From: Marc J. Zeitlin Subject: Re: Short Canard Performance Date: Mon, 10 Apr 95 11:48:38 EDT Eric wrote (to me): >I read with great interest your rah posting on the shorter canard performance >- it was very good and cleared some things up in my mind. I am not an >aerodynamisist, I'm not even sure that is a word, but it is great to read >peoples thoughts like yours. You should post it to our builders group just >in case someone missed it. Since Eric requested it, I've appended the discussion I've had with George Venkatesh on r.a.h. (rec.aviation.homebuilt). Coincidentally, I think Cliff C. and Sid L. started the whole thing. Comments are welcome. P.S. - This stuff is LONG! -- Marc J. Zeitlin E-Mail: marcz@an.hp.com >From news.an.hp.com!news!marcz Mon Apr 10 09:56:36 1995 Article: 3241 of rec.aviation.homebuilt Path: news.an.hp.com!news!marcz From: marcz@news.an.hp.com (Marc Zeitlin) Newsgroups: rec.aviation.homebuilt Subject: Re: Deep Stall, another point of view Date: 5 Apr 1995 20:07:55 GMT Organization: HP Patient Monitoring Division Lines: 154 Distribution: world NNTP-Posting-Host: hpwarhw.an.hp.com Warning: Long, theoretical post to follow! (SidLloyd) writes: >(Charles K. Scott) writes: |> >The published results unfortunately did not provide the effect |> >on the rotation speed nor discuss other implications of a shortened |> canard. |> >There may have been subsequent results published. Sid? |> |> Yes, a LOT of us have this exact concern. Nat has not published anything |> on increased takeoff/landing speeds, increased takeoff/landing distances, |> and lower front seat weight capacity due to shortening the canard by 6" as |> mandated by Nat. I've been thinking about this issue a lot since the original posting. Let's examine the questions of rotation speed, landing speed and landing roll, stall speed, as well as front seat weight capacity. General Theory: The overall wing area of the Cozy MK IV is ~120 sq. ft. The area of the canard is ~12 sq. ft. before cut-back and ~11 sq. ft. after cut-back. While I don't know exactly what lift coefficient the canard flies at with relation to the wing, it's probably only a bit higher (so that it will stall first), so let's assume that the canard produces about 10 - 12% of the total lift of the aircraft (~ the area ratios). So, what effect will making the canard 8% smaller have? Since canard lift is proportional to both "canard lift coefficient" and "canard wing area", a decrease in the wing area will lead to a directly proportional increase in lift coefficient (at the same velocity). This means that the canard lift coefficient at any given speed will be ~8% higher (with the shorter canard) than it would have been with the longer one. The canard has a maximum angle of attack (whatever it may be), and the whole point of reducing the span of the canard was to reach this angle at a lower AIRCRAFT angle of attack, reducing (and possibly eliminating) the ability of the main wing to stall. Since the canard angle of attack is determined by both the aircraft angle and the elevator deflection angle, the elevator will have to deflect more to reach the same canard angle of attack (and higher lift coefficient) at a lower aircraft angle of attack. Now we're getting somewhere. A drag issue arises - is there more or less drag at the higher elevator deflection, and what does this mean if so? While the induced drag coefficient on the canard will be higher at higher elevator angles (effective angle of attack) [we'd need the lift/drag polar of the canard airfoil to see exactly how much] the area of the canard has gone down, so these two effect would tend to cancel each other out. Also, the profile drag of the canard will decrease, due to the decreased area. Without a rigorous analysis, we can't say whether the canard drag will increase or decrease slightly at the same airspeed, but the effect WILL be slight. The canard (as we showed before) produces about 10% -12% of the total lift, and probably 5% - 8% of the total aircraft drag, and we'd only be changing that small portion of the total drag by some small amount. 1) Rotation Speed: This is the speed at which you'd like to get the nose up, create a positive wing angle of attack, and begin producing lift. No one ever recommends doing this either below (or even close to) stall speed (for obvious reasons). This means that at the 80 kt (or so) COZY rotation speed, you are WELL above the canard stall speed, so you do not need anywhere near full elevator deflection to cause rotation to occur. You will (with the shorter canard) need slightly more deflection (on the order of ~5% -10%, depending on the lift-curve-slope of the airfoil) to produce the same amount of lift as with the longer canard, but since you're not near the stall, this elevator travel is certainly available to you. Prognosis: No noticible change in rotation speed - slightly more elevator deflection (possibly noticible in direct comparison, possibly not). 2) Landing Speed (and Landing Roll): Similar arguments apply to the landing speed. Since with canard aircraft you do not do full stall touchdowns, but essentially fly the aircraft onto the runway at speeds well above canard stall speed, we again have a situation whereby we have slightly higher elevator deflection to achieve the same canard angle of attack at the same aircraft (main wing) angle of attack (and velocity). This isn't a problem, as we're not near the maximum elevator deflection anyway. Remembering the drag argument made before, we've got about the same amount of drag at the same airspeed. Prognosis: No noticible increase in drag, and no noticible change in landing speed - slightly more elevator deflection to obtain the equivalent speed. (Since the landing roll is primarily determined by the touchdown speed, this should not be affected either). 