MY 2dof wiper motor seat mover

[pipis2014]

As for most of the guys in this forum … a motion rally simulator was one of my childhood dreams. Something that was once a dream … but it is now possible to construct thanks to the devotion and contribution of many people.

So I want to thank in advance:
• the DEVELOPERS of x-sim software.
• the members on this forum for sharing their experience.
• RacingMat for the fine work and detailed presentation of his project.
• Tombo and Monaco85 for the "Pololu Dual VNH5019" code customisation.
• kalarius for his 2dof wiper motor profile.
• Plus everyone else who has and will contribute to this community

I also want to thank my friend Babis for his help, especially the welding part. We had a great time discussing testing and refining my original plan.

Having seen a lot of DIY projects come to life and evolve to working simulators, I decided to start my own project. I have to admit that I was reluctant to start such a project because I don’t feel that I master all the skills needed to complete the task.

Anyway, after having seen many different designs I decided to build a 2dof wiper motor seat mover. It is the simplest design to construct plus I don’t think that a tilting platform with only 2dof (moving seat+wheel+pedals) suits my needs or even offers better feeling than the simple seat mover. There is a strong debate on how “far” can someone go with his DIY design … but in the end its all a personal matter limited by cost and imagination.

Before I start the presentation I have to say that I already own a gaming seat so my project is designed as a detachable module that can be installed and uninstalled beneath the seat. That means that I have to go for the least possible footprint and the whole module has to be as compact as possible.

Lets get to the part list now to estimate the cost:
• 12v 47A 575Watt server Psu (20eur+15eur shipping)
• Arduino Uno R3 (23eur+12eur shipping)
• 9v ( 2.1mm centre positive plug) psu for arduino UNO (5eur)
• Pololu Dual VNH5019 motor driver shield for Arduino (36eur+12eur shipping)
• 2x single turn Cermet 10k linear pots (2x7,5eur)
• 2x SWF VALEO 24v wiper motors (2x35eur+23eur shipping)
• two 40cmx40cm steel plates + 15cmx15cm steel plate for Ujoint + 40cmx10cm bended steel sheet for motor chassis (20eur)
• USB isolator (form 25 euros on ebay)
• AWD Suzuki vitara Ujoint (free from scrap yard)

And some low cost parts:
• brush-less fan (12 or 24 volt) for h bridge
• aluminum heat-sink for h bridge
• 0,1μf 50 volt “ceramic disc non polarized capacitors” for optional “three-capacitor filter” on motors
• electrolytic capacitors 12v 500μf for optional signal shielding on arduino input
• CAT6A STP (SHIELDED CABLE) for pot-arduino ADC communication

Plus some bolts, nuts and washers and cables that will sum the final projected cost to around 300-350 euros. Considering that we are using new parts the cost seems reasonable.




->

[pipis2014]

*** SAFETY FIRST ***

When using power tools for your projects try to take some precautions in order to avoid "stupid" injuries.







For my build i didn't have to use any heavy machinery. Just "usual" stuff.



All "special" jobs were done by a local machinist.

[pipis2014]

*** Part 1 ***

The u-joint is probably an overkill but it was for free and its future proof meaning it can handle a heavier construction if I change my mind.



The u-joint can tilt up to30 degrees making it ideal for more than one designs.



So the first thing to do was calibrate and drill holes to the steel plate.



But as we later found out we should have first aligned the plate to the seat (seat + driver center of balance).



No worries tho. We just had to measure more.

And after some more drilling ...at last an early preview...



The tilt point should be located on top near the seat.

The total rise of the seat from the chassis that is is bolted on is approximately 15cm ( which is about the size of the oem seat raisers... this means that the seat doesnt get out of the designed specs)



u-joint Base plate before



and after welding



And finally the chassis steel sheet drilled and bolted with the seat positioned for preview.







Motor chassis with wiper motors under the seat.



closeup on the motor chassis. check the drilling details.



both motors mounted.



Motor chassis placed under the seat.... very tight fit ... (i hope all calculations are right).



Connecting rods (for motor)

Dimensions:
5mm thick
2cm x 6 cm

Drilling:
10mm hole for motor
(6mm) changed to 8mm hole for rod-end bolt.
center to center "C-C" distance is 3,5cm (which can also be reduced to 2,5cm if the motion is slower than expected or the tilt is more than expected).

