Aerodynamics & the "Dive Test"

[Erfolg]

I have never suggested terminal velocity as being necessary.

What I have pointed out that a model glider in a steep dive or even a shallow dive will reach a very high speed after a few seconds.

I am certain that many who have flown large open type gliders in the past will have observed models exhibiting flutter, whilst in dives, be they shallow or deep. With shallow dives the model will stay in the critical zone for longer, before either failing or moving out of the region.

Sometimes i do get the impression that there is a deliberate taking out of context, very real issues.

The sort of trimming that model gliders search for is very different to an IC model. There is no change of CG due to fuel changes. More importantly, the modeller is not looking for subtle changes of behaviour from a steady state condition.

[Biggles' Elder Brother - Moderator]

If it concerns you Erf - and for your information -

1. my comments about someone claiming that aircraft break up before reaching terninal velocity were not at all about you or what you said - my comments were referring to a completely different post! So no need for you to feel so sensitive!

2. Also, they were not "out of context" (as you imply) in any sense at all! To prove my point here is the entire paragraph quoted verbatim from the post in question,....

"You think? i wouldnt have thought so. I would expect most models to come apart before they reach their maximum speed. Although this is complicated by the question of power off vs power on and clearly the decent angle. if it was gravity only i would not expect to see them accelerate for too long, but with power on its a whole other story."

As you can see - no misrepresentation. The poster clearly states what I said was stated. So what's your problem?

BEB

Edited By Biggles' Elder Brother - Moderator on 09/04/2016 17:45:42

[Erfolg]

BEB

The problem with the dive procedure is that there are examples when it has nor worked.

There are also respected academics such as yourself who take a view that is not reliable.

Not all glider guiders see the process as the most appropriate way of achieving a CG position that is seen as optimum. On reflection, from casual comments, you realise that some glider guiders ballast up with a CG position, slightly further forward than when unballasted (or perhaps they did).

To some extent I am reticent to say this is the case today. Mainly as until recently I have seen thermal gliders that have a operating envelope far wider than i ever dreamed of. I have seen F£b models flown at speeds and pulling "gs" that i would have considered to be abuse.

At the end of the day, it may be one method, there are certainly others, also based on observation of a models behaviour. Non providing, a do this and success will result. Rather than seeing a simple fool proof single method, i believe that much of trimming is about observation and experience.

[PatMc]

Posted by Erfolg on 09/04/2016 18:19:41:

BEB

The problem with the dive procedure is that there are examples when it has nor worked.

There are also respected academics such as yourself who take a view that is not reliable.

In what way "not worked" ?

Where have any "respected academics" taken the view that it's not reliable ?

[Don Fry]

In my experience model aircraft are tough beasts. Sticking them into vertical dives, power on, is not a problem. They, if built as spec, don't break easily.

BUT, I have followed this post with great interest, a great deal of evil amusement as to the passions involved.

[Biggles' Elder Brother - Moderator]

Ah, OK I see Erf.

Well first of all I would say we have a lot in common in our views - but I would express it very differently!

Firstly - the dive test always works - by definition. Leaving aside for a moment the fact that it is possible to do the test incorrectly - lets assume that is not the case - then the "faults", such as they are, do not lie with the test in my view, but in how people interpret what happens.

I tried to explain this several pages ago - it doesn't seem to have worked! I'll try again in a slightly different way!

What does the test actually do? It reveals the pitch interaction between the flying surfaces of the model - in particular the relationship between the incidence angles of the main wing and the stabiliser.

There will, as has been said, be secondary effects from things such as' the tail moment (which changes the leverage of the stabiliser and hence its degree of influence) and the wing section (which determines the position of the centre of lift etc). But unless we can change these things on the model (not very likely!) these secondary effects are only really important if we are comparing one model with another. So for 99% of applications of the test we can ignore this - because it is not a variable we can change. OK we need to keep it in mind in interpreting the results - e.g. if the model has a long tail moment we have to bear in mind that if we are adjusting the stabiliser incidence such an aircraft will be more sensitive to small changes than a aircraft with a much shorter tail moment. But other than in this way we can put these factors to one side.

So - we arrive at a point that says the dive test examines the interaction between the main wing and the stabiliser; and that their relative incidences are our main variables. So, what's this got to do with the CoG - which is where we sort of come in? And is what most folks believe the dive test is actually about - form what I've just said its not about that at all!

Remember - it is about pitch behaviour, determined by the interaction between the lift forces from the stabiliser and the main wing. But where will this pitching take place about? Answer - the CoG! All the rotation in pitch will happen about the point which is the CoG. We spoke of the tail moment - the point from which the tail moment is measured is the CoG. The stabiliser has to work against the (almost certainly) nose down lift moment created by the main wing wing - the size of that moment is determined by the distance to the CoG. This is how the CoG location gets into the dive test!

So what can the dive test tell us? It can tell us, for a given set of tailplane and main-wing incidences, what the CoG we have will give us in terms of stability, neutral behaviour, or instability. I stress this is only for a given set of incidences. Change the incidences and the the whole situation changes.

