Jump to content

Photo

Refurbishing steering wheel controller with new electronics?


62 replies to this topic

#26 Swami ONLINE  

Swami

    Dragonstomper

  • 835 posts
  • Location:Twin Cities, MN

Posted Mon Jun 26, 2017 6:34 PM

i use those little rotary encoder in my video

https://www.ebay.ca/itm/171906808593

the problem is the number of pulse in one rotation

this is why i need a doubler circuit and a way to be sure to not fry the fragile cv controller chip...

The module 2 driving controller has 40 ppr on the outer row and another 40ppr from the inner row, 90deg out-of-phase (80 total?). What does the one you used have, it looks like 30, but is that 30 or 30x2, including both phases?



#27 BigO ONLINE  

BigO

    River Patroller

  • 4,266 posts
  • Location:Phoenix, AZ

Posted Mon Jun 26, 2017 7:46 PM

The Indy 500 controller has a 16 position rotary binary switch. It appears to use multiplexed signal over four pins to provide the 16 outputs. The side button works as the gas pedal (pin 6), although the spinner is not natively functional with the CV.

 

Here is the interupter flywheel: It has ~40 slots/row, but I don't know if the staggering of the two rows provide some type of multiplexing or not, regarding position within a " game life". 

 

 

 

I don't believe the two rows provide any additional resolution. It has to do with how the two emitter/detector pairs are arranged. What they did was offset the two sensor pairs radially with reference to the encoder wheel. One sensor set is interrupted by one row, and the other is interrupted by the other row. It lets the job be done with less precision manufacturing.

 

Other encoder wheels have one row of shutters/interrupters/whatever. The actual sensors are spread out axially so that they are affected 90 degrees out of phase from one another with respect to the interruptor/window. They happen to be very close to one another but could easily be almost straight across the circle from each other. Open up an Atari driving controller to see an easily visualizable quadrature encoder.

 

The same exact thing is happening in the more typical one row encoder wheel. It's just harder to visualize what's happening. This two row design actually matches up visually well with this diagram I put together to help me understand quadrature encoding/decoding:

Quadrature Decoding Method

 

 

40-ish slots means forty half cycles per row yielding a total of 40 full cycles per revolution. It takes two bits to determine both motion and direction. The Atari controller has only 4 full cycles per revolution (16 state changes, 4 states per full cycle = 4 cycles).

 

If there are actually 40 then there are 40x4=160 detectable motion/direction events per revolution of the wheel.



#28 Swami ONLINE  

Swami

    Dragonstomper

  • 835 posts
  • Location:Twin Cities, MN

Posted Mon Jun 26, 2017 10:32 PM

Something like this for the colecovision would be nice. These are now of the unicorns.

 

http://segaretro.org...mer_Turbo_Wheel



#29 chart45 OFFLINE  

chart45

    Dragonstomper

  • 748 posts

Posted Tue Jun 27, 2017 5:43 AM

The module 2 driving controller has 40 ppr on the outer row and another 40ppr from the inner row, 90deg out-of-phase (80 total?). What does the one you used have, it looks like 30, but is that 30 or 30x2, including both phases?



30

Edited by chart45, Tue Jun 27, 2017 5:44 AM.


#30 chart45 OFFLINE  

chart45

    Dragonstomper

  • 748 posts

Posted Tue Jun 27, 2017 6:50 PM

ok today i did more test so to work properly we need external power Source to have a stable signal...

with out power source it work with turbo but very bad with game like mindwall...

i found a little doubler pcb on ebay now its waiting game...

#31 Swami ONLINE  

Swami

    Dragonstomper

  • 835 posts
  • Location:Twin Cities, MN

Posted Tue Jun 27, 2017 7:13 PM

 

 

 

I don't believe the two rows provide any additional resolution. It has to do with how the two emitter/detector pairs are arranged. What they did was offset the two sensor pairs radially with reference to the encoder wheel. One sensor set is interrupted by one row, and the other is interrupted by the other row. It lets the job be done with less precision manufacturing.

 

Other encoder wheels have one row of shutters/interrupters/whatever. The actual sensors are spread out axially so that they are affected 90 degrees out of phase from one another with respect to the interruptor/window. They happen to be very close to one another but could easily be almost straight across the circle from each other. Open up an Atari driving controller to see an easily visualizable quadrature encoder.

 

The same exact thing is happening in the more typical one row encoder wheel. It's just harder to visualize what's happening. This two row design actually matches up visually well with this diagram I put together to help me understand quadrature encoding/decoding:

 

 

 

40-ish slots means forty half cycles per row yielding a total of 40 full cycles per revolution. It takes two bits to determine both motion and direction. The Atari controller has only 4 full cycles per revolution (16 state changes, 4 states per full cycle = 4 cycles).

 

If there are actually 40 then there are 40x4=160 detectable motion/direction events per revolution of the wheel.

