LearnDriver's 2DOF SCN5 Project

Hi all,

After a few weeks of considering my options and looking at the financial picture, I decided to take the leap and start with my own motion project, inspired by you guys.
I'm trying to document it as well as possible, who knows it might benefit someone else.
After I read a lot of topics I came to this parts list:

Code:

+-------------------+----------------------+-----+
+ Supplier          +        Part          + Qty +
+-------------------+----------------------+-----+
+ simxperience.com  + SCN5                 +  2  +
+-------------------+----------------------+-----+
+ display.nl        + Meanwell S-100F-24   +  1  +
+-------------------+----------------------+-----+
+ x-simulator.de    + Cardan joint         +  1  +
+-------------------+----------------------+-----+
+ ebay.com          + 4-point harness      +  1  + 
+-------------------+----------------------+-----+
+ mouser.com        + USB-RS485 adapter    +  1  + 
+                   + 895-USBRS485WE1800BT +     +
+                   + SCN5 Power connector +  2  + 
+                   + 798-DF1B-14DES-2.5RC +     +
+                   + RS485 connector      +  2  + 
+                   + 571-172167-1         +     +    
+-------------------+----------------------+-----+
+ wurth.nl          + Balljoints M10xP1.25 +  4  +
+                   + Rubber dampers       +  2  + 
+-------------------+----------------------+-----+
+ localhost         + 2k2 Ohm Resistor     +  2  +
+                   + 120 Ohm Resistor     +  2  +
+                   + 40x40x2 steeltubes   + 15m +
+                   + Recaro Seat          +  1  +  
+-------------------+----------------------+-----+

Am I forgetting/missing something? Or is a part just plain wrong?
Haven't incorporated a killswitch in the design yet, as from what I've read most designs here are wrong according to bvillersjr. Unfortunately I havent found the proper way to make one.
Also can someone tell me if the USB-RS485 cable I plan to use us a proper one? http://nl.mouser.com/ProductDetail/FTDI ... YwQw%3d%3d
My proposed design (oldschool drawn with a pen and a ruler ) Measurements are in CM

[img600px]http://www.basvansambeek.com/images/2DOFSCN5.jpg[/img600px]

--Bas

Finally after 2,5 months of waiting for parts and various other things getting in the way like work, my sim is nearing completion.
Tonight I've bolted everything together and came to a small flaw in my design.


Measurements are balljoint to balljoint.

In the descriptions on the forum it says that the distance between topballjoints should be around 11.25 to 12 outer dimension from top-side ball-joint to ball-joint and for the bottom 30.50 to 32 outer dimension from bottom-side ball-joint to ball-joint.
Because I've accidentaly took this as the dimensions for the bar itself, and didn't take the measurements of the balljoints into account the 2 bars are 8 cm or 3.14 longer as planned.
The top bar is 29.5cm (11.61), dampening bar is 80cm (31.49) (excluding balljoints). Measurements in the image are between balljoints.

So as far as my reasoning goes, this error isn't affecting the forward motion as that angle stays the same and only the roll is affected because the bar is longer.
Because the bar is longer, I will now achieve a lower max roll but also a lower force needed to roll.

So what will this mean for the actuators? And, as I don't have regular acces to welding tools, how urgent would the need be to fix this (regarding actuator stress etc)?








Sorry for the amazing cellphone quality, batteries of the proper digicam were empty.

LearnDriver,

There are many variables that will affect the performance of the actuators so I would test these angles usning the test plugin and see if they can sustain your weight in your specific configuration.


....Mike

Been a while since I completed my sim. Sim has been running perfectly for the last 2 months, untill recently.

First the good news:
[img600px]http://www.basvansambeek.com/photo/sim/PICT9241.JPG[/img600px]
[img600px]http://www.basvansambeek.com/photo/sim/PICT9246.JPG[/img600px]
[img600px]http://www.basvansambeek.com/photo/sim/PICT9248.JPG[/img600px]
[img600px]http://www.basvansambeek.com/photo/sim/PICT9251.JPG[/img600px]
[img600px]http://www.basvansambeek.com/photo/sim/PICT9252.JPG[/img600px]
[img600px]http://www.basvansambeek.com/photo/sim/PICT9254.JPG[/img600px]
[img600px]http://www.basvansambeek.com/photo/sim/PICT9256.JPG[/img600px]

Because my sim room is also the place where I sleep, watch tv, work, etc, I've chosen for a modular design so everything can be moved to the side of the room, nice and compact.
You got the back of the sim (the seat and motion parts), steering 'column' and the pedal set.
The pedals fit between the steering rack and therefore can't move.

The Redbull F1 (sorry :brows: ) inspired side panels are kept in place with velcro tape, so if I'm bored I can switch them for eg. Ferrari ones.

If anyone is interested I can scan the drawings I've made.

Now the sad part of the story.

