How does it work?
|John Cole||17/06/2009 17:50:21|
|615 forum posts|
One of my models, trimmed to fly straight and level on half-throttle, glides at a very steep angle when power-off. An experienced modeller recommended several changes so that, when trimmed as above, the glide will be much shallower (all of which I've made). One of them was to increase the downthrust considerably. When I asked him about that change and why it would help he gave me an explanation, but I'm not sure he's right (in his explanation, I mean - I'm not disagreeing that downthrust helps).
So if anyone out there can tell me WHY downthrust helps in this situation I would be grateful for your comments.
In case it affects the answer, here's the model configuration: high-wing electric glider with motor mounted behind and slightly above the wing centre-section (a pusher).
|Tim Mackey||17/06/2009 20:00:42|
20920 forum posts
If you think about it logically, if, under "power off" scenario the model tends to dive quite steeply as you describe, then it may be simply nose heavy. Upon having power added, the model pulls itself "up" by virtue of their being a little up thrust in the motor thrust line. Tpo compenastae for this climbing action, you have trimmed in some "down " elevator. So, in reality, you have compensated for a forward COG by applying down elevator...which is kinda backwards IYKWIM. but does allow the model under power, to fly staright and level.
Most models ( and full size ) will climb under power - thats what they are supposed to do, but a well balanced and properly trimmed model should glide fairly flat under power off conditions. I would start by looking at your COG.
Get that sorted to enable a decent power off glide, and then see what happens under applying power - if it climbs too much then apply more down thrust, but the fact that the motor is ona pod and a pusher, may require some lateral thinking about what constitutes downthrust in that configuration. I will now go and have a lay down and let someone better at aerodynamics and trimming answer the rest, and probably blow my thinking apart too
|Stephen Grigg||18/06/2009 00:09:33|
8691 forum posts
|Phil Wood.||18/06/2009 00:27:35|
3638 forum posts
Got to admit Tim, when you stick the thrust source behind and above the COG you open up a can of worms.........I can work it out because I did thrust vectors as part of my physics but it's not a simple subject to explain.
I'm standing back .....
|Gemma Jane||18/06/2009 01:05:30|
1349 forum posts
The last time I tried to explain something to Mr. Cole he let me know that he already knew the answer to his question and I was wrong Stephen and then proceeded to display to me a complete lack of understanding of the subject matter.
I'm hardly going to go into thrust/drag and lift/weight couples as I suspect it is the same sort of question. He has already stated that he has had the explanation but believes it to be wrong.... so he already knows best.
Tim pretty much has it summed up any case IMHO, it is all you need to know to get it right rather than argue the physics.
I don't see either really why one cannot trim for the two different flight conditions.
Trim for straight and level at a given power setting, trim for glide.
It is how full size are flown.
If I get a dead stick the first thing I will do is get the nose down and trim the model to glide at the angle I want it to, then I concentrate on bringing it in - exactly the same routine as when my instructor grins and shuts the throttle just after take-off to see what I will do.... it goes quiet, trim trim watch the airspeed trim! Fly the aeroplane, watching the airspeed then do any of the other stuff, but trim for best glide speed and fly first.
On any given day the exact amount of trim needed for straight and level will vary with numerous factors. So trim it as you fly. I don't see this much from modellers generally, some get it set-up just fine and don't need to worry, but if you want to do a low pass it's much easier with the model trimmed correctly than if it is with one trying to climb and one is putting in elevator inputs near the ground to keep the nose down. It's common sense to trim as required to make models fly how you wish.