So for one of the optical rotary encoders with quadrature output like I posted above in a previous post, how much of the driving controller circuit board does this cover? I am thinking it is the area with the two E/D pairs and the IC to the lower right.



#32 BigO ONLINE  

BigO

    River Patroller

  • 4,266 posts
  • Location:Phoenix, AZ

Posted Wed Jun 28, 2017 1:12 AM

 

 

So for one of the optical rotary encoders with quadrature output like I posted above in a previous post, how much of the driving controller circuit board does this cover? I am thinking it is the area with the two E/D pairs and the IC to the lower right.

Transistors were never my strong suit, but I'm guessing that part of the circuit is a voltage regulator.

The IC just looks to be buffering the output of the optical encoder.

The diodes on the outputs of the IC's I assume are there to provide some sort of isolation, but I'm not sure exactly what's going on inside the console. A guess would be that they're there because the console at some time tries to apply a strong output signal on those pins. 

 

Other than those sorts of implementation details, yeah, the guts of the thing look to be the optical elements and the IC to beef up the outputs from the optical detectors. It looks like the software would need to be set up to read the quadrature encoded signals (Gray code counter).



#33 Swami ONLINE  

Swami

    Dragonstomper

  • 835 posts
  • Location:Twin Cities, MN

Posted Fri Jun 30, 2017 4:43 PM

i use those little rotary encoder in my video

https://www.ebay.ca/itm/171906808593

the problem is the number of pulse in one rotation

this is why i need a doubler circuit and a way to be sure to not fry the fragile cv controller chip...

Hi Chart45. Since you've got the proof of concept going, you would know the transistors and what controller output to the console is required. So, for your consideration, BigO and I have some other options here if you think they will help or make it more rugged while keeping costs reasonable. Optical switches and optical rotary encoders with and without quadrature detection. Let us know how the stuff goes, good luck with it. I'm going to look into physical interface mods (steering wheel, etc.).



#34 Swami ONLINE  

Swami

    Dragonstomper

  • 835 posts
  • Location:Twin Cities, MN

Posted Fri Jun 30, 2017 5:14 PM

These are some easy, generally budget-friendly DIY ideas for creating a larger, higher and possibly adjustable tilt for the module steering wheel, if so inclined. I have not tested any of these yet, just done some browsing:

 

I would suggest using a small suction cup to pull off the ColecoVision decal from the steering wheel center to get at the screw holding the wheel on, as trying to dig it out will probably mangle it.

 

Steering wheel ideas: These are "deep-dish", so they bring the wheel further out while improving access to the hand controller with the 12" steering wheels.

 

https://www.amazon.c...d=22ATZ0Z2ATW48

 

https://www.amazon.c...HRT4FD5A91ZKGWB

 

https://www.amazon.c...d=22ATZ0Z2ATW48

 

https://www.amazon.c...d=22ATZ0Z2ATW48

 

Steering Wheel extension:

 

https://www.amazon.c...d=22ATZ0Z2ATW48

 

https://www.amazon.c...d=22ATZ0Z2ATW48

 

Something like this that could angle the module would be nice, if could also hold it in place, but it is a little spendy and would require straps...and it only comes in aqua and brown :(

 

https://www.amazon.c...06XR6GJ3S?psc=1

 

So far, not counting an adjustable wedge, this would run $20-$35, depending on wheel choice. Of course you might be able to pick up a couple inches of PVC pipe and end-caps for a couple bucks at home depot, which basically makes it the cost of the wheel + $2.



#35 chart45 OFFLINE  

chart45

    Dragonstomper

  • 748 posts

Posted Sat Jul 1, 2017 2:23 PM

as you can see my english is not really good

so if you look at the diagram the transistor are on off switch for the 6v batt... when the coleco is on the orange wire is high so transistor let the power from the battery pass trough... the coleco need both signal on purple and white wire its the way he knows if you turn left or right... my goal is more to have a padle than a steering wheel but both project could be achive at the same time... the only thing we need its a small rotary encoder with enough complet cycle in one rotation... it should be easier in the steering wheel project since you have more space to work...you can use gear to change the ratio and have more complet cycle in one complet rotation

Edited by chart45, Sat Jul 1, 2017 2:29 PM.


#36 chart45 OFFLINE  

chart45

    Dragonstomper

  • 748 posts

Posted Wed Oct 18, 2017 6:38 AM

i found a good rotary encoder with more cycle than the steering wheel and its small enough to build a paddle... this encoder have 90 pulse by rotation so more than the steering wheell 72 pulse by rotation and easy to find

here the resulte of the labscope test i will try to build a simple proto paddle and share result



#37 chart45 OFFLINE  

chart45

    Dragonstomper

  • 748 posts

Posted Fri Oct 20, 2017 5:51 PM

ok here a quick demo



#38 Pixelboy OFFLINE  

Pixelboy

    Quadrunner

  • Topic Starter
  • 7,699 posts
  • Location:Montreal, Canada

Posted Fri Oct 20, 2017 6:11 PM

Wow, this is great! I suppose your paddle controller requires outside power?