Now, if we go all the way back to the OP - he referred to my article in the mag. What was I trying to do there? Well, I wasn't trying to coach people in how to set up an ultra high-performance glider, that's for sure! I was trying to give people a test they could practically use as a guide. You see they, like me, are club flyers. The articles were about aspiring to be a bit better - "a bit" - not world champion! So in practise people like us are not going to fine tune the tailplane and main-wing incidence - we are just going to accept it at the value it is. So what the test does is help us basically get the CoG right-ish for what the incidences actually are. It enables an average flyer to set up a model so it is stable - but is perhaps closer to neutral stability than they would get otherwise. I totally agree that neutral stability isn't actually that nice a set up - its also a singularity so is almost impossible to achieve, in practice any model will always be either marginally stable or marginally unstable!

No, the dive test is not perfect. In reality it is very complex and involves intricate interactions between subtle forces. BUT,...in prcatice it is actually a pretty good guide to CoG position for a "close to neutral but actually stable" set up for a given, fixed, rigging arragement of incidences!

BEB

Edited By Biggles' Elder Brother - Moderator on 09/04/2016 20:22:17

[Don Fry]

BEB mopping sweat off brow.

[PatMc]

BEB, surely the rigging is not fixed?
The elevators (or tailplane in the case of AMT) are trimmed up or down according to whether the results from the dive test indicate moving the cg forward or back.

[Biggles' Elder Brother - Moderator]

Yes Pat of course you can change the elevator trim - but in powered models how many owould actually change the tailplane incidence? Few I suspect. The situtaion may well be different among serious "glider guiders" who, in my experience, are altogether a more scientifically minded and adventurous lot!

BEB

[PatMc]

BEB, I'm usually perfectly happy to fly with the elevators permanently set to suit any change I make to the cg as a result of dive testing. But I consider changing this elevator trimming in itself as a change in the rigging & a necessary part of the dive test procedure.
BTW I was & am refering to sports type models though I see no reason to treat competition models any different be the powered or glider.

[Biggles' Elder Brother - Moderator]

Yes It is a change in rigging effectively - I agree. But I do think you're probably more conscientious and particular than most sports flyers! Not a bad thing of course!

BEB

[Erfolg]

Patmac, at present i will be moving house next week end.

I have packed much all my books, magazines and report type papers, these are all or mostly at the new home. I have no sensible access until later, to find the report. I am not sure if it is a BATS paper, possibly for many of glider guiders a highly regarded source of data, or may be a Simon Cocker article, who has written for many magazines in the past as a technical contributor, rather than what appears now to be general reports on gliding.

Certainly in 70-80s the most prevalent competition set up would use all flying tails. Principally as the tail provides a downforce to resist the turning/pitching moment from the wing. As cambered sections tend to have higher drag than flat plate sections. Although at near to the stall, it is the vortex drag from the wing tips that is the greatest source, although reducing with increased speed.

I can see that diving a model has the appeal of being a simple process. On the other hand, without a clear understanding what is seen or measured as acceptable or desirable, is the process foolproof.

Although a 30 degree dive sounds begin and a 45 degree looks a lot more challenging, will either do? I have seen the old Carl Goldberg Gentle Lady disintegrate, just being pushed hard. This was when the single spar was located on the Neutral Axis or pretty near. I have seen others, some 20-30 years on, moulded gliders are both strong and light, and can take abuse, which will take such abuse without apparent issue. At the same time many present electric gliders are not much if any better than the original GL, not requiring the spar strength of a tow line model, often having a plastic film covering (not good for torsion resistance).

As with most things with memory i cannot remember all the details or even the bulk of them. Although I have an impression that the concerns revolved about the safety issues and what was actually being measured.

For my part with my interest now in electric power models, using on line CG calculators work extremely well, seldom requiring anything doing, other than sometimes moving the CG forward a little, to make the pitch sensitivity similar to the aileron performance. In the past i used the traditional numeric methods that took into consideration downwash from the wing and the static margin required. As the starting point.

What i have been told and observed is that many of the glider guiders, would trim their models by observation, with particular emphasis on the damping of a model after disturbance.

For me the seminal moment was the 2m experience with a very far forward CG. Having the situation explained to me that a very far forward CG does not self damp. The model continues doing what its latest equilibrium position is, until the system is disturbed, or the parameters change sufficiently. This is a extreme stability. Certainly there was no tuck under, nor recovery, the ground was just rushing up. Until my input snapped the model into a new stable condition, until that became unviable.

So in short with a power model, i am not to bothered, it is as much lack of sensitivity in pitch, setting the throws etc, to obtain a reasonable similarity of control inputs, for all surfaces.

For gliders, I no longer have any comparative models, although i would still be looking for a stable hands of glide, with good inherent recovery stability.

I will be looking for the more informed piece of literature, although it did convince me that the dive test, needed caution. In my opinion, you still need to know what you are wanting and the experience or guidance with respect what to look for, and guidance as to what is actually happening.