It's fair enough to say a model should climb with full power and nose down into a reasonable glide with the power removed though and that has been covered by Tim.
|Phil Wood.||18/06/2009 01:10:43|
3638 forum posts
Putting it another way.......if you don't trim it to glide then you're picking up the bits if you get a dead stick. ............would that be fair to say Gemma?
|Gemma Jane||18/06/2009 01:22:18|
1349 forum posts
One could put in a billion control inputs and keep it flying if one wanted to Phil, but I would trim it out so it will glide in practically on it's own. Then I'm just looking at shallow turns and judging drift to get it back down, I don't miss often either (I don't get real dead sticks often, just practice them a lot)
From the full size perspective, it's better to hit a tree flying than it is to hit the ground from a stall, the first case one can get away with it, the second one can't.. it applies to models too, the way to avoid the situation is to get it trimmed for the glide first, then one can guide it in without constant pitch adjustments and the worry of a stall.
|Doug Ireland||18/06/2009 01:22:18|
2088 forum posts
Engine failure in flight? Swap speed for height and trim for best glide. Engine failure after take -off, trim for best glide and don't look any further than 30 degrees either side for a landing area. Thats what I used to say as an AFI anyway!
|Phil Wood.||18/06/2009 01:50:00|
3638 forum posts
Ok John, try to look at it like this.....if we built a very aerobatic model that flew perfectly inverted then we would have what is known as a "Zero rigged" model........the problem with this type of model is that it has no lift until you put in some elevator........if you put it into a dive and let go of the sticks it will carry on towards the ground until you pull it out.
Most of the models we fly have a little bit of "up" elevator or a slightly positive wing attack angle so we can fly, "hands off".
You set the model up so that it glides well with just a tiny bit of up elevator and no power.
Now if you put on the power and it climbs too fast then you add down thrust to compensate and pull the model down.
If it pulls to the side then you add opposite side thrust to keep it straight.
With care and attention you can get the model to just climb gently under power but also glide perfectly when the power dies..............It is down to the C.O.G.
Set it up for the glide first. ....alter the thrust later.
|Gemma Jane||18/06/2009 03:15:24|
1349 forum posts
Agreed there Phil, the issue here is the model wasn't trimmed for glide in the first place, the first step.
I expect though the whole issue will be complicated soon by statements such as high thrust line configurations (as described) that are above the CofG pitch down on application of power and up when it is removed. I best leave it alone Of course it doesn't matter, all that matters is that the model wasn't set up to glide correctly in the first place. Once it is then one can play with the thrust line as both you and Tim have stated.
|John Cole||18/06/2009 10:17:59|
|615 forum posts|
Hi, Gemma. Nice to hear from you again.
Timbo: yes, the CoG may have been a bit forward: spot on 25% as per the plan, with a cambered-centreline wing section (as you would expect for a powered glider). I've moved it back to 30%. You say the model flew S&L under power "because of the upthrust". Sorry, I did not make clear but the starting point was a couple of degrees of downthrust, which my friend advised me to increase.
So, yes I've taken the CoG back and trimmed first for the glide - and I've put in quite a bit of downthrust and that seems to have reduced the power-on / power-off trim difference. But I'm none the wiser on why the downthrust has made a difference (which I think it has).
Phil: I wondered the same as you, whether the location of the motor made a difference, but it seems to me that the fact that the thrustline is above the centre of (parasitic) drag would give a nose-down couple under power, but that would give a result exactly OPPOSITE to what I'm seeing. And I can'tsee that the effect of downthrust links to that couple either. But if you do then I would be most interested in your explanation (pm me if you wish).
So, thatnl to you all for the advice on trimming - but I'm still looking for comments on why downthrust helps!
|ken anderson.||18/06/2009 10:42:26|
8636 forum posts
sorry you're all missing the point-downthrust is pronounced "doonthrust"....in this neck of the wood's............
ken anderson...north east plc........'s.....
|John Cole||01/07/2009 17:23:17|
|615 forum posts|
Well, here's what I've found: the down-component of the thrust on a pusher model is behind the CoG and so might be expected to push the nose up under power. But it doesn't, at least on my model (with about 7 degrees of downthrust, more than you would usually find in a tractor model, I think). In flight tests, the downthrust keeps the nose down under high power settings. But why?
I did a static test, marking a vertical line on the fuse side through the CoG, and marking a line on the fuse side representing the thrust line. I then stuck pins in where the lines intersected, and balanced the model on those pins. Static thrust from the pusher prop should then not pitch the model up or down
But it does! It goes nose-down!
