Super Off-Road Multi 360

[n3wt0n]

Can I run 3 5V LED lights and the three 5V rotary encoders straight off the Arduino? I am just powering it via USB. If not I could just use the power supply for both 5V and 12V.

[processedmeat]

Can I run 3 5V LED lights and the three 5V rotary encoders straight off the Arduino? I am just powering it via USB. If not I could just use the power supply for both 5V and 12V.

I think if its only 3 led lights you should be okay.  I think the max amperage for USB 2.0 is 500mA.

[processedmeat]

Thanks for taking the hires photo of the CPO for me.  Here's my riff on the off road CPO for my super multi 360 wheel cab.

[n3wt0n]

Can I run 3 5V LED lights and the three 5V rotary encoders straight off the Arduino? I am just powering it via USB. If not I could just use the power supply for both 5V and 12V.

I think if its only 3 led lights you should be okay.  I think the max amperage for USB 2.0 is 500mA.

Thanks. I couldn't wait so I started wiring it up to use the power supply. I'm pretty sure it would have been fine but I am 100% confident the power supply can handle it.

Thanks for taking the hires photo of the CPO for me.  Here's my riff on the off road CPO for my super multi 360 wheel cab.

That looks really cool! I like how it pays homage to the original but get the other types of vehicles in there. Nice job!

[n3wt0n]

Okay, its been a bit but I have been moving along nicely on this.

SailorSat said to not leave any analog inputs floating and I believe that was what was causing all the odd behavior on the Arduino Clone because when the new legit Arduino arrived it acted exactly the same way. I can say that the legit Arduino seemed built a lot better and it came with a nice plate for mounting - very convenient.

I was really excited to learn that I could use the original harness that came in the cabinet. I purchased the right sized female connectors to work with the original harness from DigiKey and soldered the wires from the Arduino to the right connector.





I cut a board the right size for all the computer guts and I attached nuts and bolts to the cab that would allow the board to be removed from the machine easily. Then I attached all the computer parts and the arduino and tried to get the wiring cleaned up the best I could.



It fits perfectly where the original PCB would have been mounted.

[n3wt0n]

Later I had a friend help me mount the beast of a TV that I was using for a monitor. From what I could measure, it was going to be a very snug fit. It couldn't have been any better (unless it was an original screen of course). It is a 27" screen while the original was a 25" but since this isn't a restore I wasn't worried about using a bit larger screen.







By this time I have also rewired the cab with the stock harnesses and rewired the control panel with stock harnesses. I only have the center player to finish up but more about that in a later post.

[rob1234]

I am trying to make a similar setup as well from scratch. Does anyone have the measurements for the control panel or the angles the wheels on each side sit?

[rob1234]

Does anyone have the scanned cpo or vectorized custom cpo files available to share?

[n3wt0n]

So this has been sitting for a while but I am back at it and pushing to get it done. 

I had two different sizes of control panel and was going to use the wider version of the control panel because I had all the panels that the wheels attach to but after seeing how much wider it was over the more common one I decided to switch. The only problem was that I was missing two of the three panels. I found the yellow one from a person in my area but ended up getting the other one from a nice guy on KLOV.

Once the control panel parts were sorted I needed a third encoder. Earlier in the thread I was talking about replacing the old, original optical encoders with one of these newer rotary ones.
I ordered one of ebay and it took some time between 30 and 60 days to arrive.



I picked up some new gears from a local electronic supply store and drilled appropriate sized holes in them. One size for the encoder and one for the wheel shaft. The gear on the wheel shaft is held on with a washer and a bolt and for now the gear on the encoder is friction fit but if over time I have any problems with it I can add a dab of glue.



To make the bracket I went to home depot and picked up a piece of thin metal and some small nuts and bolts to build a bracket. Its simple but effective and the metal bracket can be bent slightly to make small adjustments in fitment. When the gears mesh perfectly the setup is almost silent - unlike the 2 older encoder setups I have.





I think that brings us up to speed and into a strange problem that I will discuss in the next post.

