wizzards wrong Kanäls and wrong direktions!

[Helle]

Hy,

after testing the Wizzards

they are definitv wrong
in Chanals and in directions

its a mix of right thinking and wrong doing

1.
World Standard is:
A Stick goes to positiv ( +100%) then the movement must be to up or to right side
or
A Stick goes to negativ (-100%) the the movement must be down or to left side

but this is first the reaction for the math/logic and at the end of the movement
but nor for the servo itself! Servomovement is depending of the mounting!

2.
Starting of Ail channel is:
rigth Ail is Nr. one because up is +100%
left Ail is Nr. two and at same Time down at -100%

3.
you have a logic/math direction for mixing and logic the Ail math correct
and you have a other Servo/mechanik directioon to adapt the real world
that are two separete things dont mix them at the Kanals but you to it indirect

4.
look at the aileron:
took Ail Stick to the rigth and hold it:

you will get +100% to the right Kanal (for math/logic)
and at same time you will get -100% to the left Kanal (for math/logic)
it must be independent from the Servo direction!
its for correct calkulation for mixers later

so you have to set

CH2 Ail1 +100 %
CH5 Ail 2 -100%

and what you get is both wrong Ail 1 left +100% Ail 2 right +100%
(I know why you are doning that becaue of the Servo-movement left/right side)
(so if you dont more mixing to ail its correct and you have to reverse no or both servos)

But:
Now mix flaps down with -25% to both Ailerons
Correct math /logic is
CH2 Ail one right side +100% (fromStick) and -25% from Flap
Ch5 Ail two left side -100% (from Stick) and -25% from Flap


but what will you get:
Ch2 Ail one, left side +100% and -25% from Flap
CH5 Ail two, right side +100% and -25% from Flap

and so there its the big error at all of your Wizzard!
You start with wrong sides ail one in left
you mix math/log and Servo both +100% and wrong together!
and so You get wrong math in mixing


First Start with
Ail1 CH2 right side set it to +100%
Ail2 CH5 left side set it to -100%

and all you added mixers to Ail will work math/logic fine

and then after alle mthy/logic mixing is done, then

set your Servo so that the movement are corrects


But now all the wizzars are wrong!
Start with wrong side
Start with wrong both Ail +direction

Sorry my english is horrible!
Helle

[Kilrah]

Sorry my english is horrible!

Unfortunately...

Sorry, don't understand anything.

The wizard does not do any mixing, at this point it only handles simple scenarios where each function uses separate servos. It is still "beta".

[Helle]

Hy Kilrah;

1.
when you use a normal Modell with 2 Ail no other things, no mix
and you take the Ail-Stick to the right, the stick give you +100%
then at the Modell
the right Ail must go up +100% (go to positiv direction)
and at same time
the left Ail must go down -100% (go to negative direction)

for the math and for the logic
you have to set at the programming
first Ail right side weight +100%
second Ail left side weight -100%

At the servo limits you have to set the servo into the "right" direktion
so that Ail right goes up and at same time the Ail left goes down
and then maths and real movement is ok.


if you then set a Mixer, for example Ail as Flaps with -25%

you set
the mixer
at ail right side +100% Ail += Mixer Flap -25%
ar ail left side -100% Ail += Mixer Flap -25%

and you allways got the right calculations!!!


But your standard simple Wizzard generates
Ail riight +100%
Ail left +100%

and then You got in Trubble when you will add a Mixer
check it you will see!


2.
And at the Wizzard Picture and the colors of the Modell chanels

you see that the first Ail ist at left side
and the second Ail is at right side

Standard is:
first Ail right side
second Ail at left side

hope something will be more clear


3.Standard is:
positive direktion is movement up or to the right side (Ruder)
negativ direction is movement down or to the left side (Ruder)


these are the 3 things to do and alle math-calculatios at Programming are allways correct!!

Servo movment is independent from intern math calculation!
its only for adapt the mechanic to the real world, not more!

