frets24

My Feedback: (15)

I started a 65" WS Royal/Marutaka P-51 Sr. when I was young and relatively inexperienced as a builder, then put it aside for decades. About 10 months ago I decided to revive the project. In the course of construction or during the various moves and storage over 23+ years there appears to have developed a twist in the left horizontal stabilizer.
Unfortunately I didn't notice this prior to glassing it 23 years ago or during the recent finishing stages until I started checking(yes, late...I know) elevator alignment and decoulage. As you can see painting and detailing are also completed leaving me with a real conundrum.

Is it possible to correct the twist without damaging the finishing work that has already been completed? If this were a simple Iron-on covered area the solution would be equally as simple but I have never had this kind problem on this type of build/finish.
The construction is much the same as a typical main wing, LE, TE, and ribs but NO spar, then sheeted, epoxy glassed and painted. Wondering how this would respond to heat and opposite force or just long term, gentle opposite force bracing?

The pics are a bit "busy" with the checker pattern but you can see the negative LE washout at the tip in relation to the root. The elevator is neutral at the root, and is flat and true, so the relationship is pretty obvious at the tip; looks like down elevator.

The twist measures at @ -3 deg. at the tip. Is this enough to worry about? Should I just let it be? How much would this affect the flying characteristics?


Posting in tips&techs and scale also.

REAL QUESTION FOR THIS BOARD:
The twist measures at @ -3 deg. at the tip. Is this enough to worry about? Should I just let it be? How much would this affect the flying characteristics?

wellss

Don't worry about it. It'll cause a slight left roll, nothing the ailerons can't compensate for......same with elevator trim

rmh

I have fixed warped wings, stabs of various types -including those like yours

Pattern planes at one time were glassed and painted and this same situation occasionally arose.
The trick is to determine which way the twiat must be made-then - on the bottom surface, make a diagonal slit thru the covering from LE to hinge line
this MUST be done such that the twist will close the slit when pressure is applied.
Block the stab into position after feeding in slow thin drying epoxy along the entre slit- then wiping excess awy with a damp rag -or with de natured alcohol
The result will be a a hairline crack on the surface if you were careful.

summerwind

can you twist it back to the normal posistion by hand without cracking anything?
if you can, and it has been epoxy coated, then twisting it back to normal with the use of a heat gun is going to help immensly.
epoxy will allow movement quite easily with the use of heat.
we repitched our epoxy glass props in Free Flight with the use of boiling water all the time to get the rpm and speed we wanted.

it works on finish surfaces too..........the trick is heating the surface evenly and patiently...........it may take 15 minutes of time to heat both sides evenly as you don't want to get it reall hot, just thoroughly heated so you don't burn paint off which you will do if you try to go too fast. then it's a matter of holding the surfacing in the new posistion till it cools.
the other trick is to use a heating pad (wrapped around the stab) and jig the the surface to stay in posistion.
once it cols, the epoxy will take on a new memory, and the wood too.

the method Dick just mentioned worked well on our pattern palnes as most of them were always foam/balsa sheeted and heat would not be applicable. with the Royal kit, the stab was sheeted over a rib structure, so the only precaution to take is to put a couple pin holes (bottom side) between the bays so hot air escapes and doesn't lead to puckering during the cooling off period.

if you leave the warp in and compensate with aileron trim, then you have a model with offset ailerons that really looks bad on the ground. 3* will cause some unhappy flight characteristics that i don't don't think you'll be happy with................it's your call

frets24

My Feedback: (15)

Thanks Wellss, D Hanson and S Wind,

I am really uncomfortable with a left roll tendency built in to a scale warbird and am somewhat reassured that it shouln't be a major issue. I still think loops would be adversly affected and it would always be a mental distraction on take-offs.

This board and the folks on it have always been a wonderful resource for inspiration when I hit creative brickwalls. you guys have once again pointed me in a good direction. I plan to pursue an incremental approach from least invasive/agressive to more so if or as required. Heat pad and opposite twist bracing/pressure, heat gun and greater force, cut it up. After 20+ years to get this far it doen't make sense to get in a hurry now so if it has to sit in blocks and braces for 3-4 hours or 3-4 weeks and the problem fixed and finish saved then that'll be great. If not, it can always be painted, decaled and weathered again.

Thanks again, I'll post results in a few days/weeks.