[n3wt0n]

I am having trouble getting the rotary encoder from ebay to work. I actually thought the first one was broke so I ordered a 2nd one. Now maybe they are both broke in the same way but they are giving me the exact same results.
This is what the ebay ads say about the wiring...

Features :
400 p/r (Single-phase 400 pulses /R,Two phase 4 frequency doubling to 1600 pulses)
Power source: DC5-24V
Shaft: 6*13mm/0.23*0.51"
Size: 38*35.5mm/1.49*1.39"
Output :AB 2phase output rectangular orthogonal pulse circuit, the output for the NPN open collector output type
Maximum mechanical speed: 5000 R / min
Response frequency: 0-20KHz
Cable length: 1.5 meter
Notice:AB 2phase output must not be directly connected with VCC, otherwise, will burn the output triode
Because different batches, and may not have the terminal
connection:
Green = A phase, white = B phase, red = Vcc power supply, black = V0

I have the red wire hooked up to 5V and the black to ground.
There is also a green wire and a white wire - those are the signal wires and are hooked to the yellow and white wires on the original harness.
There is a also a braided silver shielding or ground around the other wires. I haven't done anything with that.

I get nothing in the game from the encoder and from what I think I understand about encoders is that when the encoder is provided with 5V/ground I should be able to test read between 5 and 0 volts on the signal wires when using a meter (changing as the wheel turns). I get a reading of 0V on those wires even when I rotate the wheel. Am I doing something wrong because I am getting the same thing with both encoders? Another thing I don't understand is if I disconnect the ground wire from the encoder and test using just the wiring harness ground I get a reading of 3.5 on each signal wire even when I rotate the wheel.

For reference, the old encoders test fine and read in this way with a reading between 0 and 5 volts as the wheel is turned on each signal wire.

Any help you could provide would be great. I originally thought using these encoders would be a good idea. I hope I am missing something simple.

[PL1]

I have the red wire hooked up to 5V and the black to ground.
There is also a green wire and a white wire - those are the signal wires and are hooked to the yellow and white wires on the original harness.
There is a also a braided silver shielding or ground around the other wires. I haven't done anything with that.

Sounds like you have the wires hooked up correctly.

The braided shielding is usually tied to ground, but the encoder should still work without it.

I get nothing in the game from the encoder and from what I think I understand about encoders is that when the encoder is provided with 5V/ground I should be able to test read between 5 and 0 volts on the signal wires when using a meter (changing as the wheel turns). I get a reading of 0V on those wires even when I rotate the wheel. Am I doing something wrong because I am getting the same thing with both encoders? Another thing I don't understand is if I disconnect the ground wire from the encoder and test using just the wiring harness ground I get a reading of 3.5 on each signal wire even when I rotate the wheel.

You should see a quadrature waveform when you *slowly* turn the rotary encoder -- less than 2 degrees turn to change between logic high and logic low.

Auto-ranging digital multimeters will not be able to keep up if the logic level changes too quickly.



Try disconnecting both encoder data lines (green and white wires) from the harness and check if they are outputting the expected logic levels.
- If you don't see the expected logic levels, the encoder is probably fried.


Scott

[n3wt0n]

Thanks Scott! I haven't had much luck getting readings on my super cheap meter but considering there are 400 increments per revolution I may not be making small enough changes.

[baritonomarchetto]

Two broken encoders is very unlikely. Try placing a 10Kohm resistor between a signal wire (A or B) and ground and see if you can record, by slowly turning the wheel, a logic state variation  on that line.
You can even use a led (with a 220 ohm resistor in series) to record slow state changes instead of your meter.

[n3wt0n]

Thanks Baron, I will give that a shot. I'm assuming the led should turn on and off with the state changes. Good idea.
Can you explain what adding the 10kohm resistor will do? Just so I have some idea of why I am doing this?  Thanks for your help.

So, assuming that the encoder(s) is working properly, what could be some reasons they aren't registering in mame?