3) Stall Speed: Since the Lift-Curve-Slope of airfoils becomes highly non-linear (the lift coefficient becomes constant, and then decreases with increasing angle of attack) at high angles of attack (near stall) we arguably may actually be operating the main wing at a slightly HIGHER lift coefficient at the slightly lower wing angle of attack obtained at the canard stall speed. In any case, it is very doubtful that the lift coefficient would be lower, since the wing angle of attack is not substantatially lower, and we're still operating in the flat region of the curve. Prognosis: Stall speed does not increase, since the main wing can produce the same lift at the same speed, but at a slightly lower angle of attack. Possibly, the wing may produce more lift, and the stall speed could DECREASE. This, however, is unlikely to be noticible. 4) Front Seat Weight Capacity: Prognosis: As long as the C.G. is kept within the recommended envelope (which will guarantee that the canard can be stalled at maximum elevator deflection or less) then the weight capacity of the front seat will not be affected. I realize that without the equations and numbers to back all this up, some may not be convinced, but I believe that the logical arguments made here show that the worries some may have over the canard shortening may be baseless, and that the performance of the aircraft in it's recommended C.G. range should not be noticibly effected by the change in canard span. If someone would like to run the equations with the real COZY numbers (I really don't have the time) I'd be happy to review them. (Flame proof suit on) :-). -- Marc J. Zeitlin E-Mail: marcz@an.hp.com >From news.an.hp.com!hpscit.sc.hp.com!news.dtc.hp.com!canyon.sr.hp.com!col.hp.com!sony!nntp-sc.barrnet.net!nntp-hub2.barrnet.net!olivea!rahul.net!a2i!news.clark.net!news.sprintlink.net!uunet!walter!lowell.bellcore.com!heron!venky Mon Apr 10 10:48:11 1995 Article: 3374 of rec.aviation.homebuilt Path: news.an.hp.com!hpscit.sc.hp.com!news.dtc.hp.com!canyon.sr.hp.com!col.hp.com!sony!nntp-sc.barrnet.net!nntp-hub2.barrnet.net!olivea!rahul.net!a2i!news.clark.net!news.sprintlink.net!uunet!walter!lowell.bellcore.com!heron!venky From: venky@heron.bellcore.com (G. A. Venkatesh) Newsgroups: rec.aviation.homebuilt Subject: Re: Deep Stall, another point of view Date: 6 Apr 1995 15:59:12 GMT Organization: Bell Communications Research Lines: 60 Distribution: world NNTP-Posting-Host: heron.bellcore.com In article <3lutar$5o6@hpaneqb4.an.hp.com> marcz@news.an.hp.com (Marc Zeitlin) writes: >Warning: Long, theoretical post to follow! > >General Theory: > >So, what effect will making the canard 8% smaller have? Since canard lift >is proportional to both "canard lift coefficient" and "canard wing area", a >decrease in the wing area will lead to a directly proportional increase in >lift coefficient (at the same velocity). This means that the canard lift >coefficient at any given speed will be ~8% higher (with the shorter canard) >than it would have been with the longer one. There are some imprecise statements there. I am not an aerodynamicist but my understanding is as follows. Anyone is welcome to correct me. The lift-coefficient for a wing depends on the aerofoil design and the AOA. Beyond the stall AOA, the lift-coefficient drops off rapidly. Neither of those have anything to do with the wing span (ignoring boundary conditions). The magnitude of total lift on a wing at any time is proportional to the wing span (or more precisely, the wetted area?) and square of the relative velocity. > >The canard has a maximum angle of attack (whatever it may be), and the >whole point of reducing the span of the canard was to reach this angle at a >lower AIRCRAFT angle of attack, reducing (and possibly eliminating) the >ability of the main wing to stall. Despite popular belief, simply shortening the canard has no effect on when the canard will stall in relation to the main wing. The canard will still stall at the same AOA as the longer canard. However, the force balances are affected. In stable (neither climbing nor descending) flight, a plane with a shortened canard should fly with the nose up slightly to generate the same amount of lift as a plane with a longer canard. What the shortened canard does is simply reduce the margin between AOA for stable flight and maximum AOA. In practice, a shortened canard also reduces the pitch-up moment that can be generated by the canard. Depending on the differences in the maximum AOAs in the canard and the wing, a large pitch-up moment MAY take the plane beyond the maximum AOA for the main wing even though the canard stalls earlier due to momentum alone, especially in aft CG conditions. However, curtailing that tendency by shortening the canard has other side-effects. >Since the canard angle of attack is >determined by both the aircraft angle and the elevator deflection angle, >the elevator will have to deflect more to reach the same canard angle of >attack (and higher lift coefficient) at a lower aircraft angle of attack. >Now we're getting somewhere. > Actually, therein lies the problem. There is a finite amount of travel that the elevator can do before (depending on the AOA of the plane relative to the wind), the canard will stall or the elevator will hit the stops. For take-off rotation this can be a liability. Most (if not all) canard