Helle

[Kilrah]

Servo movment is independent from intern math calculation!
its only for adapt the mechanic to the real world, not more!

Yes I know, I wrote just that in the manual...

So basically the only thing you would want changed is that "AIL2" mixers created by the wizard are set to -100% instead of 100%? Makes sense indeed, but it was hard to understand from those big posts that this is what you wanted.

[Helle]

Hy Kilrah,

we need two things:

1.
Wizzards colurs the Pictures and Channel
first Aileron at right side
second Aileron left side

2.
first Ail at right side with weight +100%
second Ail at left side with weight -100%


compare with Graupner, Futabe, Multiplex, Hitec and all others

there is a "Standard", first Ail reight side, second Ail left side

Graupner:
Ch2 Ail1 right side
Ch5 Ail2 left side


Futabe :
Chx Ail1 right side
Chx+5 Ail2 left siide

and so on

By the way:
Yes, i know some Tx have to set on both Ail +100% ,
but then if they have to use their Flapmixer, there is intern and secret a changing of the directions!
but no one see that!)

Here at openTx we have a "open mixer structur" and we have to calculate all time correct!

Helle

[jhsa]

both aileron channels have to output the same value because normally the servos on the wing are mounted complementary to each other.. that means facing away from each other for example.. therefore they are both positive or negative depending on the direction.. And for me that is the correct way to do it. But each one to his own
As for flaps with the same setup, as both surfaces go down, the servos rotate in diferent directions. So, one must be reversed.. For me that is the logical way of doing it.. Thank God for these firmwares with a lot of freedom, as I don't have to do it the way that I think it is wrong

João

[Helle]

Hy jhsa,

Yes jhsa, you all have the problem to mix in your brain the math funktion with the Servo funktion in one togehter
thats wrong and you get wrong in math calculations.

I know that normal each servos at Ail are mount to complemetary, (normaly but not ever! )
but so this is your doubel-inversion and so you set both to Ail +100%
and get one in up and the other in down because one Servo invers itself!

So you have in the real world to invers no Servo or both Servos to ge the correct movements

------------------------------------------------------------
look at the mixer picture I added.

Standard for the maths is:
positiv is up, negativ is down
positiv is to right, negativ ist to left
Ail1 is at right side
Ail2 is at left side

so my Channel setting is:
Ch2 Ail1 right weight +100%
Ch5 Ail2 left weight -100%

and then I add this 3mixers
Mix a Flapfunktion at Ail into 3 steps 0% -20% -40% both Ail have to go down
Mix a Rud to Ail +25% Rud to the right, then Ail1 up and Ail2 down
Mix a Ele to Ail +30% Ele up, then both Ail up

this is from the logic absolut clear
test it and it works in math correct all togehter or separat

Test it with non Servo invers before!
----------------------------------------
And now set one of your Servo to invers, not before!
What all have you now to change get the math and movement in correct, lots of!
--------------------------------------------------------------------------------------------
But your problem at simulation is you dont not realy know which of the Servos you have to invert?
non , one or both

I know Companion Simulator screen shows not the math only,
it shows the output inculsive of the Servos with NOR or INV
and this is the second big Error with the real world.

And so you have the Problem that all your calkulations with +mix here and -mix there
are wrong in the maths, but you think your movements are ok!

Helle

[jhsa]

No, my way is not wrong it works for me..
I have a couple planes with flaperons programmed the way I think it is correct and they work well.. no problems at all.. So, I guess it is not wrong..
It is not mathematically correct? not a problem for me as long as it works and the results do what want.. so, i guess the other way is also correct
with your way you have to reverse one of your aileron channels twice if you have complementary servos installed on your wing.. now, for me that is not logical and it is a waste of mixing and puts even more confusion into the equation
So I guess each one to his own

João

[Helle]

Hy jhsa,,
you are right each one his own.