[baritonomarchetto]

When i toyed with taito spinners i had to place those resistors to record state changes. It should be something with active high/ low type of encoder, but i had not time to dig the question.
How are you interfacing the spinner and the pc?
Have you tryed to juice the encoder with an external psu and see what the data lines return? Time to buy a supercheap cinese oscilloscope dude
Check your 5V too...

[n3wt0n]

Okay, completely stumped on this one now. Both encoders show nothing close to what they apparently should read on my meter so now I am focusing on getting a proper reading from either encoder before worrying if they work in the game or not (one step at a time right?).

I have verified that we are getting the 5V required on the red line with my meter and even switched to another power supply that read slightly above 5V (5.20). 
This 5V lights up my LED (with 220ohm resisitor in series).
I then put the LED on the A or B line and slowly turn the encoder and I get nothing.
I also tried putting a 10Kohm resistor between the ground and the A or B line and got nothing on the meter.

I did notice that sometimes my meter can read 0.01 on the data lines. I can turn the shaft and have it go back to 0 and then back to 0.01 but this is nowhere near the 5V it should be topping out at but may indicate that something is broken?

[SailorSat]

Are those encoders active LOW or active HIGH?

You might need some 10k resistors to +5v to get them working in this setup.

[n3wt0n]

Are those encoders active LOW or active HIGH?

You might need some 10k resistors to +5v to get them working in this setup.

I don't know a whole lot about the encoder. I put the description in below. They didn't ship with any additional information.
Where would I use the 10k resisitors? I'm ready to try just about anything.

Encoder 400P/R Incremental Rotary Encoder 400p/r AB phase encoder 6mm Shaf NEW

Description:

100% brand new and high quality
Features :
400 p/r (Single-phase 400 pulses /R,Two phase 4 frequency doubling to 1600 pulses)
Power source: DC5-24V
Shaft: 6*13mm/0.23*0.51"
Size: 38*35.5mm/1.49*1.39"
Output :AB 2phase output rectangular orthogonal pulse circuit, the output for the NPN open collector output type
Maximum mechanical speed: 5000 R / min
Response frequency: 0-20KHz
Cable length: 1.5 meter
Notice:AB 2phase output must not be directly connected with VCC, otherwise, will burn the output triode
Because different batches, and may not have the terminal
　
connection:
Green = A phase, white = B phase, red = Vcc power supply, black = V0
Application:
Is used to measure the rotational speed, Angle and acceleration of the object and the length measurement
Suitable for intelligent control of various displacement measurement, automatic fixed-length leather automatic guillotine machines, steel cut length control, civil measured height human scale, Students racing robots
Quantity:1pc

[PL1]

Are those encoders active LOW or active HIGH?

I have the 600 p/r version of this rotary encoder.

Pretty sure it's the same manufacturer since the label is identical to n3wt0n's except:
- P/N has a "600" instead of "400"
- The barcode at the bottom is slightly different

It is an active low device. (about 4.5v high and 7mv low)

It works with StefanBurger's Illuminated Spinner firmware on an Arduino Pro Micro.

It works without the braided shield being grounded, but it is generally a good practice to ground the shield to protect against electromagnetic interference.


Scott

[n3wt0n]

It is an active low device. (about 4.5v high and 7mv low)

Thanks Scott. if you were to test your encoder with multimeter can you see the variation between 4.5 and 7mv on the data lines?

[PL1]

It is an active low device. (about 4.5v high and 7mv low)

if you were to test your encoder with multimeter can you see the variation between 4.5 and 7mv on the data lines?

Yes.  Those are the actual measurements taken with the rotary encoder connected to a Pro Micro.

For probes, I used straight-pins pushed into the Dupont housing from the wire side.


Scott

[pixel]

The Pins may have a clearcoat on them.  A lot of MFG. are now using lower quality metal blends, and they clearcoat them to
keep them from easily rusting...  and or just to keep them from rusting on the cargo ships, from the Factory origins.

 Take some sandpaper to them,  and pre-test them for continuity, and resistance, top to bottom.


 As for your gears... I suggest making wooden shaft collars,  that you can glue on the gear face,  and additionally,
drill and screw in two tiny woodscrews from the backside of the gear, into the wood collar.