#39 chart45 OFFLINE  

chart45

    Dragonstomper

  • 748 posts

Posted Fri Oct 20, 2017 6:41 PM

Wow, this is great! I suppose your paddle controller requires outside power?


yes like the steering wheel

#40 Swami ONLINE  

Swami

    Dragonstomper

  • 835 posts
  • Location:Twin Cities, MN

Posted Mon Oct 23, 2017 6:46 PM

yes like the steering wheel

Pretty slick. Is this all in the box you are holding, or is it also wired to the board of the driving controller?



#41 chart45 OFFLINE  

chart45

    Dragonstomper

  • 748 posts

Posted Tue Oct 24, 2017 3:58 AM

Pretty slick. Is this all in the box you are holding, or is it also wired to the board of the driving controller?


for the first test I only used the rotary encoder direct to the colecovision .. to be sure to not damage the colecovision controller chip i think i will use the steering wheel pcb desing...

#42 Swami ONLINE  

Swami

    Dragonstomper

  • 835 posts
  • Location:Twin Cities, MN

Posted Tue Oct 24, 2017 7:54 AM

for the first test I only used the rotary encoder direct to the colecovision .. to be sure to not damage the colecovision controller chip i think i will use the steering wheel pcb desing...


Just curious. Is it possible to give a quick assessment of whether the driving controllers board components are common enough one could build a driving controller board with your device to put inside another racing wheel like a thrustmaster without cannibalizing an existing Colecovision driving module?

#43 chart45 OFFLINE  

chart45

    Dragonstomper

  • 748 posts

Posted Tue Oct 24, 2017 8:08 AM

Just curious. Is it possible to give a quick assessment of whether the driving controllers board components are common enough one could build a driving controller board with your device to put inside another racing wheel like a thrustmaster without cannibalizing an existing Colecovision driving module?



here the diagram...

http://www.le-grenie...n-module-2.html

there is nothing you cant find

Edited by chart45, Tue Oct 24, 2017 8:10 AM.


#44 Swami ONLINE  

Swami

    Dragonstomper

  • 835 posts
  • Location:Twin Cities, MN

Posted Tue Oct 24, 2017 8:31 AM

here the diagram...

http://www.le-grenie...n-module-2.html

there is nothing you cant find

Okay, thanks.



#45 chart45 OFFLINE  

chart45

    Dragonstomper

  • 748 posts

Posted Tue Oct 24, 2017 9:25 AM

Okay, thanks.


i did a quick breadboard for the ic part and it works

https://youtu.be/xSgy-dbJ9b0

#46 Pixelboy OFFLINE  

Pixelboy

    Quadrunner

  • Topic Starter
  • 7,699 posts
  • Location:Montreal, Canada

Posted Tue Oct 24, 2017 10:22 AM

Just curious. Is it possible to give a quick assessment of whether the driving controllers board components are common enough one could build a driving controller board with your device to put inside another racing wheel like a thrustmaster without cannibalizing an existing Colecovision driving module?


Just a note: There are so many broken ColecoVision steering wheels out there that finding one to cannibalize (or in other words, to make functional again) is really a non-issue.

#47 Swami ONLINE  

Swami

    Dragonstomper

  • 835 posts
  • Location:Twin Cities, MN

Posted Wed Nov 1, 2017 6:09 PM

Just a note: There are so many broken ColecoVision steering wheels out there that finding one to cannibalize (or in other words, to make functional again) is really a non-issue.

One of the three EM2s I bought was broke, but it was due to batteries probably left in it for 25 years ...corroded board. Are most dead ones due to the optics failing?

#48 chart45 OFFLINE  

chart45

    Dragonstomper

  • 748 posts

Posted Wed Nov 1, 2017 6:38 PM

ok here a little update .. i used an atari plug n play and installed a rotary encoder inside ... it works with 4 AA battery...



#49 Swami ONLINE  

Swami

    Dragonstomper

  • 835 posts
  • Location:Twin Cities, MN

Posted Wed Nov 1, 2017 7:01 PM

ok here a little update .. i used an atari plug n play and installed a rotary encoder inside ... it works with 4 AA battery...
 

 

That's pretty awesome. I don't think the Colecovision ever had a paddle controller, and with replaceable hardware to boot.



#50 BigO ONLINE  

BigO

    River Patroller

  • 4,266 posts
  • Location:Phoenix, AZ

Posted Wed Nov 1, 2017 7:50 PM

I believe the super action controller may have had an encoder driven by a left-right rolling wheel.




0 user(s) are browsing this forum

0 members, 0 guests, 0 anonymous users