planes use maximum deflection to start rotation. Reduced elevator authority (due to shortened canard) will mean higher speed required before rotation even if the speeds under consideration are well above canard stall speeds for that configuration. Similarly coming in too low and too slow with the nose down may result in an inability to lift the nose up with the elevator at maximum deflection. venky >From news.an.hp.com!news!marcz Mon Apr 10 11:45:00 1995 Article: 3419 of rec.aviation.homebuilt Path: news.an.hp.com!news!marcz From: marcz@news.an.hp.com (Marc Zeitlin) Newsgroups: rec.aviation.homebuilt Subject: Re: Deep Stall, another point of view Date: 10 Apr 1995 15:31:26 GMT Organization: HP Patient Monitoring Division Lines: 142 Distribution: world NNTP-Posting-Host: hpwarhw.an.hp.com (G. A. Venkatesh) writes: |> (Marc Zeitlin) writes: |> >Warning: Long, theoretical post to follow! |> > |> >General Theory: |> > |> >So, what effect will making the canard 8% smaller have? Since canard lift |> >is proportional to both "canard lift coefficient" and "canard wing area", a |> >decrease in the wing area will lead to a directly proportional increase in |> >lift coefficient (at the same velocity). This means that the canard lift |> >coefficient at any given speed will be ~8% higher (with the shorter canard) |> >than it would have been with the longer one. |> |> There are some imprecise statements there. I am not an aerodynamicist but my |> understanding is as follows. Anyone is welcome to correct me. |> |> The lift-coefficient for a wing depends on the aerofoil design and the AOA. |> Beyond the stall AOA, the lift-coefficient drops off rapidly. Neither of those |> have anything to do with the wing span (ignoring boundary conditions). The |> magnitude of total lift on a wing at any time is proportional to the wing |> span (or more precisely, the wetted area?) and square of the relative |> velocity. You are correct. The LC depends on the design and the AOA, not on the span. What I meant to say (and apparently did not do precisely enough) was that: since the CANARD is smaller but the main wing is the same, the canard needs to create the same amount of lift at the same speed, but with less area. The only way to do this is by flying at a higher AOA, and the only way to do this at the same AIRCRAFT AOA (to maintain the same airspeed) is to have slightly more elevator deflection. The constant of proportionality in the equation you cite is the LC. |> >The canard has a maximum angle of attack (whatever it may be), and the |> >whole point of reducing the span of the canard was to reach this angle at a |> >lower AIRCRAFT angle of attack, reducing (and possibly eliminating) the |> >ability of the main wing to stall. |> |> Despite popular belief, simply shortening the canard has no effect on when the |> canard will stall in relation to the main wing. The canard will still stall at |> the same AOA as the longer canard. No and yes. While the canard AOA at canard stall may be the same, it will be achieved at a lower aircraft AOA, due to the reasons cited earlier; i.e. higher canard LC and at the same speed. So, due to elevator deflection, the canard AOA will be higher, the aircraft AOA will be slightly lower, the wing LC will be about the same, the speed will be about the same, and the main wing will be further away from stall than it was before the canard was shortened. |> In stable (neither climbing nor descending) flight, a plane with a shortened |> canard should fly with the nose up slightly to generate the same amount of |> lift as a plane with a longer canard. This can't be the case. If the plane is flying more nose up (higher aircraft AOA) then the LC of the wing will be higher (at the same speed) and the aircraft will climb (not level flight). As I stated in my previous post, the canard LC will increase and the canard effective AOA will increase NOT because of a nose up attitude (which would change the wing AOA) but because of increased elevator deflection. |> .......................... What the shortened canard does is |> simply reduce the margin between AOA for stable flight and maximum AOA. I'm not sure what you mean here. There are an infinite # of level flight AOA's, depending upon speed. They continue right up to the max. AOA. What am I missing? |>...................... In |> practice, a shortened canard also reduces the pitch-up moment that can be |> generated by the canard. Depending on the differences in the maximum AOAs |> in the canard and the wing, a large pitch-up moment MAY take the plane beyond |> the maximum AOA for the main wing even though the canard stalls earlier due to |> momentum alone, especially in aft CG conditions. You are correct. Again, I was not clear enough. I was speaking of non-accelerated stall. Momentum and/or rough air may take ANY plane into an accelerated stall. |> >Since the canard angle of attack is |> >determined by both the aircraft angle and the elevator deflection angle, |> >the elevator will have to deflect more to reach the same canard angle of |> >attack (and higher lift coefficient) at a lower aircraft angle of attack. |> >Now we're getting somewhere. |> Actually, therein lies the problem. There is a finite amount of travel that |> the elevator can do before (depending on the AOA of the plane relative to the |> wind), the canard will stall or the elevator will hit the stops. Absolutely. |> ........................ For take-off |> rotation this can be a liability. Most (if not all) canard planes use maximum |> deflection to start rotation. Reduced elevator authority (due to shortened |> canard) will mean higher speed required before rotation even if the speeds |> under consideration are well above canard stall speeds for that configuration. If in fact this is the case, then I was incorrect. I assumed that these aircraft did NOT use full elevator deflection at rotation. If they do, then the shortened canard will definitely affect rotation speed (but since lift goes as the square of velocity, the difference will be about the square root of 1.08 [~ 1.04] or about 4% higher to achieve the same lift). If rotation speed was 65 kts, it would now be 67.5 kts. Measurable, but small difference. |> Similarly coming in too low and too slow with the nose down may result in an |> inability to lift the nose up with the elevator at maximum deflection. I'm not following how this would be possible. Assuming constant speed level flight (which for all practical purposes, a glide is), there are only two speeds possible for a given power setting, depending upon which side of the power curve you're on. Too low is bad, but too slow is only possible if your nose is already up and the elevator deflection is already high. This is, of course, bad, but that's the case with any airplane during a landing approach. One thing I'd like to make clear is that this canard shortening business only works in a small range. You obviously could not make the canard with only a one foot span and hope to have the aircraft fly, and neither could you extend it to 23 feet. The speed range of interest here is near stall, and since both wings (canard and main) have non-linear LC slopes with AOA near stall, the interaction between the two is complex (but definable). If you get either wing far away from this region, the analysis Mr. Venkatesh and I have presented here disintegrates (as would the plane) :-). I don't know about anyone else, but to me, this is a fascinating discussion. :-). -- Marc J. Zeitlin E-Mail: marcz@an.hp.com From: Lee Devlin Subject: Re: Short Canard Performance Date: Mon, 10 Apr 95 11:56:33 MDT All this talk about deep stall and the shortened canard length on performance is interesting, but I'm curious as to how easy it is to get the Cozy into an aft CG condition to begin with. My understanding is that the back seat is essentially on the CG so that loading that up doesn't move the CG, but would, of course, increase the gross weight and thus increase the stall speed. However, the front seat is in front of the CG and it's quite possible to have a wide range of weights in it. I can see where you might have a small pilot flying solo, or a pair of 190 lb people up there and have to handle this entire regime with the same canard. I don't know the specifics, but I understand that there is a ballast weight for the nose to compensate for the presence or absence of a front seat passenger. If a pilot were to be flying a passenger around and storing the ballast in the rear seat, then let the passenger out and forgot to move the ballast back to the nose, would this cause an aft CG? If so, this seems like something that's pretty easy to do and I don't think that adding a few kts. on the minimum landing/take off speed would be enough justification to avoid shortening the canard. Has Nat ever stated what happens to the take-off and landing rolls with the new canard or is he remaining quiet about it? Lee Devlin Date: Mon, 10 Apr 95 13:20:34 est From: "Phillip Johnson" Subject: RE: Short Canard Performance That was a fascinating discussion on canard performance. It is something that I have been addressing over the last couple of years. One of the weaknesses of the argument put forward is that there is constant reference to canard AOA changes as if the whole canard could be made to tilt with respect to the airframe. The discrepancy lies in that the canard is fixed and a slotted flap provides this function. (In fact the mechanism of the elevator is tantamount to a Fowler flap.) The CL curve of the canard changes with every new position of the elevator and I think the arguments given break down as a result of this oversimplification. My interest in the impact of front seat loading as a function of canard length/area. I found the simplest way to consider the effect was to consider the wing loadings on the wing and the canard and the respective weight and balance envelope. Since for the aircraft to fly with almost identical conditions, the wing loading at forward and aft C of G and at max and min weights, must be maintained. You can then plot new C of G windows as a function of canard length. This analysis indicates that the C of G window moves backwards as the canard is shortened. This is in line with Nat Puffers findings. I recall him saying somewhere that the original canard length was good for a front seat loadin excess of 400 lbs but he did now say what the min front seat load was, nor what the weight and balance window had been before he cut the 6 inches from the canard. I know that many people seem to want to operate with a heavier front seat capacity which would also give a commensurate forward shift in the C of G window. I have done the calculations but I am unhappy about publishing until they are tested. I would not want anyone to take my word as gospel only to find that a mistake had been made. Those builders who have already built their canard and have not cut off the mandatory 3 inches per side might do well waiting to the last minute (if they want a larger front seat capacity) before they do the amputation, there seem to be a number of people out there willing to give it a try and the results may be OK, after all