My 3DOF project is born!

Ehhh this is a good question!.
I think each of us have a personal idea of what is the more realistic way to simulate a real racing car feeling. I have my personal idea from all my personal real life experiences of when I was inside a fast car racing somewhere... (a lot of mistakes of youth age.. :uups: ) and the result of the mix of all my experiences is one word : ROUGH!!! hehe.
Plus my racing style is really hard and jerky so the result is sure a lot of shakes!!
With this profile, today, I got a lot of fun and 6 precision times in where I was really near to a real feel.
1) In fast cornering.. feeling the bump of tracks.. hoping the tyres continuing to get grip.. :rofl:
2) When I back on the track after a out of track section.. the difference of height and I feel very well the wheight of the car hitting the ground!
3) Finally I can feel good and realistic the breaking.. In old profiles I never felt the correct sensation of position respect the breaking force I using. Now It's all more more realistic!
4) When I am very fast in straight so I can feel the fear of the speed!! the sensation that I am near to lost control and crash!!
5) Going very slow (10/20 km/h) and hitting the curbs so I can feel every single steps of the curbs.. really impressive!
6) climbs and descents are pretty scary now! Sure more near to the classical hearth in throat sensation when you take a fast descent than with the old mechanical and profile configurations!

About the my personal concentration when I am running with my subaru, believe me, never I was so focused and passionate in a videogame as today..
A good mix of many little realistic components and the immersion can really touch good levels!
:thbup:

Did you read this bit of the thread?
I don't really understand your relay schematic but I think it will not work. Bidrectional control requires four relays. The jrk outputs a 20 kHz PWM signal that allows it to have fine control, but relays are basically on or off.
Yes a relay would work , but it would only give you a on/off function, full power or full stop, no real control.
Try fiddling with PID first. You are only running you jrks at 25amp from memory, so they are still safe. I run mine at 32-38 amps for spikes in current and they still live - just.

Yeap, the biggest pitfall of jrk's: You can't upgrade to bigger h-bridge... LOL

Don't hesitate, don't wait, GO!

Hi mate! The matter is not IF the thing work or not! I already test this method and it's working! Sure you can just obtain an ON/OF full power as you told.
But the problem is not too big from my personal prospective.
X-SIM just handle the motor positions, not the speed to reach these positions. If you work fine on the profiler you cat get smooth movement too when you are using realy method full power.

Last time I tested the relays method so I get the same behaviour of my 3dof with MAX CURRENT 25amps - cycle 600/600 - accelleration 600/600. Just one BIG difference :


My 3dof without relays with MAX AMPS 25amps - duty 600/600 - accelleration 600/600 :
Very brutal power but in some circumstances all this power is not sufficient to get fast movements and the motor and the jrk are suffering very bad (example when all 3 motors must to reach the minimum height positions challenging the spring resistance (at minimum height position the spring resistance is very hard to beat..) and the jrk get the conncetion lost error.

My 3dof using relays method :
Very brutal power too but no one time I got my motor suffering bad. Only little problem you can hear the little click sound when the relay is working but is a really low noise.

So for my 3dof typology the behaviour is not too different but I get a lot of overpower and this overpower is exactly what I need to finally archive my 3dof!

Hi, sure the little jrk not born to these applications.. It's really a miracle what it made since now! But I am pretty sure that one of the many solutions provided by this forum and the pololu forum can solve the connection lost problem.
I notice that lowering the duty cycle and accelleration I get less errors. The actual situation of my last videos is DUTY 500/600 ACCELLERATION 450/600 and I get really less errors than when I was using duty 600/600 and 600/600. If I put lower these parameters a bit again I am sure I can avoid these problems and I will lost just a bit of the actual power.

Yeah but i am really devastated dude..

I believe that to overcome the difficulties we have to work and use the imagination, nothing is impossible, each simulator design is different to others, but the end is the same, make it work :yes:

aldoz your simulator is a good job, and like other people you spend a long time, go ahead and each test you do to improve your simulator will be a step forward, good luck! :thbup:

I think a good way to avoid consuming too much amp motor, is to have less weight, so the engine will make less power, and I think in this way the card work jrk pololu safer over time, I think if we use spring would be a solution, how about? ;D

Yes, but does yours handle up to 38 amps?