My final setting was with 2,5cm C-C.

Small C-C distance=fast movement(more torque), small angles of tilt and less stress on the motor
longer C-C = less torque with probably slower movement, biger angles of tilt and more stress on the motor.



for my case i started with 10mm holes for the motor axle to fit.
Then the holes require some more "refining" because their inner surface has to obtain a conical profile.
When fitting the rod, make sure it sits very tightly on the "dents" of the axle.


You have to work patiently with precision to make it a tight fit or else you may end up with premature failure due to insufficient friction between the rod and the axle.

Your rod if properly treated should look like this on the inside.


The easiest way to achieve this is to choose a "softer" metal alloy for the rod (not steel). So when you install it on the motor axle and as you progressively tighten the securing nut, the grooves of the axle will be "registered" on the softer alloys surface. Just don't go for a very weak alloy for this rod because it is a very crucial part of the motion transmission assembly.

self locking nuts (for motor). At a later point these were changed to accommodate the potentiometer pinion axle.


assembled ... side view


and... top view




Make sure that you get some heavy duty parts so your power transmission system is long lasting and maintenance free.


*** EDIT ***
as it turned out

M6 rod-ends that were initially chosen were not as strong as expected

so i ll be testing some M8 rod-ends

M8 rod-ends:


And a close preview.



Link detail

5mm thick L-shape connectors to be fitted under the seat


Just a few steps before finishing with the mechanical part of the project and everything seems to fit as planed.


... some more testing to be sure that travel and tilt performance is as planed.


preparing motor chassis for installation ...


chassis installed


Very tight fit as expected ...


360 degree view



The last component to be fitted was the potentiometer.
I tried to find the best solution amongst many presented in this forum as well as others.

After trial and error i ended up with a custom solution from the r/c racing world(cogs with 45t/17t~2,7:1 ratio).
I intend to use 90* to 100* (+/- 45* to +/- 50* from resting/center point) of rotation of the wiper motor axle, so this equals to 243* to 270* of rotation on the potentiometer axle.
This way i can take advantage of full 270* potentiometer range -and arduino ADC range- for better resolution).


You can see in the details that the cogs have a fine pitch of 48p (small dents) to increase "sensitivity" for finer motion tuning.
These 5mm pinion gears were the closest i could get to fit my 6mm potentiometer axle (OK, some drilling will make them fit) while also having the finest pitch possible for max motion precision.
I couldn't find any 64pitch pinion gears that could fit a 5mm axle.

So i bought some round head nuts to replace the self locking nut on the motor axle.
Some drilling and some treatment with the lathe and:






And this is how it looks:






The potentiometer mounts are pieces cut from a fairly soft pvc drain pipe.
They serve two purposes:
1)They protect the pots in the rare case something goes wrong and the pots go out of travel.
2)They electrically isolate the pots from the chassis.





Module assembled:







...everyone who owns some short of racing/gaming seat can simply put this kind of module under his seat and convert it into a motion simulator.

[sirnoname]

Hi there, good job.
You may look to this setup for a easy pot placement: viewtopic.php?t=1536

[pipis2014]

*** Part 2 ***
Preparing the server psu i had first to remove the security latch (for aesthetic purpose). This is done by unscrewing the necessary screws to get access to the inside. With the help of a big screwdriver work your way slowly to remove the latch. Then screw back the cover carefully.



The next thing was to have the psu running. When the psu is plugged in the fan starts to spin but there is no power output.
With info from this guy:

https://sites.google.com/site/tjinguytech/my-projects/HP47A

I made a “Y” cable to short pins 6-9-10.




But as the psu powered on the fan started spinning at 100% and was too noisy.
So I shorted pins 4-8 to reduce the fan rpm to minimum. The fan still spins up when needed but stays quiet when the psu doesn't require cooling.



High current cable terminals were chosen for safety (and practical reasons ... easy plugging and unplugging)



Heat-shrink is a must ...



Cable-work closeup ...



Psu with one line



And two 12v lines ... one for h-bridge and one for utility circuits (board/motor cooling etc).



