• Welcome to PiBoSo Official Forum. Please login or sign up.

Project Realistic Rig

Started by speedfr, December 08, 2017, 02:10:31 PM

Previous topic - Next topic


December 08, 2017, 02:10:31 PM Last Edit: December 08, 2017, 02:14:43 PM by speedfr
Hello riders,

yes, i'm having the guts to call it "Realistic"... We'll see.
But i've been riding my bike and others this last week to check the mouvement. I'm driving bike for more than 30 years and i never really paid attention to the cinetic even after watching every week-end every possible bike race in the world...better later than ever...
And  then i tried to find a good way to reproduce what's happening on a bike AND that i could "ride" on my chair in front of my screen... --> GPB !!
Because i don't want do a real frame bike and having to use hydraulic or electric servo to shake, lean and move the frame. I don't have the room and more than that i don't have the money.

And so...
In red you have the handlebar that will be counter-steering to lean on the sides. This mouvement (vertical axe) will be short angle but hard to do (piece of metal into a big rubber that makes the mouvement harder - trying to reproduce what is real and what a steering absorber will do). The green piece will then turn left and right in a horizontal plan.
The blue part will be the driver leaning and can go on a horizontal axe from top to bottom, this will be a free mouvement, inside a "teflon" piece that will make the mouvement smooth (as in a bearing wheel).

Obviously, the feet parts are missing - brake and gear - and all the buttons on the handlebar but that easy stuff, what was annoying is to find a way to reproduce counter-sterring as it is in real life. I want a be able to ride in GPB without thinking of the device, just being absorbed by the screen. Like Assetto Corsa with a steering wheel, you forget about the device, you drive.

Now it's going to be time to make a prototype and then improve the stuff.

Thanks again to DD because he helped me a lot with the Leobodnar stuff and since HS1, HS2 and soon HS3, we have of excellence to follow, doing my best !!! (and if i could buy one and race instead of burning my brain into that rig, i'd prefer !!! )

Missing Gp500 (Microprose)  Testing EDTracker Pro on YT   R7-3800X/16Go/GTX1080/1440p/Full WC


December 08, 2017, 02:23:50 PM #1 Last Edit: December 08, 2017, 02:27:42 PM by poumpouny
good idea but i doubt it will be natural to move into 2 direction on the same mouvement, i think on axes for the steering and ed tracker for the lean will be more natural ! the other problem is that if you want to use counter steering, you must switch to DSA or DST hardcore mode, and thoses modes without ffb is very difficult to handle cause you don't have any feedback to know where is the bike lean limit ! anyways i wish you the best with your rig

Edit : Or may be it would be more natural if you move the steer in the mid of the lean axis, that way you controll the lean by pushig the handlebar instead of swinging it !


December 08, 2017, 03:22:37 PM #2 Last Edit: December 08, 2017, 04:52:16 PM by speedfr
Hello Poumpouny,

and thank you for your feedback, i always consider than more brains are on a problem, better the solution will be at the end !!

And about what you says :
About natural mouvement, as i said i spent hours this last week analysing the mouvement, even put the GoPro of my son to film my hands and handlebar on my Pan European or on my other son ER-6 Kawa and i found on images what i was doing without really knowing.
And this mouvement, despite what you think, is or begging natural to me as every motorbike riders that ride for a while.

About the horizontal axe moved by the fact that i counter-steer on the vertical axe, this other idea (i post the "old" one - from 2 days) and this one is the news one from this morning at the breakfast time.
And i think that will absorb more the bike leaning mouvement to not trouble the rider leaning movement. Or, if it perturb anyway, i will have to make the rider leaning harder. I want to try anyway because that's how it feels to me.
About EdTracker Pro, that i use since it's out on the market, i will only put the Back and Front rider leaning.
I will be seating on my chair, not on the frame that's gonna move left and right.
So yes, want to ride a bike seating on a chair looks akward but that's the challenge.

