|
Forums 1C:Maddox Games IL2 Maddox General Discussion Thoughts on a realistic flightstick...is it possible?|
Go
|
New
|
Find
|
Notify
|
Tools
|
Reply
|
|
I didn't say you wanted a variable spring, I said you wanted a non linear spring. There is a difference. ---------------- Flying online as nate85 "I can buy a scalpel, that doesn't make me a surgeon." - M_Gunz |
||||
|
As I wrote, that was a description of the --standard digitizer circuit-- that the pedals might hook up to. Some input devices have the digitizer built in, USB devices that use pots/halls for example. It's kind of like when I write about ohms of resistance I get reply about resistance to foot movement and then accused of throwing random information. |
||||
|
You are absolutly right Wolf-striked. I am building a joystick setup at the moment which is at the stage I can use it in game.
In the first instance it is based on the force feedback mechanism from a Logitech Wingman. This mechanism determins the throw. The rest of the mechanism gimble is constructed from 25mm square aluminium angle and bar. Once the throw was established I could purchase some gears to give a full turn to some Vishay pots which are output through a Leo Bodnar board. The mechanism is mounted on a 200mm high box with a further 200mm extension (old chrome curtain rail) between mechanism and grip and buttons from old Logitech stick. This brings the grip to a nice level just above knee height without an excesively long movement. The feedback force proved insufficient to give really positive centering so, as per your suggestion, I added springs to improve this and make the whole thing much stiffer with increasing resistance as the stick is moved away from centre. I also added some drawer runner based pedals which are attached to the box. The overall result is a quantum improvement over other sticks I've used (Logitech Wingman, Saitek Evoforce, Microsoft Sidewinder). Control over the plane is absolutely precise. There is no fighting to get and hold a target in the sights as my plane yaws and wanders around the sky. Tighter turns are made easily without sudden spins and landing even on carriers are almost guaranteed with a rock steady path in. The only slight downside is that rolls in particular seem a tadge sluggish but this can, hopefully, be improved with some more tweaking of the hardware and software. There is still a bit of work to do and the addition of a control panel using switches and buttons but perhaps I'll try and get some pictures loaded sometime soon. |
||||
|
Shadrach52,while I thank you for mentioning me its ALL your work.Thats great news and I feel we ALL are on our way to enlightenment.  Go north(tighter)my friend and seek more.Your looking to ncrease spring strength until a fully linear setup is very precise to fly....and then go abit stronger. My logitech sidewinder had one spring to control the tension...is this the way you have it setup?I think the best results would be achieved by using a gimbal that has two springs,one spring to control pitch movement and one spring to control bank movement and that the pitch and bank are on different axis.The reason for this is that you want to be able to pull back on stick while feeling the banking springs guiding you...or letting you know your not applying any bank movements if you don't need them.Also by having separate axis for both you can fine tune the amount of spring strength for each.Here is a pic of a NXT gimbal http://www.wingsofhonour.com/h...in_3072x2304x24b.jpg The springs are way too small but you can see the separate axis setup. You might try reverting the length back to standard tabletop joystick,going full linear and testing as this will make the springs you have installed right now become even stronger due to decreased leverage. EDIT--also rudder pedals need to start being made with very heavy duty springs.A setup where you have to actually build a cockpit so you can apply the pressures needed to move the rudder pedals is what I mean.You want to have to really put you muscles to work to move the rudder pedal.Goes back to ability to run full linear rudders and the precision to fly it with precision caused by the high resistance of rudder pedal setup.Think about a plane just slightly off center to yours and you want to nudge the yaw over in the slightest amount.Then this setup will allow that. ~S~ |
||||
|
Actually that's not how it works. The analog voltage signal produced by the potentiometer gets run through an ADC or Analog to Digital Converter where the analog voltage drop is encoded into a binary number. That binary number is used by the game. ---------------- Flying online as nate85 "I can buy a scalpel, that doesn't make me a surgeon." - M_Gunz |
||||
|
I use a four spring setup - same in priciple to the Cougar mod.