For teaching the funktions and the maths ist simpler to understand whats running
and yes, in the real world I have to invert also one Servo

Helle

[Helle]

Hy,

now here the same settings with one Servo is invers (CH5, but you cant see it)
because of the complementary Servo movement at Aileron.

So if the wizzard set both Ail to +100 it must also set one Servo to invers, but which one?
And so in the reality you have to change then non or both Servos to get the right movements
later you read it back form TX to simuation and all movements run in opposite direction
what will you do, all signs change from + to - and from - to +
Thats a typical simulationprobelm you cant solve it!

And lock at the math of the 3 mixer is it realy clear whats happens here and in which direction all will work?
I think no, but both will run well in reality

Yes I agree jhsa each his own

My Solution is simple:
I set no Servo in the simulation to invers,
then I get the right math in simulation,
then I get the right movements in the simulation

And in reality at the real modell I have to set then non or one or two Servos to invers
but I dont know which one and how in Simulation

2 Servos in same direktion mounted but from diverent produktion runs in opposite direktion!
Graupner, Futabe, Noname
A problem you cant solve




Helle

[jhsa]

That's the good thing about these firmwares you have many ways of doing the RIGHT thing..

João

[Romolo]

I'm sorry i do not agree at all...
It's a very old discussion, but
On a 4 channel receiver you can use two servos on ails using just a Y
For flaps on a 5 channels receiver you must use an inverting Y

In my mind that means that AIL1 and AIL2 can be driven by exactly the same signal: so if one is 100% the other should be 100% no servo to be inverted
For flaps instead it means: 100 and -100%

When differential comes in place, then right wing need a direction ie 40% and left the opposite.

Maybe this way of proceeding derives from the fact that I have started doing models when 4 channel radios were radios for rich people, and an Y was the only solution. (for flap you needed to open one servo and reverse pot and engine)

[jhsa]

or just turn it the other way.. or if the servos are facing down, use the arm on the same side for both..
It all depends how the servos are installed.
I do agree with Romolo.. it is exactly what I was saying before

João

[Helle]

Hy romolo and jhsa,

I also come from the old 4 Ch or 6 Ch TX
1 or 2 Aileron with a Y
1 or 2 Flaps with a Y
I know this discussions since more as 30 years too
and well, all will work correct!

Only programming direct at the TX, Step for Step, Mixer for Mixer,
and always watching the movements, Servo change, Servo not change and so on.

Now since some Years we can programm at the PC with a Sim
----------------------------------------------
In my opinion we have to separate
the mathematic of mixing in one Step
and the real movments in the second Step

its simple to understand what happens when we say
+100% is up or right
-100% is down or left

and its not depending of left or right side, the Servo direction or mounting

So if you have lots of mixers to the 2 or 4 Ail and the 2 Flaps and Breaks or Chambers you always have a clear working manual
What is right side,
what is left side,
what moves up ,
what moves down
and what is the result of each channel and you can calculate it direkt

not more, not less

and so is my sinple mixing example

At openTx its all possible, thats realy great!!

But when i am teaching this I always got the same questions about this
"unlogic programming"

So i say forget the Servoreverse first.
Programm it with this 3 ruels, test it at a Sim
take it to the TX and then make servorevers you need

And all are Happy and works "logic"

Helle

[ReSt]

Helle,
I believe, I understand your problem, and it has confused me more than once.
Especially if you are adding several mixes onto one of these channels.

Your point is, looking at the simulator, not the real model, ailerons have to move in opposite directions. So on the simulator, the signals for the ailerons should also move in opposite direction.
Flaps that have to move in the same direction should also move the signals in the same direction on the simulator.
Its another theme wich surface is #1 or #2 and if a +100 value should go up or down, left or right.
But it should be consistent.

This way, it is much easier to verify for complicates mixes on the simulator, whether the directions of the single mixes are correct or not.
THen, in the real model, if the surface is moving in the wrong direction, you only have to invert the servo and the total mix will be corrected.