About FFB, yes, i would love to FFB my vertical axe but i'm not sure (that was my question to DD) that the FFB is actually to copy the reality, meaning bloking me to turn if i brake hard, as in reality. And on a race, guys are allways braking to late and have to force down the bike on the side they want to go. Its natural centrifuge and centripete forces, plus the gravity and the wheels inertial mouvement. On the regular street bike, you brake, turn and the accelerate, on a race bike you gonna brake and turn but still brake on your turn (on the angle) so braking will have a tendancy to make the bike stand back up to the vertical and is going to refuse to steer and then you have to force it to put it into the curve, that's what i will love to find with FFB but i'm not sure it acts like that. (where to  catch the data for that ?, as i wrote somewhere else, as a Motec in racing cars).

And about DSA or DST i have no idea what is it, information in here and about the game is an archeology type of thing, so if you know what it is, i'll be more than happy to learn about it...

(what you say about vertical axe is interesting and i started to draw something like that but the draw isn't finished, i'll give it a try to the end of design and post it in here)

Edit : what you couldn't see in the other draws is that the horizontal axis on the handlebar is at the same axis as the riders leaning axis. So if i push my handlebar in a horizontal plan (to counter steer), it won't be pushing the rider leaning left and right.
And if DSA or DST doesn't give any feedback, i was thinking of using limiter after calibrating is done in order to give myself a max angle of 60° or something around that. (today, i'm using a joystick and a drum pedals with EdTracker on it for the rear brake, it's a catastrophy !! and i have no feedback anyway so...)

Missing Gp500 (Microprose)  Testing EDTracker Pro on YT   R7-3800X/16Go/GTX1080/1440p/Full WC


Interesting idea!

I will be following this one :)
Hitting the apex is just a figure of speech, Don't literally hit it...


February 01, 2018, 06:20:29 PM #4 Last Edit: February 01, 2018, 06:29:40 PM by speedfr
Hi Matty,

i'm in the middle of testing how to display an image AND resizing it.

Working on [img] code and i'll be back with the second chapter  ;D

And now it works (the forum doesn't like the short BB code.)


Missing Gp500 (Microprose)  Testing EDTracker Pro on YT   R7-3800X/16Go/GTX1080/1440p/Full WC


February 01, 2018, 06:43:12 PM #5 Last Edit: February 01, 2018, 06:49:12 PM by speedfr
Yes, I'm a HAPPY BUNNY !!

Second Chapter of this intense story...

Always heading for my most realistic possible system, I just gave a few draws in here but I did a bunch, submit them to DD know-how and after long skype session, he convince me that the way the HS work (pendular system) is the one that gonna hurt me the less if playing for a while. Sure I'm not 20...

So, after all that, DD allowed me to purely copy his system, and at the same time, I ordered him some parts to start my project.

Yes, it's a start. Because I'm going to need to have 2 handlebars to play GP Bikes. My youngest son (18) is thrilled by the game and once he will see and try the handlebar on GP Bikes, he will refuse to use the Xbox pad which he doesn't like already...

So yes, DD parts are there for me to start with and have a base.

Then I'm going to equip that, set the system in place and I would guess that the next two weeks are going to be around this main objective.
And, proud as I am, I want to make a little unboxing kind of thing, of the masterpiece printed with ABS by DD, the quality is incredible and it seems super sturdy. He found me an extra main-plate and all the electronics parts from Leo Bodnar that I need.  Plus all the little parts for the cables, screws, it is great.

One more time big up for DoubleDragon to offer me this opportunity, thanks again for all the time you spent on Skype, on mails, this special package you made for me and all your kidness for this kid that I am.

Better than words, pictures :

Opening the package, find that on top.  ;D

Here is the Leo Bodnar, all setup in R&D lab at IAS Computer Controls by Allan.  8)

A box with 2 P260 and the system for the spring on the lever. (Brake and Clutch)

Two handlebar support for 22mm tube.

The main object...  :o

The main plate with three ABS printed system for the potentiometers.