 The only tricky part is centering the collar.   You can draw two cross-lines,  and center-punch a dot on a large dowel section.
It would be easier if the gears still had the small center-holes.. and you could use a nail or small shaft,  as a centering pin.

 Flattening the tiny raised center hub area,  for the Inside surface... via sanding / chiseling.

 Drill a small enough hole for the centering pin to go down about 3 mm,  and then lock the gear in place to the dowel.
Glue / Hot-Glue  could be put on the inside faces.. then pressed together.  Or, glue from the gear side, on the exposed
gear holes.

 Lock the gear completely,  with narrow and tiny woodscrews, on the far edges of the gears spokes.  At least two,  should be fine.

 Now you can remove the center-pin.

 Lock the dowel in a drill-vice,  over a scrap 2x4 over-drill section.

 You could either drill from the dowel side down through the gear...

 or

 Drill from the gears center hole,  down through the dowel.

 or

 If you do not need the shaft to go all the way through the gear... you could drill a certain depth,
that only goes through part of the dowel... leaving the gear largely as-is.

 I drilled mine completely through... for a homebrew Starwars Yoke.  (Pass-Thru shaft)


 Finally... you can now drill and Thread Tap,  a set-screw hole into the collar,  that will be used to lock the
shaft in place.   Make sure its a little snag,  as wood isnt as robust as metal.  You could use a long machine screw,
depending on how deep your collar is... or you can pick up a box of various sized set-screws from harbor freight.

 IMO, Gluing a shaft in place, is not a good option.  Both in terms of maintenance, and in longevity + accurate performance.
Shaft collars are not that difficult to make,  and can be made to be very Robust.

[n3wt0n]

Scott, Based on your meter readings and my inability to get any readings at all I am back to thinking that my encoders are both not working properly if at all. Thanks for the information though - It is much easier to troubleshoot knowing that a very similar product can produce the desired results.

Pixel, I am using bare wires that I stripped myself. There aren't any contacts or pins involved right now but that's for the heads up. As for making collars, I appreciate the advice and if I have problems with the gears not staying in place I will revisit with your suggestions in mind.

[n3wt0n]

necrobump of my old thread  ... Has anyone used this sketch on an Arduino on a Windows 7 PC? I know it was originally made to work with XP... just wondering if I am going to run into any problems.

[n3wt0n]

Okay, having a little trouble with the arduino and the new computer.

I cannot get any button presses to register anywhere. Nothing from the coin door and nothing from the nitro buttons. I have checked for continuity from the buttons, through the harness, all the way back to the arduino. The buttons are set to be different mouse presses but are not registering as anything.

All three steering wheels move the cursor left and right across the screen.
All three pedals show up and can be calibrated in the game controllers properties tab.

This is all outside of MAME in windows. Any suggestions?

Thanks.

[Howard_Casto]

Ok it's sounding more and more like you are having coding issues and not hardware issues.  From the hardware end it's pretty dang hard to screw up a switch.   Maybe post your sketch in plain text and we can check it for you.   


*edit*   Nevermind, looks like you are using KADE software.   I can't help you there.

[PL1]

looks like you are using KADE software.

He's using SailorSat's sketch on an Arduino Mega2560 clone.

(This is a link to SailorSats github repository)
https://github.com/SailorSat/daytona-utils/tree/master/OffroadArduino

The buttons are set to be different mouse presses but are not registering as anything.

So you're getting the mouse X-axis movement from the three wheels, but not the expected mouse click outputs when you ground pins 22 - 28, right?

Are you getting the expected gamepad button outputs when you ground pins 31-37?

The sketch is reading form PortA and PortC which map to 22 to 37.

Pin 22 to 29 are mapped to various mouse buttons.
Pin 30 to 37 are mapped to the gamepad.

