Repair Atari 2600 that did not work with paddles – Problems and Solutions

The paddles do Portal 2600 are game controllers that have an analog and a digital wheel action button. They exist in pairs, where the wires come together connect in one of the Atari controller port 2600, so that it is possible to connect up to four paddles in an Atari 2600, two on each door.

Paddles Atari VCS / 2600

Paddles Atari VCS / 2600

A little bit of my story with paddles

In the USA, the first Atari 2600, which at the time still called only Atari Video Computer System, They came with a pair of joysticks and a pair of paddles. later the paddles They were not included, probably to cut costs and also because there were many more games for joysticks than to paddles.

No Brasil, a lot of people probably do not know that this kind of control existed for Atari 2600, because I believe that they have never been officially sold here. I even spent my entire childhood and adolescence without knowing they existed.

Games support paddles it was not common here, probably for the same reason. Just remember to have had contact with one of them at that time. In a multigame cartridge that a cousin lent someone a few times there was Circus Atari. Not knowing that the game used the paddles, We could not play. We tried everything and concluded that the game was faulty, because the platform does not leave the place. We could throw the acrobat because the buttons on each paddle actually correspond to the same signal push the joystick left or right, respectively. But we could not move the platform, because the wheel signal paddle has no counterpart in joystick.

I only have my first pair of paddles when acquiring the Atari 2600 US VCS with S-Video, Composite Video, Stereo Audio and Pause, in 2014, as I have shown here in Skooter Blog. They came in the package. work, but they are somewhat inaccurate, need of maintenance. The paddles They use a potentiometer to control the wheel, and with time the pots are becoming bad and need to be replaced.

I just acquiring another pair of paddles of Best Electronics, as well as already described here in Skooter Blog. This has come with pots that last 3 times more than the original and which are a result of search 6 a 7 Best made for years Electronics. These rather work perfectly and they could finally play Atari games 2600 that USAM paddles, I had never had the opportunity to play before (except through emulators).

The Atari 2600 that did not work with paddles

In 2017 acquired another Atari 2600, the Atari 2600 the Polivoks with American Board Rev. 17, who has also shown here in Skooter Blog. I bought it with the intention of de-transcode-lo (PAL-NTSC M) e install it one 2600RGB kit. Both procedures were successful and described herein in Skooter also Blog.

But the truth is that I had not gotten to test this Atari 2600 with paddles. Not before installing against, or after. So I do not know to say if it has come to this defect or if it appeared later, perhaps during the installation of against. I never know…

The fact is that recently I downloaded pack Atari 2600 from the SmokeMonster, where ROMs are separated in different ways. So we have a directory with all games paddle separated, which helps a lot, because before I had to stay looking listinhas to know what are the games that use the paddles.

Only this June, specifically on 03, It is that I decided to finally put paddles in this Atari 2600 the Polivoks with American Board Rev. 17 and, to my disappointment, they did not work.

connecting paddles on the left controller port, only the buttons worked, the wheel does not make any effect. On the right controller port they functioned normally. I tested with various ROMs and no work, I tested another Atari and worked. The problem was actually in the console.

Investigating the problem

To investigate the problem, first sought the scheme pins door controllers Atari 2600. I found it easily in Atari Age forums:

Scheme pins Atari controllers 2600.

Scheme pins Atari controllers 2600.

With the meter tested pin 7 and I found that the + 5V were present. This voltage is used to feed the paddles, through their pots which are attenuated in accordance with the wheel position, and returning the pins 5 and 9, respectively. So the problem in Atari 2600 I was not in the circuit before paddle, but the circuit after paddle.

The next step was to take the Atari scheme 2600 to see where to spend and where are the signs of the paddles. I also found the scheme in Atari Age files. Check out:

Scheme Atari 2600.

Scheme Atari 2600.

Watching the scheme, I realized that pin signals 5 and 9 each door pass by only four capacitors (a capacitor for each pin) and then go directly to the pins 37 a 40 no aunt, again a TIA pin for each paddle. Soon the problem should be in the capacitors, no aunt, or on the tracks between them.

I tested the continuities and everything was right, each pin of each door opened controllers continue until the respective capacitor terminal and to the respective pin TIA. Even redid some suspicious welds, but nothing has changed.

I also tried to get the 2600RGB to verify that the against It was not affecting anything, but there has been no change.

