Power Pad problems

[Ataritard]

I thought it would be cool to modify a NES Power Pad to work on an Atari 2600 (and completely ridiculous at the same time. I really just wanted to see if it could be done). Well I've run into a big roadblock so here we are.

 

If you have never taken a Power Pad apart (and Why would you, for that matter?) here is how it works: There are 12 traces going to the individual buttons and one common ground. The problem is that these are carbon or graphite (I don't know really) traces on plastic and they have an inherent amount of resistance. (Round trip about 5k ohm if I'm measuring right.) BTW I've removed all the chips and resistors.

 

My question is there any way to "power through" this resistance so that the Atari can recognize a connection? Thank you.

[Rex Dart]

Would it be possible to just run some wires up from the "buttons" & bypass those carbon traces? Maybe route those wires along the edge of the pad so they don't get stomped on. That ought to give less than 5kΩ, at least.

Edited June 19, 2012 by Rex Dart

[Ataritard]

Thanks Rex, That would work but I'd rather not cut it up to get the wires in it, and I don't know how I could replace the 6 inch diameter "buttons" that it uses.

 

I am using a cord with all 9 wires coming from the Atari so I have access to 5 volts if you think that would help.

[Stephen Moss]

According to http://nesdev.parodi...om/powerpad.txt the power pad is a somewhat messy Parallel in Serial out design which does not make things easy for you. Although why you would want to do this anyway alludes me as to move diagonally and fire at the same time surely you would have to be able to stand on three buttons at once which would be a little tricky.

 

If the connections you mentioned are like this...

 

Trace 1 -- 5k Track resiscance --- Switch --- Common Ground

Trace 2 -- 5k Track resiscance --- Switch --- Common Ground

 

and so on then connecting +5 to the top of each trace and taking the output from the non ground side of the switch should work as that is the same as the 2600, although you should put the switch output through a buffer chip before sending it to the 2600 to prevent any possible issue with the parallell resistances, plus it cleans up the signal.

 

However, if the connections you mentioned are like this...

 

Trace 1 -- Switch --- 5k Track resiscance --- Common Ground

Trace 2 -- Switch --- 5k Track resiscance --- Common Ground

 

and so on then you could either...

 

1) take the output from the non 5K end of the switch and connect it into the 2600 although for that to work you would have to replace the 2600's internal pull up resistors with some of a value of 20K in order to get the voltage down far enough however, doing so will probably mean that you then have insufficient current to drive the logic inputs or

2) connect +5 to the top of each trace and take the output from 5K end of the switch, passing it through an inverting buffer chip before sending it to the 2600.

 

I think these are the only viable options now, at least they are if you want a simple low components count solution.

Although it can get a bit messy as you end up with the Pad plugged into a adaptor plugged into the 2600 I would have been tempted to of left it intact and used a PIC to interface between the two.

[Ataritard]

Thank you Stephen, I seriously doubt this will replace my CX-10 or modded NES controller as my go to controller for the Atari, even if I get it going, it's just a fun thing to do. I have considered the problem of how to push three buttons at the same time so my plan was to wire the fire button to a handheld plunger.

 

The resistance is actually on both sides of the switch (if by switch you mean button). From the circuit board at the top there is, let's say 2 or 3k ohms of resistance to the four buttons, and say 2 or 3k ohms back up to the board. What you said about a simple low components count solution is a priority as this is just for fun.

 

As for a buffer chip, My question now is would 5V damage my Atari if it were connected to Ground or any of the directions without a buffer chip?

[Stephen Moss]

Yes, Switch = Button

 

Connecting +5V to the VCS input without a buffer chip should not damage the VCS, the reason I suggested using one is best explained with by referencing the attached image.

 

The top circuit (VCS + Joystick) shows the standard Atari controller interface, when the fire/direction switch is open no current flows through the internal 5K resistor (R1), consequently no voltage is dropped across the resistor and the input is pulled up to +5V via R1. When the switch is closed current flows through R1 to ground, consequently 5V is then dropped across R1 and the input is pulled down to ground (0V), thus providing nice easy levels for the logic chips to deal with.

 

The bottom circuit (VCS + Power Pad) shows one way in which the power pad could be connected to the VCS (note: R2 is now the VCS internal pull up resistor, R3 & R4 represent the power pad track resistance). The problem here is that R2 and R3 are now connected in parallel and so effective act as a single resistor.

 

If I split the difference between the resistance figures stated in your last post and thus assume the 5K reisistance of the Power pad track is split evenly (2K5 either side of the switch) then the effective resistance of the VCS's internal 5K pull up resistor is now 2K (5K in parallell with 2K5) and the resulting input at the VCS with the switch closed would be 2.7V. Consequently the VCS would never see a logic low input and you would have no game control, that is why a buffer chip is needed to isolate the output from the power pad switch (and therefore its resistance) from the VCS input/pull up resistance.

 

Unforturnately if memory serves the logic levels for the standard TTL logic that was probably used in the VCS are 0 - 0.4V for a logic low and 1.7V (may have been 2.7V) and above for a logic high.

Consequently as previously stated the only easy ways I can see of you getting the input voltage lower than 2.5V are...

1) Connect the power pad as per the joystick in the top circuit (VCS + Joystick) and change the 5K internal pull up resistors to 27K or

2) Connect the power pad as per the lower circuit (VCS + Power Pad) but place a 27K resistor between the +5V and the power pad and don't forget the buffer chip (preferable one with a schmitt input).

PowerPad.bmp