BMatthews

The amount of roll induced by the stab twist will be very, very minimal. It's the sort of thing that a pattern flyer would notice but most others would not ever be able to detect with any degree of certainty.

However what you ARE going to find with that twist is a strong tendency to climb that will require a pretty good amount of down trim to equalize.

Go easy on the heat if you try that option. I strongly suspect that by the time you get it hot enough to let the wood and resin ease up to twist it back that the paint will be bubbling and peeling off. You may be "stuck" with Dick's option along with a bit of paint touch up.

summerwind

just remember not to overheat the surface.............it does take some deal of time and patience to use the heatgun method, but it is very effective.

the roll rate induced by this warp is going to be more sensitive if the CG is aft.
either way, it's best to straighten this problem out before the first flight.....no point in adding suprises

rmh

Taint the roll
the BIG problem is just like a car with a low back tire the side of the stab with hte most drag ,will try to center its self on the line o flight
to restate
the center of the drag at the rear will try to center itself. It act's like rudder trim

summerwind

thank god you know what you're talking about Dick, but then again you always give solid advice.........you should be the moderator

seriously............posts like the previous one to mine really blow my mind

frets24

My Feedback: (15)

Not that it's going to matter since this twist IS getting fixed, whatever the method, but I would like to make sure I understand D Hanson assesment/logic. I think that the better one understands the hows and whys the better one is at (in this case) building, flying, and CORRECTLY trimming a plane...seen far too many rudder issues "corrected" w/ ail and vice versa... along with other miscorrected probs.

so, in this case, the simple issue is;

1)the Left stab will have greater induced drag which increases toward the tip where the negative inclination is greatest [fig 1] moment offset by arm
2)the center of that drag will attempt to move to the right in order to be at the centerline of travel (much like a brick on a string dragging behind a bicycle) 3) this causes the nose to swing left and would require RUDDER correction. [fig 2] moment attempts to establish linearity

compound issues are;

3)stab is acting as a stabilator with a negative ang of atk causing tail down/nose up=climb tendency [fig 3 & 4] requiring up(actually, down...thx Bmthws) elevator trim,(more drag on left, more rudder?) while at the same time...
4)the greatest twist is at the left tip again creating a moment offset by an arm causing left roll tendency [fig 5] requiring aileron trim

none of this looks good for an air frame that has a bad rep for left snaps on take-off

rmh

Here is another trim which is fun to play with

IF on puts a small plate at the top TE of the rudder
What is the effect on pitch?

frets24

My Feedback: (15)

First, did I get the previous stuff close? Or am I completely confused!?

Assuming the plate has no relative mass to affect the CG and what you are envisioning is like a 1'' x 1'' x 1/32 square (like a large trim tab) on a 60 size plane then:

a) if the plate is parallel to the flight path sticking straight back out of the rudder it seems that it would create a stronger effect further from the roll axis causing greater rudder-roll coupling. also, possibly, it would create some drag, offset from the datum line, thereby causing some nose up/climb tendency when the rudder is deflected. (if I got the brick behind the bike analogy correct in understanding you the first time round)

b) if it were placed perpendicular to the rudder then, following again on the drag trying to center idea, would it act as if there were some up trim on the elevator?

yes, no, kinda-sorta?

By the way...Your puppy avatar is adorable...black Labs have been the only type dog I've ever had. just one presently, he's #5

BMatthews

Summerwind, you may want to calm yourself down and adhere to polite RCU discussion rules or earn a time out. It's fine to disagree and discuss but do not belittle as you're doing.

I stand by all I typed. And frankly I don't see what you found so wrong in it.

Yes, the twist will have the effect of increasing the negative incidence in the stabilizer so it'll try to climb and considerable down trim will be needed to counter this. Looking at the first picture shows this clearly.

No the twist will not induce a significant amount of rolling action. First off the twist in the stabilizer is working over a very short moment arm compared to the span of the wing. Second the wing tends to damp out rolling inputs from sources that are not wing related. A fellow posted on another board about a short span biplane with a very long span stabilizer. Since the stabilizer was so big he thought he would be able to go with elevons and avoid the need for aileron servos. Even with a stab that was well over 1/2 the span of the wing the roll inputs did very little and the model was almost unflyable. It certainly wasn't in any way capable of the 3D style maneuvers that were intended. The wing acted to damp out the rolling inputs from the elevons with great gusto. The designer/builder said it was just barely able to turn at all even with large amounts of throw in the elevons on his Pitts'ish like design. So the P-51 with a stabilizer span that is far less than the wing span will produce an almost insignificant rolling moment from the amount of twist shown.