it is easier to cut off 6 inches that add it again afterwards. Remember Nat flew for about a year with the longer canard. From: Marc J. Zeitlin Subject: RE: Short Canard Performance Date: Mon, 10 Apr 95 14:42:12 EDT Phillip J. writes: > That was a fascinating discussion on canard performance. It is > something that I have been addressing over the last couple of years. > One of the weaknesses of the argument put forward is that there is > constant reference to canard AOA changes as if the whole canard could > be made to tilt with respect to the airframe. The discrepancy lies in > that the canard is fixed and a slotted flap provides this function. > (In fact the mechanism of the elevator is tantamount to a Fowler > flap.) The CL curve of the canard changes with every new position of > the elevator and I think the arguments given break down as a result of > this oversimplification. I thought that I made it clear that I was talking about EFFECTIVE AOA, as set by the elevator (slotted flap). Obvoiously, the canard AOA can't change with respect to the aircraft, but the effective AOA can and does. As you stated, that's what changes the CL. Maybe I wasn't clear on that point. I think that our arguments are not orthogonal, and that there isn't an oversimplification to avoid. . [Good (alternate) analysis of CG range deleted.] . > I have done the calculations but I am unhappy about publishing until > they are tested. State that they are preliminary, and not for use. I, for one, would like to see them. > ............... I would not want anyone to take my word as gospel > only to find that a mistake had been made. Those builders who have > already built their canard and have not cut off the mandatory 3 inches > per side might do well waiting to the last minute (if they want a > larger front seat capacity) before they do the amputation, there seem > to be a number of people out there willing to give it a try and the > results may be OK, after all it is easier to cut off 6 inches that add > it again afterwards. Remember Nat flew for about a year with the > longer canard. Reasonable advice. We should also remember that Nat only had problems when loaded 2" AFT of the furthest aft recommended CG. With the long canard in the recommended CG range, there were no problems. On the other hand, I'm 150 lb., my wife is 115 lb, and I doubt my son will ever get over 150 lb, so I'm more interested in the following comments by Lee D. and the question of an aft CG: >However, the front seat is in front of the CG and it's quite possible >to have a wide range of weights in it. I can see where you might have >a small pilot flying solo, or a pair of 190 lb people up there and have >to handle this entire regime with the same canard. I don't know the >specifics, but I understand that there is a ballast weight for the nose >to compensate for the presence or absence of a front seat passenger. Look in your flight manual - Nat discusses using a ~40 lb. weight in the nose compartment when flying with less than 250 lb in the front seat. >................................. If >a pilot were to be flying a passenger around and storing the ballast in >the rear seat, then let the passenger out and forgot to move the ballast >back to the nose, would this cause an aft CG? Sure would. Even with the ballast in the FRONT seat, much less the back. >......................... If so, this seems like >something that's pretty easy to do and I don't think that adding a few >kts. on the minimum landing/take off speed would be enough >justification to avoid shortening the canard. I agree. Although taking 3" off each end isn't going to protect me if I try to take off with just me in the front seat and no ballast in the nose. I'll be WAY aft of the aft CG limit, and in deep trouble. I'm probably going to permanently mount my battery in the nose, to minimize this problem (as well as the one of the battery trying to take my [and anyone else's] head off if the plane comes to a sudden stop). >Has Nat ever stated what happens to the take-off and landing rolls with >the new canard or is he remaining quiet about it? I don't think he's said anything. The handwaving analysis I did seemed to indicate that there shouldn't be a big difference. -- Marc J. Zeitlin E-Mail: marcz@an.hp.com Date: Tue, 11 Apr 95 08:39:38 est From: "Phillip Johnson" Subject: Re[2]: Short Canard Performance Marc It sounds like I ruffled a few feathers, that was not my intention and I apologise. Clearly the canard is fixed and I understand that you were talking about effective AOA, my comments about being an oversimplification refer to the fact that, as the elevator is extended there is an effective increase in AOA but, unlike the equivalent, conceptual, tilting of the canard i.e. effective AOA increase, the CLmax also increases so the relative angle at which the stall occurs is changed and bears little correlation to the argument. Without the full suite of CL Vs alpha curves the limits of the canard cannot be ascertained, hence my comment. The whole concept of the increase in CLmax as a function of alpha, for the slotted flap / Fowler flap elevator, is the main reason why the Rutan derivatives can support such a wide range of C of G's. Again I apologise for any bad feeling, my intent was to promote rational argument and discussion from which we all learn. Phillip Johnson ______________________________ Reply Separator _________________________________ Subject: RE: Short Canard Performance Author: Marc J. Zeitlin at smtp_gway Date: 10/4/95 2:42 PM Phillip J. writes: > That was a fascinating discussion on canard