My gimbal consists of the joystick shaft which sits in a 120 mm frame made from the 25mm angle aluminium and holds a ffback mechanism one side and a bearing the other. Two springs are attached to the shaft and to a post in opposite corners of the frame. The frame rocks between two other post arrangements, one with the ffback, the other with a bearing. Two further springs are attached to opposite sides of the frame and down to the mounting board. This arrangement means the direction of force from the springs are very close to being inline with the direction of movement giving maximum effect. I used posts each side of the joystick shaft as my first attempt had the springs running down at 45deg to the frame as per the cougar mod which in effect halved the spring force. The spring force is probably two or three times as strong as any joystick I've used. I could make it more but don't think that necessary and would begin to spoil the force feedback effect which is already weakened by the leverage of the stick extension. I extended the shaft to make the range of movement more realistic but I am sure it would work as well without and the base could be clamped to a table top. Unfortunately my digital camera is broken or I could give you a better idea what I mean. |
||||
|
I was just simplifying it and saying that a capacitor is not an electrical device thats needed.Your right,a joystick(some are digital nowadays??)is analog and needs to be converted to digital since a PC needs binary.But capacitor?That stores an electrical charge to be released at a later time. And I say again....I learned this stuff reading books on circuitry....meaning u gotta give me a break. |
||||
|
Are you able to run full linear and still be precise.By forgoing the FFB you may be able to up the spring strength by large amount and full linear could be super precise...instead of the overcompensating mess it is if done with weak spring joystick. Very soon I am gonna order a NXT gimbal.I don't own a cougar but will install it into a small box with a tiny nub for the joystick.I think this,while not being realistic compared to a full handheld stick,will give me the tension I want since working with finger strength is harder than moving your arm. Keep going with it.... |
||||
|
I hope this was not only intended as a joke but also understood as a joke by the audience? the voltage difference between "ground to collector" and "collector to high voltage" does not change with higher reistance pots, only the current would shrink (maybe making noise more of a problem). What you can to is getting a pot with a very short angle of travel, but the better idea would be to ditch pots completely and use less antiquitated technology, like a hall sensor, or a load cell, if you are really deep into the "high force, little movement" thing. But what you should really ask yourself is: "why the focus on desktop sticks?" If you have pedals and maybe even a trackIR you are already way out of the "socially acceptible", you might just take a long throw stick as well. PS: if they have gearing then it's probably just cheaper to manufacture that way (less calibration issues), that's the only possible explanation i see for the X52's play-inducing "mechanically augmented hall sensor" |
||||
|
How about giving a break to those who have dealt with these things hands-on even part time for decades already? I was trying to help you telling you how and why changing the pot makes adding mechanism the wrong way to go. If you want a good starting point for hands-on then I can recommend something like these. Books alone are not the way to learn, certainly not the way I learned either. One thing I did learn is how many people who never learned think that everyone who did got everything from books and are so quick to prove that "books don't have everything" and then go on about things like boiling water freezes faster than cold water or whatever else their Granny told them. Next step is always the fist fight since that's really how you decide how things work. Hey, emotion takes over where knowledge leaves off and there's lots of shallow puddles out there. Since I lived in a poor (only faggots read books) neighborhood the last 7 years of school I learned to argue either way, that goes back to 1967 so guess how much bull I'm ready to take? |
||||
|
Because it's a lot cheaper to check a capacitor for "voltage above threshold or below threshold?" ten thousand times a second than to actually measure the voltage (with more than two possible results). One can be done by chips that maybe cost half a dollar, the other needs a chip that might cost two dollars or more (disclaimer: those numbers are completely made up). Things like this make a world of a difference in consumer electronics that will go over the counter for a two-digit amount. About spring setups: i use a two spring setup (three with pedals), one spring for each axis. Stick deflection is transferred into pull on bowden wires and an axis' spring gets extended by deflection to both sides. It is very easy around the center (barely enough to keep the stick straight) but it becomes much stronger towards the limits of motion. Right now i'm in the process of adding FFB motors to modulate spring strength. The stick is running full linear of course, so i am never trimming for higher precision around the center, only for reduced load on the springs. |
||||
|
I should have just linked to something like this the first time.
There are better and faster ways but not much cheaper ways. The same thing or better can be found built-in to many cheap user-programmable microcontrollers like the STAMPs and AVRs as a small part of the whole package. Leo Bodnar uses one of those unless he manufactures his own chips. The plus of using those is that you don't have to go the ADC route, his updated controllers also allow pure digital measure as well much the same as non-optical mice count spokes movement of internal wheels (turned by the mouse ball) past two light bridges. LOL, which is better, the optical mouse or the one that uses the mechanical ball and wheels? No contest! |
||||
|
I'm of the same opinion as Gunz, but would have expressed it differently.
Gears (especially the gearing used in the average gaming controller) will have slack, and give unwanted dead-zones when ever you change your foot's movement direction. Rewording what Gunz said, if the original pot was 10k, and zero pedal deflection gives ~0 ohms, then you can pick a 100k pot and position it so that zero pedal deflection gives ~0 ohms also, however the resistance will rise 10 times faster, meaning you get full in-game response with lesser deflection. To do the above well, would require no-slack in the pots mounting and shaft connection (ie, slight interference press fits), however doing this is easy and cheap compared to using gears without slack. Either way, why gaming controllers are still using pots for critical axis (x/y/pedals) measurement is beyond me.