The quality is incredible, ABS is perfect, and it moves perfectly with no friction ! Awesome work   :-*

See you soon for the WIP  and the third Chapter.  ;)
Missing Gp500 (Microprose)  Testing EDTracker Pro on YT   R7-3800X/16Go/GTX1080/1440p/Full WC


a shame you let DD convince you to ditch the "true countersteering".


February 01, 2018, 08:50:28 PM #7 Last Edit: February 01, 2018, 08:55:31 PM by speedfr
I understand what you meant Vini97 but two things :

-First, the vertical axis is the problem as well as the solution.
It will need to steer easely at low speed and get tuff at higher speed (starting around 40km/h but actually when Gyroscopic forces will says !! ) which means having data coming from the software to engage some kinds of resistance (torque system with FFB inside) on this axis when the horizontal axis (leaning) will be on regular axis-bearings-spring.
On my mind, i was going to need something like a silent bloc, a hard one, and needed to measure the movement to induct the steer/counter steer.
But even with those, the risk is big that the strenth input will force the leaning so that's not solved on a simple idea, it needs programming and driving a FFB engine to give you feelback. Not that simple.
Probably possible since GP Bikes can send data outside (like the ones used in MaxHUD) but not yet for me, need more investigation and more capacities that i don't have even if I am a DIYer and love to work with my hands.
In conclusion, later or maybe will stay as a dream, but not in the sight now.

Second is that i paid attention to my movement on my bike, for like 2 weeks, constantly "studying" my rides.
And DD pendular system forces you to move in a way that correspond exactly to a bike movement.
On the video where Gren rides at the Holovis Lab (on DD Youtube Channel), look how Gren moves taking and exiting a curve.
That's exactly what you and I do on our IRL bikes.
His look (because he doesn't have a screen to fix with his eyes) goes to the apex, he his inside in Vr thing and his head goes exactly as in reality.
I mean the move is "natural" for a rider.
And so far, because it's "makable" for me, i think it will be the best.
I saw a lot of DD videos, and its about years that he works on his system, he knows a lot about that, his knowledge and experience is precious.

The real counter-steering request first a lot more thinking in order to have, sitting on a chair, the same feelings as sitting on my bike.

What i understood, is mostly that i need to go back to my studies about it and for now, i'm going to have a handlebar to finally ride as much as I want, as regular as I want - repeating the same lap more than once - and mostly with the same feeling or at least something familiar.

Always better than a joystick, a drum pedal or a broken wheel to ride GP bikes.  ;)

Missing Gp500 (Microprose)  Testing EDTracker Pro on YT   R7-3800X/16Go/GTX1080/1440p/Full WC


February 01, 2018, 09:46:38 PM #8 Last Edit: February 01, 2018, 09:48:51 PM by vini97
Quote from: speedfr on February 01, 2018, 08:50:28 PMSecond is that i paid attention to my movement on my bike, for like 2 weeks, constantly "studying" my rides.
And DD pendular system forces you to move in a way that correspond exactly to a bike movement.
I believe that. The pendular system should emulate the correct relative position and angle of the handlebar to the rider. The resulting forces one has to apply to the handlebar should be very similar to what you need to do on a real bike (countersteering).
My ideal controller for GPB would essentially look similar to DD's but work a bit differently. The handlebar would be swung left/right by a strong electric motor, firmly holding it in place against direct user inputs (output only). The swinging arm holding the handlebar would be a tube that can rotate around itself, acting as the (vertical) steering axis just like on a real bike.
If calibrated properly, this system should result in exactly the same user movements as DD's system does but with the big difference of giving actual feedback from the front. So for example if the front starts to chatter or if the rear steps out, you would feel (and see) it on the steering axis, independently of the lean angle. Also, steering would become harder as the speed increases.

As you said, the challenge is using the steering axis simultaneously as a torque input and output (resisting the input). Theoretically it is possible since GPB outputs all necessary data and accepts steering torque as an input (using DST mode). It's a pure engineering problem.