performance. It is > something that I have been addressing over the last couple of years. > One of the weaknesses of the argument put forward is that there is > constant reference to canard AOA changes as if the whole canard could > be made to tilt with respect to the airframe. The discrepancy lies in > that the canard is fixed and a slotted flap provides this function. > (In fact the mechanism of the elevator is tantamount to a Fowler > flap.) The CL curve of the canard changes with every new position of > the elevator and I think the arguments given break down as a result of > this oversimplification. I thought that I made it clear that I was talking about EFFECTIVE AOA, as set by the elevator (slotted flap). Obviously, the canard AOA can't change with respect to the aircraft, but the effective AOA can and does. As you stated, that's what changes the CL. Maybe I wasn't clear on that point. I think that our arguments are not orthogonal, and that there isn't an oversimplification to avoid. Date: Tue, 11 Apr 95 13:17:33 EDT Subject: Re: Re[2]: Short Canard Performance Phillip; > It sounds like I ruffled a few feathers, that was not my intention and > I apologise. No apology necessary! No feathers were ruffled, and if I sounded like it, I need to tone down my responses. > ......... Clearly the canard is fixed and I understand that you > were talking about effective AOA, my comments about being an > oversimplification refer to the fact that, as the elevator is extended > there is an effective increase in AOA but, unlike the equivalent, > conceptual, tilting of the canard i.e. effective AOA increase, the > CLmax also increases so the relative angle at which the stall occurs > is changed and bears little correlation to the argument. I'll buy that. I guess that considering the audience in r.a.h. (mostly non-aeronautical technical) I didn't want to start trying to explain full aero theory in ascii text, but your points are well taken. > .......................... Without the > full suite of CL Vs alpha curves the limits of the canard cannot be > ascertained, hence my comment. The whole concept of the increase in > CLmax as a function of alpha, for the slotted flap / Fowler flap > elevator, is the main reason why the Rutan derivatives can support > such a wide range of C of G's. Also sounds reasonable. > ..................... Again I apologise for any bad feeling, > my intent was to promote rational argument and discussion from which > we all learn. I agree completely, and again, no apology is necessary. One of the problems with e-mail is that there's no facial expression or body language, and even with emoticons, it's hard to tell what's inside someone's head. We'll all slowly learn what pisses each other off as we go along. You were NOWHERE close to my hot buttons, and I hope I was nowhere near yours. -- Marc J. Zeitlin E-Mail: marcz@an.hp.com Date: Fri, 21 Apr 1995 15:16:42 -0400 From: JIMWHI@aol.com Subject: Short Canard Performance Effect of shorter canard on flying performance. There were several aspects of the recent postings on shortening the canard that need clarification. Just because the canard makes up approximately 10-12% of the lifting area does not mean it provides only 10-12% of the total lift. The canard has much higher wing loading than the main wing because it flies at a higher angle of attack (AOA). Without the lift curves for the Ronz airfoil with flap (ie elevator), it is difficult to predict the actual wing loading of the canard. Shortening the canard will significantly reduce front seat capacity and extend take-off distances of what I believe to be an already marginal acceptable takeoff roll. I hated shortening my canard by 6", and in retrospect I should have waited until all of the aerodynamics had been sorted out. The original Mark IV canard has the same amount of canard extending out of the fuselage as the Long-Ez does. I believe we are barking up the wrong tree when we focus on the canard as the culprit. My theory is that the main wing strakes my be the problem, not the canard. Shortening the canard fixes the stall "symptom", but does not address the root cause. THE STRAKE ON THE MARK IV INTERSECTS THE FUSELAGE 10" AHEAD OF WHERE IT INTERSECTS ON THE COZY III. Also, the Mark IV has a straight leading edge along the fuel strakes, whereas the Cozy III and Long-Ez have a jog. The intersection is further ahead and the jog is missing because the strake was widened to give more main wing area. As a result, the top of the strakes on the Mark IV is much flatter than either the Cozy III or the Long-Ez. Any look at a book on airfoil shapes will quickly show which wing surface would stall first. The rather flat surface on top of the strakes causes a rather abrubt stall, which may explain the results of Nat's flight tests. The strake modifications also moved the strake leading edge closer to the air flowing off of the canard. Which may explain why the flight tests deviated from the projected neutral stability points that Nat investigated in the newsletter prior to the deep stall tests. What are the possible fixes to all of this? I'm not sure. Am I qualified to make aerodynamic recommendations? NO, but here are a few possible fixes that I can think of. 1. Change the airfoil shape of the top surface of the strake, 2. Move the canard further forward, or 3. Lower the canard. No doubt, there are other possible solutions. What do you think? Jim White, JIMWHI@AOL.com Date: Fri, 21 Apr 95 16:52:08 est From: "Phillip Johnson" Subject: Re: Short Canard Performance Jim White Says: > Effect of shorter canard