Everything was working great, untill suddenly X-Sim wouldn't recognize my SCN5's anymore.
My hardware setup is as follows:
Power is provided by a Meanwell S-100F-24.
USB-RS485 is the following FTDI based connector:

The connection of the adapter to the SCN5's was provided by an UTP CAT5E cable, with a P4 connector soldered to the end so they'll fit in nicely into the SCN5's without soldering to the SCN5's itself, voiding its warranty.

I'm using a single adapter, proving both SCN5's with data. Because the data cable was over a meter (=3.28ft) long I've added a resistor.

The first thing I did was to check the cables and the USB-RS485 adapter.
When testing with a multimeter all my cables didn't seem to be broken, so I assumed no problems there.
The USB-RS485 adapter seemed to work as well when measuring the data cables with an oscilloscope as some data was being sent.
To be sure I'd ordered a new adapter, but that one was DOA (chips were getting hot, still waiting for a replacement atm).

So, back to the cables.
Even though the cables seemed ok, I've changed the UTP twisted pair cable for ones from an USB cable.
And magically one SCN5 started living (still haven't got a clue why the UTP cable suddenly stopped working and now it does work with the stipped USB cable...).
So I tested the same cable on the 2nd SCN5 and nothing.

So the problem I'm having is that with the same cables etc one SCN5 is working and the other isn't.
I've removed the plastic cover from the SCN5 and everything seems to be ok, no fried circuits what so ever. The proc gets a little warm but afaik thats ok.
When turning on my power supply I can hear the actuator 'grip', like they used to do.

So I'm running out of idea's. All I can think of is that there is something with the RS485 unit on the actuator.

Love to hear your ideas,

Best regards,
Bas

Probably a voltage spike via the power supply, which killed the ILK transistor.
There are some other people, XSim and Frex users, whose SCN5 got killed that way.
However, there is most likely a workarround.
You will have to disable the ILK mode by using the Dyadic SCN5 configuration tool.
After setting the actuator to air-mode 1, it should work again.
There is a thread regarding this issue already.
post14382.html?hilit=air%20mode%201#p14382

regards

The sim is turning out very nice. Good work.

I believe the Meanwell is Over-Volt and Over-Amp protected, so I would be surprised if you got any kind of power spike related problem unless you were plugging / unplugging things improperly or shorted something out.

Some other things you can try:

-Deleting all the X-Sim registry keys (C:\Program FIles\X-Sim\Other Stuff\RegistryCleaner.exe) Clear all scetions
-Changing the USB ports you are using
-Starting X-Sim in debug mode and reporting back the error messages (this will verify if you have the ilk issue)

:

First of all, thanks for the quick reply.
I'd left the actuators at a friends house with more tools so he could do some checks, unfortunately he had little time, so I picked them up again this week.

I've put X-Sim in debug mode, and it gives no response what so ever. I downloaded the serial port analyzer and I get the same result. Data gets sent to the actuator, but he doesn't return a response. I know the actuator gets power because you can hear him grip when I connect the power and hear him buzz. With the same wires etc the other actuator has no problem being recognized and immediately extends to the default position.

So what should my next step be?

My friend thinks it might be something with the RS485 chip on the actuator itself, but in order to examine that he needs to solder some wires to the chip. So I decided to ask you guys first before I void my warranty :?

It could be a few things

1-RS495 adapter is fried (has happened to multiple people with the cheap off the shelf adapters

2-The serial A & B wires are crossed

3- If you accidentally connected the power to one of the communications pins on the SCN5 when making your wires, you will blow the communications circuitry. It is only designed for 5V. The actuator will still make the gripping noise because power is sent to the motor, but no commuication will occur because the chip is fried.

4-Power supply was not adequate. Can you show us a picture of it.

Well I don't think it's 1 and 2, as it works on the other actuator with no changes.
3 would be hard to do, as they are separate cables.

This is my current setup (and the setup I've always used before).



[img600px]http://www.basvansambeek.com/sim/PICT0017.jpg[/img600px]

[img600px]http://www.basvansambeek.com/sim/PICT0038.jpg[/img600px]

Your power supply has protection according to the manufacturers PDF.

You can try egoexpress' suggestion of using the Dyadic tools to set the air mode, or if you bought it less than a year ago, you can ship it in for repair.

I got them from you in March I believe. Will drive to the stores tomorrow to get my own multimeter (measured most of the stuff at my friends house) and will see if that returns me with any obvious solutions. If not I'll ship them back to the States :'(

Hi,

I was looking at you sim and noticed that you don't have any extra rod's between the seat frame and the bottom frame to keep the seat frame from turning left or right.
As you can see in the picture there are 2 rod's going from the botom frame to the seat frame. How did you solve this?

Hey Mexwall,

I don't exactly know the answer but I think it is because he is using a universal joint instead of the Frex-style rubber bushing.

Don't hold me to it.

Gus

LearnDriver, how did things turn out? I'm new here, so am getting to this thread late, and am curious to know how you got the SCN5s to come alive again.

Over the years here, I belive the consensus has been that a universal joint has been more cost effective and less prone to use configuration error that result in actuator failure and poor quality motion due to the small amount of uneven rotation. If you search the forums for universal joint, you are sure to find many posts of varying types and prices.