[Don Fry]

Sorry Erfolg, but, using any CG calculator, they are all conservative, they always give a forward position. They are safe. But to get a nice flyer, you move the position back. That's what these assessments are for. (Dive and watch, roll inverted, how much elevator, and how stable is a low 45 degree banked low pass). When right, it is right ( for a sports pilot), but forward from a calculated position, never.

[Erfolg]

Donald

I am not certain at all that anyone can be that categoric, that CG calculators are always conservative. Yet in principle that is what I would expect. In principle i expect that the formula and the options selected are designed to err on the conservative side. Although I expect that most are just a spreadshhet using the relationships found in all the older text books and Martin Simons book, with a bit of simplification, principally the effects of downwash on the elevator

Yet at the end of the day I do think you have a point, in that the calculator provides a starting point.

At present I also do roll inverted, and anticipate that only a slight amount of forward pressure will be required for level flight.. I have tried the dive test and have remained unconvinced that it does anything with a power model.

As for the rest, not something i do. For power type models, with the type of flying i do, I am only interested in a model that has a degree of stick uniformity and a predicable flying model. In general I find that most sports models just fly ok, at the 30 percent CG.

It is perhaps the often dismissed thermal glider that is for me far more demanding. In the past i would mark the Tx with three points, max duration, cruise and penetration. More recently i started using the programming for a similar effect. A mix of elevator with throttle to obtain a controllable approx 45 degree climb via a mix. Again using the mix for the three off power conditions via a three way switch. The big difference in my opinion, is that a lot of time a glider is taking care of itself. It is the other remaining time, that the model can be expected to reliably to be at the extremities of the flight envelope.

I am getting the impression, that many of us have our own ways of establishing a working CG position. I now rely on CG calculators.Others seem to find the dive test works. For my power models, that generally seems good enough, in most cases. Yet I and i suspect others then have their own ways of optimising the set up, if necessary.

[Dave Hopkin]

And here was me, thinking I was asking a simple little question..................

[Steve Hargreaves - Moderator]

Surely what matters most is that the model responds in a way that the pilot wants it to. We all have different "tastes" & one man's "touchy beast" will be another man's "responsive aeroplane".

Personally I trim the model so it responds as I like it to & then do a quick dive test to see where abouts on the scale the CoG has ended up......mainly for academic interest but partly to ensure I haven't strayed too near the bit of the performance envelope marked "Instability". I do this so I can be sure the model won't bite back in extremis.

[john stones 1 - Moderator]

The question was simple....the answers complicated it

John

[PatMc]

Remember the dive test isn't tool for predicting the cg. It's a means of tuning to the cg position so that the model handles the way you want it to - "handles the way you want it to" being the operative phrase
Different folk might prefer different cg position on identical models.
And an individual might prefer different levels of stability on different models.

[Masher]

H

Edited By Masher on 11/04/2016 06:56:35

[Bob Cotsford]

It's also a tool that can be used for other purposes, eg for the F3A fraternity, once they have established the balance point using other methods, they can use the dive test to set the decalage for minimum trim change with speed without the motor thrustline affecting the result.

[Don Fry]

Erfolg, I think I can be that categoric. They must be conservative. Imagine a maiden flight, controls not balanced, and out of trim, with a rearward CG. ONE FLIGHT ONE AIRCRAFT TIME.

And a disgruntled flyer hunting the publisher of the calculator, seeking to find if more pleasure is gained by slow or quick decapitation.

[Erfolg]

Donald

It may be pedantic, as we users have no knowledge what lies behind the front page we cannot be certain. I just accept that they are pretty good,.

Although the canard calculator did not provide a spot on CG, as posted in one of my threads, being rearward, providing a few anxious minutes to both me and the pilot before it regained the field. What is perhaps interesting, that CG calculator placed the CG forward of my calculated position. On that basis you would argue that it was conservative.

As a non believer, part of the issue is a good definition of what it is you should be not only doing, also the manner in which the model recovers.

Making the assumption that the model is in the correct angled dive, it has been held there until whatever criteria is used, being time, distance, or maybe speed. The stick is allowed to self centre, and the model starts to recover. Some guidance on the recovery rate anticipated would help some. Then what happens next, does the model return to level flight, without any overshoot of the steady state path, may be one oscillation or a number as in a damped porpoise?

For a system advocated by many, there is not a lot of real guidance or agreement.

I guess it is not a perfect system, that works for some Although without guidance, it is probably experience that really matters.

I suggest that some consideration of the start set up is worth considering, well before any practical flight testing. In my case I always set my clevis on the outer hole of the servo arm, using the standard travel. Partly because i use budget servos, where centring is probably not as repeatable, or travel as accurate as say a good digital. I also like to see typically +/- 10 degree of surface movement.

In general as others have suggested, I am looking for a well coordinated, reliable set up. I personally rely on what experience, advice I have available to me. Now I increasingly use a skilled flyer to act as a test flyer. A good chat, to review the flight after touchdown.

I am suggesting that other methods can work, for those who are not convinced with a dive procedure.