This will work, it may take some different positioning of the springs, but it will help reduce load on motors. Just watch the springs do not cause too much load Resistance on motors, I found that , but experiment.
:

I didn't say it can, not the DSMhb, but at least I can seperate the controller and use it with bigger h-bridge, 150A h-bridge for example I can even connect the AMC to a VFD AC motor if I could afford one... :mug1:

If I would make a wiper motor sim, I'd put a small gas spring where the motors pivot arm connects to the chair (one for each motor) Size of the spring would have to be calculated or tested. I'd also take the load of from the wipers gearing with external bearings (something along 20mm ID)

could be a solution!, as well says david (BSFT), the car jrk pololu robot was created to fight, and the sirnoname began using in the simulator, the experience makes them 2 see that the card is very good, but we must find ways to help the cards to work with little effort, the goal is as follows:

1. make motors work with the least possible effort.

that goal if we have the advantage of using less amperage, so the cards jrk work more safely. And we could use one power supply, and so do not use car batteries to help power source, is that this is not easy, but we can do it! :thbup:

so is david, you have to do tests!, I think the springs may help reduce the burden of weight. :highfive:

We are taking over Aldoz thread here, I will add that I have built 3 wiper style sims, all with springs to alleviate the pressure that wiper motor gearing is under as it changes direction. Springs on mine helped stopped the gearing wearing out and reduced backlash.
I have not gone further in spring design, like Aldoz has done, but all is possible. I have some thoughts and sketchy plans on a small suspension adjustable strut to under each corner of a sim, just havent spent much time going further. I need to figure out a way to calculate that works for me.
I am sticking to what I have done at this stage. Feel free to try something though.
Otherwise, lets let Aldoz continue to share his work. :thbup:

My 3dof problem isn't the weight but the spring resistance (despite all the calculations done, I overestimated a bit the jrk resistance to hold a really high amps for a time of >1,5 seconds.)


I notice that you can be 100 or 120 kg guy and no problems here! you can get lifted without problems!
The problem is the resistance of the spring when it's pushed to the MINIMUM motor position (35 cm of spring pressed, using my spring so and so 292kg of total torque needed.. yeah there is the pilot and platform wheight to help to press the spring but the various inclinations and tilts make the understand of the value of this help really hard to find..).

It's not a problem about if the jrk can handle this! In fact the jrk can provide (with a lot of effort) to motors the right amps but if this effort time is over 1,5 seconds so I get lost connection.
The solution should be to obtain a new spring (a bit less rigid) but I done a too long battle to obtain my actual spring so I need to use it now.

Ok, the solutions are:
1) to use less vertical travel. (so to push less the spring then less newton torque needed then less amps needed)I I could reduce the vertical travel to 20 cm instead the actual 35. This would still be very positive vertical effect and I would get no more connection errors.
2) to use the relays method. Then the 292kg of total torque needed would no longer be a problem and my jrk's would be quiet and fresh!
3) To reduce the motor duty cycle and accelleration to 300/600 both and slightly reduce the vertical travel and I would get no more connection errors. Many peoples prefer to have a less rude platform so many peoples can obtain a 3dof with jrk's and winches without problems!!!!! this must to be really clear! :thbup:

BUT! I want to use all my 35 cm vertical travel :brows: and to obtain my motors on full power so tomorrow I will go to buy 6 realys to do some tests!.

Again about the jrk I have to say that it's a little miracle board! :yes:
It's little but permit to people like me (with really bad electronic and electrical study limits) to reach GREAT results!.
The 3dof + jrk is in fact it's really possible! and the problems can all be overcome!

Just me, I can not be satisfied since I can't see my platform working without problems at full motor power! :mushrooms:

Tomorrow I will try the hub usb solution before to start the realys tests. Over that there is the electrolyitc capacitor method (from pololu thread) still to be tested and the usb-opto isolated method still to be tested too! Ah I can't forget to add the PWM frequency of the jrk to 5 kHz instead of 20 kHz method still to be tested too!!!

A lot of chanches still to be tested to overtake the connection lost problem at full motor torque! ;D

:thbup:

aldoz I think this capacitor is used for car audio can solve he problem, so no need to use a car battery, so you can use a power source and the capacitor, the capacitor helps deliver power to the motor ... who do you think?

Yeah mate I quote you at 100%. All is possible, we need just time and unfortunately some money but the physics and math can produce ALL!.
I am curious about your 3 wiper style sim, is it a 3dof? or your 2dof project?

Ahh unfortunately my electrical knowledge are really bad dude.. and I would not flatter myself too much by reading your comment :uups:
From the little of what I understand the capacitor method should be similar to the relay method.. right?
fascinating and sure a thing to TEST soon as possible!!


I would to open a parenthesis about the simulation of the real racing car effect on an human body to follow on the past discussion from stowaway about the right behaviour of the motion system to reach the maximum realism! (too rough, too soft ecc).
let's see this video (really spectacular):


In my motion system I am simulating the Subaru Impreza 2001, so and so the same performances than the car on the video.
I would to reach something similar to this! And it's possible!
My tests were with no seatbelt so my body can to shake and move in the seat.
So I need in first to get a good 4 point seat belts..
I am pretty sure that with a good profile work (especially in the vertical travel, little fast bumpings and big bumpings) I can get near to the effects in this video.

So from now I will take this video as my personal point of reference of what must be the behaviour of the motion of my platform!.
Stay tuned! :brows:

This is why a pneumatic spring with an air tank is better than a coil spring. You will achieve a constant force throughout the whole range of the spring.

Sorry Aldoz, I should make myself clear, 3 styles of simulator design, all 2dof with 2 wiper motors.