22 = Mouse 1 RightClick (Coin)
23 = Mouse 1 LeftClick (Nitro)
24 = Mouse 2 RightClick (Coin)
25 = Mouse 2 LeftClick (Nitro)
26 = Mouse 3 RightClick (Coin)
27 = Mouse 3 LeftClick (Nitro)
28 = Mouse 1 MiddleClick (Service)
29 = -

30 = -
31 = GamePad 4
32 = GamePad 3
33 = GamePad 7
34 = GamePad 2
35 = GamePad 6
36 = GamePad 1
37 = GamePad 5

Mouse 1 is on 2/3
Mouse 2 is on 21/20
Mouse 3 is on 19/18

Scott

[PL1]

I cannot get any button presses to register anywhere. Nothing from the coin door and nothing from the nitro buttons. I have checked for continuity from the buttons, through the harness, all the way back to the arduino. The buttons are set to be different mouse presses but are not registering as anything.

At the risk of sounding like Captain Obvious here, are you sure the wires are connected to the microswitch COM and NO tabs?   


Scott

[n3wt0n]

So you're getting the mouse X-axis movement from the three wheels, but not the expected mouse click outputs when you ground pins 22 - 28, right?

Are you getting the expected gamepad button outputs when you ground pins 31-37?

Scott

I have never used the button outputs as part of the layout - just the mouse clicks - but I can test those easily enough to see if the buttons dedicated to the gamepad are registering. I have a feeling it's something in the PC related to multiple mice but I really have no idea at this point. If the game buttons are being picked up I can likely switch to use those for 6 of the inputs. Good idea to test those.

At the risk of sounding like Captain Obvious here, are you sure the wires are connected to the microswitch COM and NO tabs?   

Scott

I am going to go through and check everything but the cabinet was working nearly 100% with this setup on a different PC (and different operating system). I didn't rewire any of the buttons in the meantime but it's good to go back to the basics so I appreciate the reminder to check the little things as well.

I will get to this tonight or tomorrow hopefully and report back. Thanks for your help.

[PL1]

Looking at a Mega 2560 pinout, there's one thing about SailorSat's pin list that some people might find confusing.

29 = -

30 = -

I'm pretty sure that by labeling digital input pins 29 and 30 as "-" she meant that they are undefined/not used, but someone could misinterpret that "-" label as negative/ground.   
- There are several ground pins next to pins 52 and 53, if this is what caused your buttons to not work.

I have a feeling it's something in the PC related to multiple mice but I really have no idea at this point.

Button clicks from multiple mice should still register in the OS, with Windows adding all of the matching inputs together.  i.e. A right-click via pin 22, 24, or 26 will register as a mouse right-click in Windows.

You will need "multimouse" enabled in mame.ini so MAME keeps all of the mouse inputs separate.  i.e. One wheel controls mouse1 X-axis, another wheel controls mouse2 X-axis, mouse1 right-click separate from mouse2 right-click, etc.


Scott

[baritonomarchetto]

Why a multi mouse device? Wasn't is easier (maybe lighter on a processing point of view?) to use a single mouse and set each axis (X, Y and center wheel) one for each wheel?

[PL1]

Why a multi mouse device? Wasn't is easier (maybe lighter on a processing point of view?) to use a single mouse and set each axis (X, Y and center wheel) one for each wheel?

Good question.

With 3 optical axes (steering wheels) and 3 analog axes (pedals) on a single Arduino board, there's no way to get around making it a composite mouse and gamepad HID device.
- If you're already making a composite HID device with one mouse and gamepad, it's probably not much more work to add two more mice and gamepads.
- If you're making a composite HID device with two mice and gamepads to avoid the potential Z-axis issues mentioned below, it's even less additional work to add a third mouse and gamepad.

Not sure how different the processing overhead is for the same total number of inputs defined as a single 3-axis mouse+gamepad vs. three single-axis mice+gamepads.   
- If the Mega can handle either version without backspin or lag, it shouldn't matter which approach is used, right?

Two considerations that may have led SailorSat to avoid the mouse and gamepad analog Z-axes:
- The Windows 4:1 scroll wheel issue.
- An unpressed Z-axis analog pedal causes constant down-scrolling in some menu systems.


Scott