bizarre paddle behavior

[+rbairos]

I acquired two sets of paddles recently.

Neither worked, but after a thorough disassembly and cleaning of the pots the first set works and the other set partially works,

*except*

One button causes the atari to short out.
Not sure how else to describe it. The screen blacks out with with a thick band of white noise in the middle while that button is pressed.
On releasing, the game is in some bad state.

I did a full tracing of the wires to the connectors, and things get very confusing for me:

Working paddles:

switch 1 black: pin 8 (button gnd)
switch 1 green: pin 3 (button 1)

pot 1 yellow: pin 7 (pot vcc)
pot 1 purple: pin 9 (pot1 out)

switch 2 black: pin 8 (button gnd)
switch 2 brown: pin 4 (button 2)

pot 2 yellow: pin 7 (pot vcc)
pot 2 red: pin 5 (pot 2 output)


Note both yellows are connected, both blacks connected, everything makes sense.


set with the button-shorting behavior:

switch 1 black: pin 8 + pin 3 (button gnd + button1)
switch 1 green: (no connection)

 

pot 1 yellow: pin (no connection)
pot 1 purple: pin (no connection)

switch 2 black: pin 8 + pin 3 (button gnd + button1)
switch 2 brown: pin 4 + pin 7 (button2 + pot vcc)

pot 2 yellow: pin 7 + pin 4 (pot vcc + button2)
pot 2 red: pin 5 (pot 2)

Im trying to wrap my head around this being some short, but it just seems horribly mis-wired from the start ?

I can't see how this may have changed after leaving the factory ??
I could imagine pin 3 breaking internally and shorting to its neigbor below pin 8,
but pin 4 and 7 aren't even neighboring.
5 4 3 2 1
9 8 7 6

Any tips on how to repair it?
I feel I need to cut off the connector head, and just make a new one, but want to understand how it got in this state.

Thanks,
Rob.

Edited February 17, 2018 by rbairos

[BLASTER]

Damaged/crushed harness inside where you can't see it?

 

Disconnect the harness from the pots,and ohm test the individual wires.

[BigO]

There's no reason the +5v would connect to a button.

 

I would guess that the problem set was rewired to work with some other device.

 

I would completely disconnect the wires from the pots and buttons and test the wiring as suggested, then wire it back correctly.

[+rbairos]

I'm a bit pessimistic. If it was a different system one switch lead and entire pot don't seem to make it to the pins. I don't see any visible damage either. Its strange in that in uses the same color scheme as the working set.

There's no way to open the head end is there? I'm assuming it's a solid encased hunk of rubber?

[BLASTER]

You cannot open the connector without destroying it.If you do not have continuity between the pots and the connector pins,your harness is damaged internally.

Internal damage to the harness can be invisible from the outside.Could have been crushed/pinched at one time and ruined it.

[BigO]

I'm a bit pessimistic. If it was a different system one switch lead and entire pot don't seem to make it to the pins. I don't see any visible damage either. Its strange in that in uses the same color scheme as the working set.

There's no way to open the head end is there? I'm assuming it's a solid encased hunk of rubber?

It could be that it was rewired to be used with some device that only supported one analog controller input on a port and the perpetrator disconnected the other wires to prevent interference with the function of the other pins. It might not have even been a game console. It's not much of a stretch for me to imagine doing something like that myself.

 

By "uses the same color scheme", do you mean that you verified the mapping of each color of wire by checking its continuity to the pin(socket) on the connector, and that the color mapping matched the working set exactly? There's no guarantee of consistency from one controller to the next.

 

Having said that, I didn't spot anything obvious in your description of the wiring that would cause a direct short. Though, I don't know what would happen if the software configured a pin to output a 0 and then 5v was applied to that pin.

 

I'm assuming that if you have an ohmmeter that you used to trace out the wire color mapping. If so, you will probably also have already verified that there are no shorts between any two pins of the connector. If all that is true then it seems like the next obvious step would be to rewire it to match the working set.

[+rbairos]

To be clear, there is no continuity between one pot and any pins in the header, nor is there any continuity between one switch lead and any pins on the header.
On the header, pins 4 and 7 are shorted. So doesn't seem to be any rewiring I can do, unless I'm missing something.

By 'same color scheme' I mean every color on the working set is physically connected to the same pot/switch destination on the problem set.
(ie: one unit has black+green on the first switch, yellow+purple on the first pot, then black+brown on the second switch, yellow+red on the second pot).

So I have to think either the pins were rearranged at the rubber header end (seems impossible, unless it was some weird factory mistake at the pin side), or the harness is internally damaged at some point.

I might desolder the harness just to be sure.

[+rbairos]

Should also add the buttons are not shorted, nor are the pots . they measure correctly on their side.
Assuming the harness is damaged internally I have to cut the cord at some point, which makes me sad.

[BLASTER]

Just spend a few bucks and buy another set of paddles.Maybe you can get a refund from whoever sold you the defective set.

[+rbairos]

Yah, its not the money. Just wary of destroying something rare in my attempt to preserve it.

[+rbairos]

So the fault ended up being right in the header/connector somewhere.
I had to clip off both cords just below it.

The units seem otherwise fine, so I'm just going to resolder the cords to a new DB9 connector.
May not look factory original, but should be reliable enough while preserving most of the original.

[zylon]

So the fault ended up being right in the header/connector somewhere.

I had to clip off both cords just below it.

 

The units seem otherwise fine, so I'm just going to resolder the cords to a new DB9 connector.

May not look factory original, but should be reliable enough while preserving most of the original.

 

Sometimes, the little pin sockets in the plug end of cable get crushed and pushed back just enough to cause headaches like that.

[+rbairos]

Ah, that makes sense. Was really puzzled.
I just ordered a sega extension cord from aliexpress to use the connector.
30 to 60 day delivery time, but only $2.46 !

[+rbairos]

Just a followup.
I received the 9 pin cable over a year ago and finally decided to splice it in tonight.
Was going well, until another break in the original atari cable.

Then another.
It was looking pretty monstrous so going to rethink this cable replacement fix.
 

[+rbairos]

Alright, ended up using a length of spare svideo cable (had 4 conductors), and a DB9 connector (with internal terminal strip).
Not factory identical, but tidy enough and works and saved the 40 year old bases from landfill.