I think what's happening in the static test is that the airflow over the horizontal stabiliser pitches the plane. With no downthrust (and a quite-forward CoG, my starting point), the stabiliser is at negative incidence to the propwash and the plane pitches up (which is what I observed initially). With substantial downthrust the stabiliser is at POSITIVE incidence to the propwash, and so the tail is pushed up (and nose down). And the flight test suggests this has a bigger effect than the nose-up moment of the behind-the-CoG downward component of the thrust.
I guess the same is true to an extent with a tractor config - with the two moments acting in the same sense, but with the propwash element weakened by greater prop-stabiliser separation and perhaps a flow-straightening effect from the wing
2083 forum posts
With what appears to be your set-up imagine,,,,,
Get a long handle floorbrush, hold in the centre of the shaft with the brush hanging. Now, Push the TOP af the handle imagining the push is coming from the engine, the "nose" on the opposite side from the thrust points DOWN. The wing tries to lift, however too much DOWNTHRUST makes it impossible.
So as said, ballance for glide first, then adjust engine thrust line to give a level flight at about half throttle. Higher engine speed it should climb, lower speed it should,,,, well you should have it by now. Use trim to suit what you are doing with the model, only after it had been "sorted"
11713 forum posts
Tut, tut. Not getting into the physics?
Why! Isnt that a really good starting point in understanding what is possibly happning. It also helps in guessing the direction of the solutions.
I do agree that a pylon mounted motor, will require more thought. But why! Because the thrust force is so far from the point which can be considered as the sum of all the forces. so the up thrust required could be significant!
You are all so right in trimming for the power off situation, excellent advice.
So why will no one shed some light on the possible issues of my flying wing (2.4 Futaba, Ohhhhh Nooo Not again).
Gemma, tut tut again, physics is great fun, little like suduko.
|Gemma Jane||02/07/2009 13:35:52|
1349 forum posts
Not really Erfolg the correct answer lies in a rather complex analysis of the static stability of the configuration which includes the downwash affects of the main plane. It's a approximately three university terms worth of mathematics.
Having attempted to discuss similar issues with Mr Cole in the past I have no stomach for it. As you can see he has already 'solved' the problem. Not. Though you may notice that I had stated one would expect nose down pitching with this configuration. It should be no surprise at all!
11713 forum posts
Done that, got the tee shirt as they say. Graduated with honours.
I tend to think that an absolute answer is not required in modelling, partly because we estimate so many of the forces. Yet a FBD (free body diagram or a sketch if you like) can give an insight to what are the principal forces and how they are probably interacting
But it is fun. One of the many things I have learnt about modelling, that defnative mathamatical calculations, seldom match with practice and that small changes in value can result in a vastly differing results. With the advent of computers (IBM PS2), I once did some bending moment and torsion calcualtions (for a model), where all the parrameters, were incrementally changed, within the believed viable interval. As suspected the plot of the values was significant. The excersice was not really about the results, it was about programming, and generating graphical output from the sums.
Still think that the theory and concepts are useful, in that it helps us get and keep our models flying. Most, if not all modellers, can easily understand the ideas, the main issue is often they suspect they can not
I am sceptical as you can gather regarding the value of complex mathamatics for our models. Differential calculus has little space in my modeling.
As for not being able to see or understand information given to me, I to am and are guilty. When I was told to just accept, if you want to pass your exams, I found it difficult. I did not and still do not really believe in "circulation theory" and do not really understand, Von karmen vortex shedding (other than lab observations matched well with calculated values). I keep on asking though, and tring to understand.
11713 forum posts
If I understand you correctly, I would expect a nose down couple if the motor is mounted above the CG on a pylon, with the thrust axis along the zero AOA for the section. I would therefore expect a lot of upthrust, to obtain a relatively neutral responce under power.
If the model however is a shoulder wing, such as some of the Graupner Foamies. With the motor behind the CG, could well be that downthrust would be required.
I hope I am not complicating things.
|John Cole||02/07/2009 19:31:32|
|615 forum posts|
Erfolg: thanks for your comments. I agree that a high-mounted motor will generate a nose-down static-thrust pitching moment (though I think the issue is whether it's above the centre of drag rather than the CoG). But what I found with my plane is that increasing power made it pitch UP and climb steeply. And that's what started this thread.