Two springs in series will let the weaker spring compress first, then move on to the stronger one (ie two stage spring constant). Two springs in parallel (ie inside each other) will have a fixed spring constant.... Unless you were meaning that the stronger spring is shorter than the weaker one, and the longer one has to be compressed a bit first before being contacted. ---------------------------------- Flying online as 453_Whittle E8400 @3.8, 9800gtx+ Phenom II X4 B50 @ 3.4  Uni-joint / hall effect sensor stick guide: http://www.jpfiles.com/hardware/uni_stick.pdf |
||||
|
See my post above.
Well put. Except I don't mind whether the stick has any centering force or not, hence I didn't bother with springs. Mine has the frictional resistance of a tightened uni-joint. ---------------------------------- Flying online as 453_Whittle E8400 @3.8, 9800gtx+ Phenom II X4 B50 @ 3.4 Uni-joint / hall effect sensor stick guide: http://www.jpfiles.com/hardware/uni_stick.pdf |
||||
|
 |
Modded Thrustmaster Cougar.
Nuff said. ------------------------------------------------------------ "Of all lovers perhaps none is more unrequited than a liberal humanist. History makes fun of him. Misanthropes deride him." - Harper Magazine |
|||
|
Ok change to a pot that has higher resolution in smaller rotation so that you don't need gears.Forget all that for now.
This post is not about the gearing needed to make a pot turn.I wanted to point out that the direction that joysticks are headed is wrong IMO.We dont need longer throw joysticks....extending a joystick does give more precision...but a better way is to make the resistance of the throw we have now very strong.Much stronger than say a Cougar. We all agree running full linear is the best way to run a sims controls.But when you do that you now have joysticks that have no center position and weak springs.With heavy duty springs you will feel the INSTANT you start applying backpressure and by tactile feel youll be able to add a bit more or a bit less. With weak springs you have to use your mind to move the joystick say 1/16 of an inch.With heavy duty springs you pull and judge this amount by "feel". Some people will understand this and some won't. M_Gunz I actually know alot about electronics just from reading.When I was younger,as a hobby I used to build circuits that convert actual wall outlet electricity to DC and stepped down to the voltage I needed....all done on a small circuit board....yes I electrocuted myself a many times learning and doing this.Basically I needed a power supply source(variable)and built one from scratch all learned by reading. But again....forget the electronics of a joystick....I am talking about the mechanical side. |
||||
|
I tried a Cougar in a store and the resistance was suprisingly weak compared to what I had heard about it.I hear modded sticks have reduced tension for longer life of springs. What I am talking about is running full linear with enough resistance that you can move the stick by "feeling" for precise control. |
||||
|
 |
i popped a higher tension spring into my logitech and there was more tension on the initial pull, but the resistance actually flattens out as i near the edge of its throw limit.
I was expecting it to have the most tension on the outside of its throw but apparently the way the joystick is built the spring actually compresses the most in the first two centimeters of throw than it does the last centimeter of throw. If you can imagine pulling through the resistance and then have it suddenly let up. Its difficult to be precise through that part of the throw curve. I think the stick was designed around a specific spring tension. This effect is actually present with the stock spring but since the tension is way lower its not at all noticeable. I would actually opt for a longer of a handle as well as increased tension for accuracy, but on another note, accuracy is really just a matter of learning the joystick you have. Some might be more accurate than what i have, but if you are use to relying on subtle hand movements then accuracy can be learned, not only engineered. |
|||
|
Indeed Bill. Try flying with just index finger and thumb holding the stick and see the change.
A biggie is whether or not the player rests arm weight on the stick. That and a tight grip are two factors in ham-handedness. |
||||
|
Like my tests with the T2 racing wheel yrs back.There was this sweet spot that lasted a few minutes.All of a sudden I was driving a car on my PC it felt that good.After 3 minutes it would just start to slightly lose tension but the feel of a real car would diminish rapidly.Yrs back I actually built a steering wheel from scratch.I used super heavy duty springs but the gimbal itself left the center position slack and then as you turned the wheel the forces became exteme so at full lock you would need two hands for prolonged turning.It just didn't cut it after feeling the modified T2. I know what you mean about the sidewinder spring loosening up at edges.If you look at it you'll see that when you first start a movement the plastic piece compresses the spring but then the side opposite to direction your deflecting starts to rise and tension reduces.Much better than the Saitek with there non-existent spring feel. |
||||
|
| Powered by Eve Community | Page 1 2 3 4 5 6 7 |
 | Please Wait. Your request is being processed... |
|
Forums 1C:Maddox Games IL2 Maddox General Discussion Thoughts on a realistic flightstick...is it possible?
