaps, the drag can in many situations be advantageous.

It would appear that the developers have paid attention to this factor and have made a reasonable judgment.

Cheers: Tom

[Snave 466]

Hi,

Good points.

In that case, may I suggest the drag with the floats extended seems extreme compared to with them retracted?

I've found that with the floats extended and the cowl flaps fully open, even at full power she is struggling to climb away (if I can get a positive rate of climb at all).

It seems there is little margin between the stall (~80 kts) and straight and level maximum speed in the water landing configuration.

I'll re-fly and provide some figures. A couple of times it has resulted in a forced landing.

Best regards,

Robin.

A lot of this is going to depend on gross weight. As you don't mention the fuel state it's perhaps reasonable to assume that the Cat might struggle as it was not a high AOA aircraft so in the low/slow regime there is some delicate juxtaposition required between pitch/(IAS) and climb rate. Even at the best of times there wasn't much margin between stall and straight and level flight!

At half-fuel I have no trouble getting 400fpm on climbout, dirty, from land or water at METO - remembering that having BOTH gear and floats out would be an unusual situation... if I have both floats and gear deployed, then she barely flies - with 1200hp per side she is noticeably underpowered at 35,000lb max take-off compared to, say, a DC-3 (MTO circa 25,000...)

Reading between the lines of the official manual there is clear indication that a go-around or wave-off is marginal in many situations so I guess it's simply a matter of habit of getting floats up and gear up as soon as possible. At higher weights you'd probably be reaching for the fuel dump lever, too! She ain't no fighter!

[VulcanB2 5]

This is true - perhaps I'm expecting slightly too much.

I must be at MTOW when trying this (certainly full tanks). I'll reduce the fuel loading.

Best regards,

Robin.

[Snave 466]

Yes, unless the thought of 3,000+ nautical miles at 90 kts is something you'd enjoy sitting in the chair for, it's probably best to reduce the fuel load.

You'd only need full tanks for a Qantas-emulating 32 hour flight, and you can be certain FSX would find some excuse to collapse and die long before the airplane ran out of juice!

[fliger747 0]

A reasonably good test of the correctness of the thrust/drag ratio is the single engine performance. Casual evaluation seems to suggest that indeed careful flying is necessary at higher weights and one does not want to get too slow. However such carefull flying does reward one with continued aviation even at fairly high weights.

The only complaint I might voice is the rudder seems too effective in that one can continue controlled flight at speeds somewhat below the VMCA.

Cheers: T.

[Snave 466]

A reasonably good test of the correctness of the thrust/drag ratio is the single engine performance. Casual evaluation seems to suggest that indeed careful flying is necessary at higher weights and one does not want to get too slow. However such carefull flying does reward one with continued aviation even at fairly high weights.

The only complaint I might voice is the rudder seems too effective in that one can continue controlled flight at speeds somewhat below the VMCA.

Cheers: T.

Possibly, but it could equally be a controller sensitivity issue on your rig. What are you using for rudder control and how is the rudder sensitivity set within FSX? Using USB CH Pedals I have had problems with over-sensitivty in FS9 and FSX and I solved it by bypassing the adjustment in-sim and simply setting the parameters through FSUIPC.

The simple tweak would be to change aircraft.cfg

[flight tuning]

rudder_effectiveness = 1

to some lower value that meets your expectations

The rudder area and angular movement is correct in the .cfg

[fliger747 0]

Controller sensitivity should have no effect on the total force applied by the rudder at any given angle, just will adjust the ratio between your controller deflection and the sims controll deflection. If one measures the angular deflection with a utility such as AFSD the maximum value will not be exceeded no matter how your setup is calibrated.

The rudder effectiveness characteristics are controlled largely by the tables in the .air file. Ie. table 519, Rudder deflection (Moment) Factor vrs. Q. The .cfg file scalar does not affect the shape of the curves. The table 460, weathervane stability also has an effect. Additionally the entries in the primary aerodyanmics (rec 1101) will have a big effect. There is not just one thing going on here.

At VMCA, one runs out of rudder force to counteract the yaw caused by asymetric power application. Since the rudder relies on dynamic pressure (Q) to derive it's force, it will (very simplified) loose force with the square of the decrease in velocity. So as speed decreases, full rudder no longer is sufficent to keep the aircraft in directional control. Either speed must be increased or power reduced to maintain aircraft control. For multi engine aircraft this is a critical property, one we must compute for every takeoff for the aircraft that I fly. It will vary with weight, CG, density altitude and power available.

Cheers: T.