on flying performance. There were several > aspects of the recent postings on shortening the canard that need > clarification. I couldn't agree more. In my opinion you are correct in your assessment. An area for departure from your argument might be the analysis of the strake. > What are the possible fixes to all of this? I'm not sure. Am I > qualified to make aerodynamic recommendations? NO, but here are a > few possible fixes that I can think of. > 1. Change the airfoil shape of the top surface of the strake, The stake will surely stall abruptly but an aerofoil that is nominally flat will also have an aerodynamic centre at close to the 40% chord compared with classical aerofoil shapes. The classical aerofoil has an aerodynamic centre which lies typically at the 25% chord point. The flat strake thus causes the mean aerodynamic centre of the wing /strake combination to be at a station aft of the expected position for a true aerofoil, thus loading the canard more heavily than it would if the strake were a classical aerofoil (Loading the canard is good). Changing the aerofoil section for the strake may exacerbate the deep stall problem rather than relieve the problem. > 2. Move the canard further forward, or Moving the canard forward will increase the moment, which is equivalent to enlarging the canard not reducing the canard, thus exacerbating the deep stall problem. Further, the downwash from the canard reduces the true angle of attack for the strake thus making the strake less likely to stall even at high angles of attack. Moving the canard forward reduces the canard/wing coupling thereby making a wing stall more likely at high angles of attack. > 3. Lower the canard. Lowering the canard would put the wing out of the canard downwash with the same consequences as above. MY OWN THOUGHTS I really can't understand why Nat decided to cut off 6 inches. The long canard model that was flown for some time simply had a weight and balance window that was further forward than the short canard version. With the long canard, light pilots would be forced to carry more ballast, but heavy pilots would not overload the design. Those builders who typically flew with ballast would do better putting the battery up front instead of ballast. In this condition the aircraft would be able to stand on three wheels whilst being unattended. i.e. Far less chance of falling on it's back when empty WHICH IS GOOD. With more ballast up front (or more pilot weight) the aircraft would be more stable since the pitch moment of inertia increases with increasing mass up front. The real solution is to modify the strake but that is a complete development programme that I don't want to get into, at least not with my life. For once I agree with Puffer (Wow that was difficult to say). One solution that is worth considering is to extend the elevator deflection for take off. This causes early rotation whilst in ground effect. If you do this though you can get right back to square 1 with respect to deep stall if you use this at any time whilst in an aft C of G condition. A friend of mine has extended his travel and shortened the canard on his Vari-eze and he finds it works well particularly of a grass strip. (N.B. Profile drag also goes down with a shorter canard.) Apparently the original Long-eze's used to have more elevator deflection than they do now. The deflection was reduced to accommodate the same problems as the MK IV. There is more than one way to skin a cat!! Since the deep stall problem exists only in the aft C of G condition the aircraft could be designed with an extended elevator range that had two stops: 1) In a condition in which the aircraft is operating in the forward C of G condition full movement could be achieved. 2) In a condition in which the aircraft is operating in the aft C of G condition restricted movement could be implemented. Of course this same philosophy could be applied to the original long canard thus increasing the front seat max weight. i.e. implement stops when flying with an aft C of G. It is probable that you could do away with ballast completely if the stop was set as a function of C of G thus optimising the aircraft for all conditions but this requires a good pilot discipline to make the appropriate W&B calcs and adjust accordingly. Basically, like all aircraft, the MK IV has design limitations that we are all faced to live with. Remember most aircraft stall, and many are dangerous when they do. Fly a conventional aircraft behind the aft C of G and you WILL die, Fly a canard behind the C of G and, IF YOU FLY SLOWLY ENOUGH, only then will you die. Readers beware, these are my thoughts and are totally unproven. Phillip Johnson Date: Sun, 23 Apr 95 22:53:58 EDT Subject: Re: Short Canard Performance (fwd) People; > Jim White Says: > > > Effect of shorter canard on flying performance. There were several > > aspects of the recent postings on shortening the canard that need > > clarification. > > I couldn't agree more. In my opinion you are correct in your > assessment. An area for departure from your argument might be the > analysis of the strake. Long analysis by Jim W. and Phillip J. deleted for brevity. > Basically, like all aircraft, the MK IV has design limitations that we > are all faced to live with. Remember most aircraft stall, and many are > dangerous when they do. Fly a conventional aircraft behind the aft C > of G and you WILL die, Fly a canard behind the C of G and, IF YOU FLY > SLOWLY ENOUGH, only then will you die. > > Readers beware, these are my thoughts and are totally unproven. Well, I gotta say, you