Yes I'm a big fan of using heat to correct a lot of misalignment woes. But in this case we've got a stabilizer that is built up with full sheeting over ribs and then has been prep'ed with glass cloth and finishing resin before being painted. As you know this produces a rather stiff structure. To soften the resin and lignum in the wood to allow the surface to be twisted back into correct alignment will require a lot of high temperature being applied from a heat gun or other source. And under that sort of temperature paints have been known to blister and peel. So I added a warning to heat the surface with care with an eye to closely monitoring the paint for signs of lifting. Nothing more and nothing less. I certainly did not suggest that heating the surface would be ineffectual or not to try it.

BMatthews

Close. In point 3) it's DOWN trim you'd need to balance the stabilizer's twisted in negative declage in order to cancel out the climb that the twist would try to induce.

Moving the elevator does two things. It alters the usually symetrical stabilizer airfoil into a cambered airfoil and it alters the center line of the overall stabilizer airfoil. The center line being from the leading edge to the trailing edge. So moving the elevator does these two things all at the same time. On the side with the twist it would be passing through the air in such a way that it would appear to be a cambered airfoil that is "flying" inverted. And that isn't going to do any wonders for the drag it's generating either. And as Dick says this higher trim drag will try to push the tail to the right/nose to the left and require some right rudder to counter it.

But again, given the amount you're showing in that first picture the effect would be quite minimal. By all means fix it but if you run into any major stumbling blocks in doing so and can only remove some of the twist then I wouldn't lose any sleep over it. Other factors play bigger parts in the requirement for trimming. For example if an aileron wasn't aligned well with the wing on one side so that there was a step in the airfoil shape it would likely have more effect than the stabilizer twist you have shown here.

BMatthews

You mean sort of like a Gurney flap but on both sides? So it's flat to the airflow?

If it's at the top of the rudder's trailing edge it's going to act like up trim and lift the nose..... unless the trailing edge is strongly swept back or forward in which case it'll cause some up trim for a swept back TE and a down trim for a swept forward TE. But that'll fight the drag induced up trim so it would take a really strongly swept forward TE to see any nose down tendency from the plate.

So... what do I win?

frets24

My Feedback: (15)

Nice catch on the down vs up trim. Parentheticaly corrected it above to preserve reference continuity.

These aeroD threads are fascinating, spent several hours over here today...learned alot....how do you guy know all this stuff down to " percentage KIP value at a 45* shear to the angular variation of the tangental load...."[sm=stupid.gif]or something like that.
I have a feeling this board has a lot of true engineers about. Anyway thanks for the education and cerebral overload[sm=drowning.gif]

This stab isn't really that substantial even though it is built up and sheeted, doesn't take a load of pressure with thumb and forefinger on opposite sides@ LE and TE to reverse twist it up to and a bit past true.
Just in messing with a 120* kitchen sink water heated gel pack and a refrigerator cold pack alternately over a couple hours or so is showing very promising results...@ 80% done.

wrap it hot for @ 15 min and it seems to penetrate and get everything just a wee bit looser. hand hold it for 5 min w/heat still applied, remove the heat and hold it for 5 min, then 5 more held with with the cold wrap on. then let it sit unbraced for @20 min and it feels room temp on the surface anyway. have 4 reps on it now. Seems to move back to old memory twist less during each 20-30 min rest period, I'll look at it again in the AM, see if the recent progress has held.

BMatthews

Nice discovery on that as a technique! We'll keep the virtual internet fingers crossed for you.

da Rock

Excellent technique. Heat is really hard to control with some methods. These packs seem to be a perfect way to apply a controlled amount.

Amazing what can be learned on these forums.

summerwind

BMathews,

i did not belittle you or anyone else.....i merely pointed out that your analogy of the warp was totally incorrect, so i guess i did disagree with you, and worded my thoughts in accordance to the minimal evidence of your analogy

however, i think it is way inappropriate of you to threaten me with a "time out" which means you are flexing your authority in a way as to tell me to shut up and say things to your liking.

i gave the OP 2 options in which to deal with the warp in which he took those options and came up with a heat related fix which is what he was after.

da Rock

Hold that thought.

This thread could use a timeout.