February 01, 2018, 10:00:54 PM #9 Last Edit: February 01, 2018, 10:07:43 PM by speedfr
Will need to have the counter engine inside the tube. And if we want real torque resistance, it will turn into a hydraulique system, because the electric motor can't have the necessary gear to create the force.
Hydraulic means pump, heavy fat tubes or metallic ones, a reservoir of hydraulic fluid and energy.
But the hydraulic distributor then need to be driven by the data coming from GP bikes = the torque on the axis that seats under our handlebar (i found the word : steering column  ;D )
it will request a programm that control the distributor action.

At the same time, the leaning axis has to force against you, same thing, by hydraulic, to make the leaning harder to take when fast on a curve, i mean you need to force on the inside foot to help the bike turn so, that request another hydraulic actuator and distributor.

That's for staying on a chair or kind of bike seat.
Because, after that you can steal a RX6 and put it on a board with 6 DOF and that might become more simple...  ;D

I think what we are looking for is possible but exactly as you said, it request a big engineering thinking.

Edit : And to do even better or more close to reality, the best will be to have the horizontal axis ... on the floor, where the front wheel hits the ground. Exactly there. And reading angle or hydraulic actuator will go there to. But the seat need to move sideways....
Crazy thing to think about already... but....  :o
Missing Gp500 (Microprose)  Testing EDTracker Pro on YT   R7-3800X/16Go/GTX1080/1440p/Full WC


Here a example of the movement with 20° steering/counter-steering max ( 2 times 10° forward axis) and 48° lean angle (2 times 24° vertical axis).

I don't know where we sit on this configuration but, the seat is going to have to go sideways a little bit or shoulders or wrist problems gonna come fast.

Missing Gp500 (Microprose)  Testing EDTracker Pro on YT   R7-3800X/16Go/GTX1080/1440p/Full WC


you would have to scale both axis of course but if DD's controller doesn't give you shoulder problems, then this system should be fine, too.
the few centimeters of additional absolute movement of the handlebar stubs can be easily compensated by slightly leaning your body, which you are probably doing already anyways with DD's system.


I'll tell you as soon as i'll do...

but i still have to puts everything together.  ;)
Missing Gp500 (Microprose)  Testing EDTracker Pro on YT   R7-3800X/16Go/GTX1080/1440p/Full WC


I dont "slightly lean my body" even on the couch. The SPSS system will have you fully leaning once you get into it. I found in VR over at Holovis I was on the limit of leaning a few times as I got a bit carried away with it lol.

One thing most forget when thinking about designing a system is that ALL software even GPB demands a certain amount of input to work correctly. If you design a system to work EXACTLY like a real bike you will crash all the time as you will have to sensetive movement at the bars due to the fact you hardly move the bars at all in real life. After 8 solid years working on all types of designs I know just how bad it can be lol.

I will give you all details on wiring the BU board and other points on skype speedfr.

GPBOC Live Streams: https://www.youtube.com/c/IASystemsComputerControls; i7 7700 5GHz z270 ASUS Maximus Code Mobo 32GB 3200MHz DDR4 RAM ASUS Strix RTX3080 OC 10GB DDR6X Kraken X62 AIO Cooler ROG Thor 850w PSU in ROG Helios Tower Case.


Quote from: doubledragoncc on February 01, 2018, 10:55:04 PM
.... I found in VR over at Holovis I was on the limit of leaning a few times as I got a bit carried away with it lol.


I will give you all details on wiring the BU board and other points on skype speedfr.


That's what i was kind of worried when looking at Gren, it needs big matress on the side of the structure because if you start to be "inside" the screen, you might fall or need to be in good shape. Gyroscopic and Centrifugal forces will always be the problem.

@DD : i'll come to you tomorrow afternoon if you are around.  ;)
Missing Gp500 (Microprose)  Testing EDTracker Pro on YT   R7-3800X/16Go/GTX1080/1440p/Full WC