guys are good! I thought I remembered something about aircraft and wing theory, but apparently not enough. Excellent analysis, Phillip. B.T.W. - if you do a pure static moment diagram while in level flight, you get a Canard loading of anywhere between 18% of the total lift and 23% of the total lift (at aft and forward C.G. limits, respectively) assuming that the Aerodynamic Center of the Main wing is at station 120 in. (an educated guess). If the A.C.M.W. is further forward, than these #'s will be lower, and if it's further aft, they'll be higher. In any case, Jim W. was right, they're a lot higher than 10%-12%. This implies a Canard Lift Coefficient of ~2.4 and ~1.9 at 60 kts. at forward and aft C.G.'s, respectively, at 1600 lb. The main wing CL would be ~1.17 and ~1.25 in these cases. Anyway, thanks, guys. -- Marc J. Zeitlin E-Mail: marcz@an.hp.com Date: Mon, 17 Jul 1995 14:01:08 -0500 From: campbell@ee.umn.edu (faculty S. A. Campbell) Well, Nat called about 10 minutes after I posted the CFQ (call for questions) so I din't have anything to forward. I just got back and I could either - prepare the lecture that I am about to give to a bunch of grad students or I could type up this report and let you know what happened. I guess you know which I chose. To answer the big question - I did not get to fly 44CZ. Nat weighs 165 lbs and so we would be out of the envelope (410 lbs). He still considered going but Fleming field is rather short and is surrounded by homes and apartment buildings. I mentioned this and suggested that flying might not be a good idea. I did get to taxi to the other end of the field to get to the hangar. It's a strange experience not having a prop in front of you. I talked to Nat about the operating envelope with respect to the canard length. The numbers in the POH are correct for the shortened canard, but they are rather soft. For example the front seat weight limit is about 400 lbs (and should be independent of fuel content BTW), but Nat was quite vague on exactly how this was determined and in the end just said that it was a judgement call. Lessons learned for larger than normal cozy builders (I'm 6'4", 245): 1) For my use the canard should not be shortened the full 6". Nat recommended shortening by 4", since the additional 2" is a saftey margin which is only important when flying a full aft cg. 2) Nat suggested the use of an IO360 rather than the O360. More HP, but more importantly improved placement of the CG (for me). 3) Definitely increase the bubble height by 2". It closes on me as is but is quite tight with headphones on. 4) Look into replacing his version of the control quadrant. It was tight on my hip, squeezed between the two front seats. 5) Keep the nose light - no electric nose gear. 6) Continue to lose weight ;-). See you at Oshkosh, Steve Date: Mon, 17 Jul 1995 18:14:00 -0400 From: KSPREUER@aol.com Subject: Fwd: Nat on the Net In Phil's e-mail regarding Cozy canard span and C.G. and the related testing that Nat did he asked for the rest of us to comment on our attempts to communicate with Nat. I've been a Cozy bulider since 1984. I have flown the 3 place 7 years and am a year or so away from flying the Mark IV. I have written to Nat a number of times over the years and more often than not been put down, told that 1200 (possibly an exaggeration) other builders didn't have my problem, or told that I had insulted Nat. I have given up writting unless it is urgent. I don't have any problem with him in the group and I hope he adopts a very open mind (and a thick skin). I'm sure he will find much of it infuriating. On the subject of Canard span: I agree with the comment that Nat made in his relayed note. There is more going on here than the canard area alone can explain. I don't make any recomendation to leave the canard full span except to those who are willing to accept the risk of doing dangerous flight testing to be sure that all combinations of C.G. and weight are stall free. I personally intend to accept that risk. The reason I will do that is because I found that my Cozy 3 place had considerably diffgerent low speed characteristics than the one that Nat had tested and that many other builders seemed to have experienced. I never fully understood the reasons, but suspect that the difference was due to some bumps at the out board end of the strakes. I have heard of others with similar problems so my situation is not typical but not unique either. There were 3 incidents that on landing flare, at forward C.G. (but within envelope) that there was not enough control to flare. The first time resulted in a nose gear failure, the others in a go around. These flare attempts were at speeds of 75Kts + which is well above the tested minimum flight speed at those C.G.s. which would run around 65 to 68Kts. Takeoff is also an area of concern. On this point my experience is different than that indicated in Nat's relayed letter. I have found that rotation speed is highly dependant on C.G. and is generally higher than both minimum flight speed and minimum flare speed. This is because it is related to canard lift which is a function of both weight and C.G. position relative to main wheel position. It is our conclusion that our airplane also had a somewhat higher rotation speed than standard. For these reasons I want to be sure that my airplane behaves like a "standard" Mark IV before I cut the canard to the new "standard" length. I think that there needs to be a lot more published data on the effect the canard change has on takeoff and landing before I'm ready to cut off my canard prior to flight.