The Pololu Dual VNH5019 Motor Driver Shield for Arduino.



It was chosen for a variety of reasons:

1. Low price … 49.95USD (for the features provided)
2. Compact … 2 Motor channels on a single pcb
3. Arduino uno compatible and stackable (allowing for a more compact design)
4. Works with up to 24volt VIN
5. 12Amp continuous output current per channel (30Amps peak per channel)
"short durations (on the order of milliseconds) of 30 A and several seconds of 20 A without overheating"
6. 24Amp continuous output current for single motor connected to both channels(60Amps total peak)
7. Ultrasonic PWM frequency 20kHz for quiet motor operation
8. Reverse voltage protection
9. overvoltage shutdown
10. thermal shutdown
11. Short protection

It looks higher spec-ed than motomonster with a better price.

After some soldering and fitting ... i noticed that a ground pin from Pololu Dual VNH5019 was in contact with arduino usb chassis.


Just to be sure i used some tape on both the arduino



and Pololu Dual VNH5019




Taking a look at the arduino board ... six ICSP pins seem to stand very close to the power input/output of the Hbridge (which sits on top of the arduino board).


After removal


arduino+Hbridge


Just to be safe, a heatsink is a must


After some trimming ... a near perfect fit (360 view)


Thermal pads:
After some search I found that the wiser choice was to use Phobya Thermal Pads.
The pads have a relative high thermal conductivity of about 5W/mk, while also being flexible enough.



Next step the fan installation...
Prefer Brushless fan to avoid future "headaches" ...





In the meanwhile ... the psu for the arduino board arrived.
It is always a good practice to install ferrite snap beads (electronic choke) on both ends to attenuate RF interference.




Heatsink+fan installed. GND & Vin cables cut to fit and soldered.



Installation preview


All things have gone well ... total length under 40cm( 40cm is the total length of the base steel plate).

Final arrangement under the seat

[RacingMat]

Very nice post! It's a good start!
Thanks for your compliment i'm glad you found my tutorial useful!
Yours
Mat

[pipis2014]

***part 3***

(software and testing)

As an extra precaution i used a USB isolator to keep my computer safe-er:


Even though both the h_bridge and the wipers can work with 24volts, i will try some tests with just one psu and see how it goes. To my surprise one psu (12v) seems to be more than enough.

My suggestion is that even if u are 100% sure that everything is correct NEVER test the motors for the first time connected to your chair, or something like this may happen(if your settings are not correct):


Be very careful ... you may read that the wiper motors are not POWERFUL enough BUT they are POWERFUL enough to bend metals and hurt you.

On a second thought ... i decided to keep the rods as the weakest point of the transmission assembly. The reason for that is that i prefer to replace two 10cent rods if something goes wrong ... instead of replacing a more expensive part.

At the same time i also upgraded to higher speced m8 rod-ends.


I started testing with the code provided by Racingmat in his project just altering the lines suggested by Tombo in this post to account for the pinout change for use with the pololu vnh5019 Arduino shield.

Of course some changes need to be made (to trim sensitivity and control) to account for the 2,7:1 ratio between the motor axle and pot axle.




After some testing i am glad to say that the motion is so intense and accurate that exceeded my expectations.
I dont think there is any need for more than 12volts. For the moment at least.
As for the pwm, i was a little frustrated by the buzzing noise.

...BUT...

This is where i felt very happy for choosing the Pololu_Dual_VNH5019 instead of the motomonster.


Pololu_Dual_VNH5019 can be stacked and used with 20kHz ultrasonic pwm...so there a choice from 12v up to 24volts and ultrasonic pwm.


At this point i want to thank Racingmat once more for his great work._

[Petar2]

Hi, I'm starting my project chassis and I've looked at your structure.
Although it seems easy to replicate, if you had asked me some kind of flat and if so if you could share.
I could really use because I think using the same system as you.

Regards and thanks in advance.

[pipis2014]

Feel free to ask whatever you like.
i will be more than happy to answer and give any information i can.

[Petar2]

Well I would like to see some video of your simulator to see how they behave Engines in this position and push rods in this angle.
If you have some PDF / CAD or similar steel sheet Holding Up Engines, as well as how thick is the steel