Reinhard

[Helle]

Ja, Reinhard,

genau so ist es.

schau mal in das FPV forum rüber
unter Taranis sind viele Beispiele
das ist immer gleich und math. immer richtig

und mach das mal mit einem 8 Ruder Segler
und 3-4 Flugphasen, da biste sonst am verzweifeln

Halt dich an die 3 Regeln und alles läuft immer richtig

1.
positive Signale gehen nach oben oder rechts
negative Signale gehen nach unten oder links
2
das erste Querruder ist rechts,
das zweite Querruder ist links
3
Servos dürfen die Berechungsmathematik nicht beinflussen
Servos passen die Mathe an die reale Ruderbewegungn an

und alles ist gut, klar, logisch!


Helle

[jhsa]

in companion you are seeing channel output, not control surfaces. That is not a model simulator, it is a channel output simulator. Anyone is ready to write a model simulator for companion where you can reverse the servos? That would be great. And correct. or at least an option to reverse the direction of the channel bars in companion

[Romolo]

Creating a model simulator inside companion is a work of months, and results probably will never be optimal.
the aim of companion is to emulate the model.

Bitte, das ist ein Englisch-Forum.
Auch wenn ich Deutsch verstehen können, es wäre besser, nur in Englisch zu schreiben.
Kann man sich vorstellen, was passiert, wenn ich schreibe in italienisch, jhsa in portugiesisch und so weiter?
Viele danke...

[ReSt]

in companion you are seeing channel output, not control surfaces. That is not a model simulator, it is a channel output simulator.

Exactly this is the reason. Ailerons go to different directions, and therefore the channel outputs should show this.
If I have to reverse the servo (in the limits) to adapt the channel to the surface direction should not be covered from Companion.

(I can live with it, but I see the irritation the actual behaviour creates)

Reinhard

[Helle]

hy boys,

of course its a channel output simulator, and I know we have to adapt the real servos to the real world, no problem.

By the Way

Whats the reason that at DR/Expo Weight is only from 0 to 100% but not from -100% to 100% as in Mixers.

For example Elevator pulling should give a positve Signal
the Poti will give what, dont know, but.....

And yes I can take a invers 3 point curve when I want that direkt at elevator page
or a self generated invers Expo when I whant that
But why not weight from -100 to 100% and all would be free to set.

And by the way:
My positv mixer logic would also working fine
positv stick levels (elevator pulling!) must be set to a positv mixer signal at the mixer page and the result must be a up reaktion

or same as throttle insers,
set a bit for invers ele and pulling ele give positve levels


I know later we can do mix with weight -100% but at each Mixer line, that to late I think

Helle

[jhsa]

Rest, ailerons go in different directions but servos rotate in the same direction for that to happen on a conventional setup. Therefore the output on the channels are the same.

[ReSt]

After playing a while with templates and wizzard in Companion and eePe I think I found at least two reasons for confusion.

I simply applied the elevon/delta mix

1. eePe and Companion create different mixes
2. changing the channel output order changes the function.

eePe with channel order RETA behaves as I like (and expect) it.
First aileron channel goes right with stick right while left aileron channel goes left. One going up and one going down.
Both Elevon channels go to the right (up) when stick (value) goes up

eePe with channel order AETR reverses the two channels and now the first aileron channel value goes left when the stick goes right. Second channel vice versa.
That means, you either have to exchange the two servo connections or you must switch them to reverse to get the same behaviour as before.

Companion programs both aileron channels to go into the same direction while the elevator channels go to different directions. It also switches the channels when the Channel order is changed.


I believe
1. both, eePe as well as Companion shoud behave similar
2. as the channel order is important for the function of the mixes, the channel order should not be part of the General Settings, but part of the Model Settings.


Reinhard

[Romolo]

Sorry but channel order is used when creating a new model so it cannot be part of a model that does still not exist.
Maybe you forget this is a free mixer radio and not a template based one.