But my question was really about how to solve the problem, and WHY downthrust seems to help. I am by no means sure that the explanation that I've offered (propwash) is correct, but given the results of my trials I can think of no other. And it was to get other people's ideas that I started the thread.
I'm not looking for help on doing the sums, but I am looking for people to give me ideas so I can do the sums myself. Like you, I've quite a strong background in Maths; it was my job before I retired.
Anyway, thanks to all all for the point that I should initially have trimmed for the glide, and then sorted the power-on issues. Thanks Gemma for the point of needing different trim for those two conditions. As well as partially resolving the problem with downthrust (much to my surprise), I also mixed power with elevator, so increased power adds down- elevator (I used a 6% mix) and my plane now flies nicely.
Now for the article in this month's RCM&E about stalling! Gemma: you might like to kick that off.
11713 forum posts
You are correct in that my use of termology or jargon, is loose. It is easy to get bogged down with terms. The days have now gone, where I and my colleuges would converse in this type of manner, high alpha, delta t, G (as in Gibbs) etc. It works where all talk the same language, but us modellers are very different.
Back to your issues, from the point of where the aircraft can be considered to pivot, on the longitudinal axis, the CG is near enough, for our practical purposes. Now considering rotation around this point as x,y co-ordinates. There is a slight problem, in that we do not know where along the y axis, the model could be considered to pivot, either staticly or from a forces perspective (it is the forces we are really interested in.
If we were to consider a model, real or diagram, where this point was co-incident with all the drag type forces and the thrust from the engine. Then no down force would be required. All would be happiness if the drag forces, just increased with speed. Yet if the motor thrust were above or below this point, then a couple would be created. Causing the model to try and pivot up or down. A simple solution would be to angle the motor thrust from along the y axis, to generate some force in the y axis (this would be drawn as a force vector). Now everything is in balance.
Ohhhhh, if the real world were that simple.
The forces causing drag, are not all along the x axis. They to can be represented by vectors. The additional trouble is that these foce vector will vary with velocity (downwash etc). The most annoying are those varying in the y axis.
To further make things arckward, we do not know what is the point that these forces can be considered to be acting, at any speed, nor do we modellers no there value, or often there relative value. The position of the motop will in all certainty not be in line with this point either.
So it is easy to throw our hands in air and say we cannot solve this problem. Yet many models, fly most satisfactorily.
A simple step towards, solving the problem is to grab a peice of paper, produce a rough otline of a model. then superimpose, what you know, the drag from the wings, the relative position of the drag from the tailplane, undercarriage etc. Draw on the thrust from the motor.
From this rough sketch you will know which rule of thumb to adopt for down or upthrust, which have been developed by modellers.
With regard to your model, if after trimming for the glide, it is found to climb faster than you want under power. Just reduce the thrust line.
I think Gemma is hinting at a number of issues, the wing should be model designers witlldraw the CL on the board/CAD system, then arrange the wing section relative to this to achieve the sit of the model in the air (for want of a better term). Then the engine etc. This process is well understood by model designers. But from a mathamatical perspective, the forces acting on this thing are changing with velocity (speed and direction) and we have not considered inertia, centrifical forces all in the x,y,z axis.
I believe in keeping the model (theroretical that is ) simple, but relevant. Then adjust the model when built to a satisfactory state. None of this is acceptable in full size aircraft, you need to know before flying. Hence Gemmas very differant view, she is professional in her approach. I see it as a hobby these days. When I get out the calculator/spreadsheet it is because I am interested in the maths, not the model. I do use other peoples spread sheets though (although none have been verified as far as I know0, there are a lot of gifted theoretical modellers out there.
I am waiting for a hobbiest FDM (fluid dynamic modelling) system to be made available to us hobbiests. Again caution is needed, lack of intelligent use of programmes has been responsible for some big cock ups.
Please login to post a reply.
Want the latest issue of RCM&E? Use our magazine locator link to find your nearest stockist!