Again, on delta I have servo specular ( aileron signal so is the same)
Elevator need a mirroring...
But again is philosophy...

[Kilrah]

2. as the channel order is important for the function of the mixes

Not at all, channel order is purely arbitrary. You can plug your elevator in channel 10 if you feel like. Nothing states it has to be 1,2,3 or 4. The general setting is there because everybody has some kind of preference, and setting that to your own preference saves you some time when creating a new model. Templates then follow that - but that's all, and you don't have to use them. If you don't like the channel order, or how the templates and wizard work then you probably already know enough not to use them and to enter your mixes yourself the way you want them.
I personally do not recommend using templates as you won't learn much with them.

All of these things are based on personal preference, so changing anything may make some happy, but will confuse others. Can't please everybody.
Jhsa (and apparently Romolo) like to do elevon mixing with both ailerons mixers positive, and both elevator mixers opposed, because he's looking at the servos. For me it's the opposite that is logical because when you look at the control surfaces when you act on the elevator they both move the same way when you apply an elavator input.

As I see it stick to servo direction is arbitrary, there is no convention to follow there. As long as you know what relative motion you want and enter that in the mixers, then all that's needed to translate to the model is to invert the servos that work the wrong way in the limits screen.

[Helle]

Hy Kilrah, Romol,Rest,Jhsa,

Yes it must be free to mix in this or in the other way, not depending of the first Ail right or left
or elevator both move same way or not.
thats ok

The same should be the elevator stick behavior
when I pull the stick I what from these elevator positve signals to the mixers
elevator stick deliver no negativ signals
(ok I can set a invers kurve)

if weight at DR Expo would be from -100% to +100% (now only from 0% to +100%)
I can invert the elevator direkt at the stickside
got a invers linear curve or a invers Expo curve automatic

can use it as "normal" positiv signal as input to a mixer
dont need any invers at the outputside of servo limit
and the control surface would run "right"

So only get weight of elevator Dr/Expo itself to -100%
and nothing of servo revers at outside of control surface must invers

So all of the two partys would be happy


two ways to do this
a) as elevator invers button same as throttle invers
or better
b) open weight at Dr/Expo to -100% to +100%


And a new aspect:

If we make a open libray with standardfunktions we cannot know the real movement of the rudders
but we can generate a logic correkt mixerset with all servos are at normal first
and the control surface will run the logic in correct way.

so one rule for this should be:
positiv input to mixers will generate up or right
negativ inpus to mixers will generate down or left

and all users will be happy

we here and the new one comming next days

each his bear

Helle

[Kilrah]

D/R and expos from -100 to +100 can not be implemented on stock board. So for consistency I think it must be left alone.

positiv input to mixers will generate up or right
negativ inpus to mixers will generate down or left

IMO it still makes no sense, as we don't know what direction the servo and linkage will work anyway. Stick right or up gives a positive signal, a positive stick signal gives a positive on the mixer, and a positive on the mixer gives a positive on the output, and that's all you need to know to make anything correctly...

[Helle]

Hy Kilrah ,

whay not in stock board, only A/D with 10 or 11 bits 1024 or 2048

direct after A/D with
value = 1024 - value
and the stick itself works invers

all before normalising to internal variable -100% +100% or -511 +512

helle

helle

[Kilrah]

Because the EEPROM field for D/R and expo doesn't have enough capacity to hold negative values AFAIK, would need big changes.

Reversing stick action is of course possible, but not desirable as you wouldn't want to change behavior of all existing models, only to respect an arbitrary convention that doesn't add anything...

[jhsa]

can't we have negative expo? I'm not sure but I think i used it once on er9x. About the direction thing. are we arguing over a system that allows us to do whatever we want, right or wrong? I think we all have better things to do.

[G550Ted]

Agreed. This "directional" thing is meaningless because of how a model may be set up. Time to move on.

Ted