I then decided to identify and test the capacitors. On board they are identified by C218, C219, C220, C221, and according to the service manual Atari 2600, are capacitors 68 nF e 100V (Cap. Epoxy Dipped .068 uF (100In)). In the body of the capacitors registration is .068K 100V.

polyester capacitors give hardly fault. The problem usually electrolytic, but anyway I thought I'd test them. In the photos below they are the verdinhos capacitors:

Here came my first faltering. I decided to test the capacitors on the card itself with the BSIDE ESR02 Pro, even though the test card is not reliable as there is influence of other components. The BSIDE ESR02 Pro He did not identify the capacitors as capacitors, but presented capacitance measurements. These measures have been:

C218 – He did not show capacitance measurement
C219 – And = 9.5 uA, Uf = 2.37V, C = 66.7n
C220 – And = 9.5 uA, Uf = 2.39V, C = 66.1n
C221 – And = 9.4 uA, Uf = 2.38V, C = 62.1n

I also capacitance tests with ET-997 multimeter MINIPA and he also introduced similar measures:

C218 – 00.0 nF
C219 – 68.4 nF
C220 – 67.7 nF
C221 – 63.7 nF

Remove capacitors to test is the story: It takes more work to get a component of that place, then we are taking to test, even if the capacitance is ok, it is better to have put a new logo. Because I I bought a package Capacitors Polyester metallic 68nF 100V, as I have shown here in Skooter Blog.

My hesitation was great not having removed the TIA and remade the tests on board before taking the capacitors. As the TIA is a socket, It is very easy to remove. If he had done that would have drawn the capacitors were perfect. But OK, living and learning…

I changed the two left controller port of capacitors. And testing the capacitors removed saw that they were perfect, indicating the correct capacitance. With the two new capacitors installed Atari 2600 continued not working with paddles.

So all my suspicion was to TIA, it is that it should be short and not letting the C218 capacitor carry and indicate its capacitance.

I did a series of tests and eventually the paddles even they worked on both doors, but strangely the image vanished when one of paddle at the door 1 He was taken to the right limit. after the paddles They stopped working on both doors. Certainly there was a gremlin on my system.

I could even get a TIA another Atari 2600 to test, but not Atari 2600 US VCS with S-Video, Composite Video, Stereo Audio and Pause TIA has soldered wires directly on the pins because of against. On other Atari 2600 with Rev. 17 TIA is soldered directly on the board. How is working is better not mess. Por fim, the my Atari 2600 the Polivoks plate It is also working and still has the seal of Polyvox, I do not want to mess with it.

Before leaving in search of a new TIA, I decided to do more tests to make sure the problem, which is what I should have done from the beginning.

Removing the TIA, all the 4 capacitors are identified as capacitors in BSIDE ESR02 Pro and show perfect, We tested directly on the board. And here we have another clue that the problem is in TIA. I never take pictures of this procedure, but the indication on screen BSIDE ESR02 Pro It is equal to this, with the capacitor disconnected:

Capacitor test Metallized Polyester of 68nF and 100V.

Capacitor test Metallized Polyester of 68nF and 100V.

testing other Atari 2600 with Rev. 17, with TIA soldier, I noticed that 3 capacitors show the correct capacitance, even with the connected TIA, but this time showing the wrong capacitance is a paddle the right door controllers. Testing the operation console noticed that paddles typically operate at the door 1, but at the door 2 one of them is with a very small strip of running the entire length of pot, with the object giving jumps on the screen. This probably corresponds to the capacitor problems. In this case it may be the same capacitor. But I'll leave this problem to solve this other Atari later. We will deal with an Atari at a time.

I then turned Pro Atari 2600 the Polivoks with American Board Rev. 17, and tested in the capacitor voltages entrance and the TIA with and without pins paddle connected, with the TIA removed and bound console.

As paddle connected have very small values, few mV. With paddles connected have values ​​ranging from 4,3V ~ 5V depending on the position of potentiometer paddle. The values ​​are the same in each terminal capacitor or the corresponding socket terminals TIA.

then I tried to test each paddle on every door and pick up the exact values:

  • Paddle on the far left: 4,4~4,5V
  • Paddle on the far right: ~5V
  • paddle offline: few mV

then again I connected the TIA, and with the same Atari 2600 Tests on the remade. This time got:

  • Paddle on the far left: 1º Capacitor: ~300mV, 24 th to capacitors: ~ 4,4V
  • Paddle on the far right: ~5V (all)
  • paddle offline: 20~30 mV

Here we see another strong indication of problems with the TIA, perhaps a short circuit on the pin that corresponds to 1 paddle.

But I did not know what are the values ​​that should be presented, so I needed to test the other Atari 2600 with Rev. 17 to see if the values ​​hit.

then tested the other Atari 2600 with Rev. 17, with TIA soldier, and the values ​​I got were these:

  • Paddle on the far left: ~20mV (all)
  • Paddle on the far right: ~5V (all)
  • paddle offline: 20 mV

These values ​​are very different from Atari 2600 problematic, then definitely there is a gremlin by.

I decided to, do resistance tests between ground (point of contact shield) and capacitors or TIA own or TIA connection. All three give almost the same value.

With the Atari 2600 the Polivoks with American Board Rev. 17 problematic c / o connected TIA, getting the following values:

  • Paddle on the far left: Paddle 1: 60k ohms; Paddles 2 a 4: 1M ohms
  • Paddle on the far right: 430 ohms (all)
  • Paddle offline: 0 (all)

Repeat the test with the TIA and got disconnected:

  • Paddle on the far left: 1M ohms (all)
  • Paddle on the far right: 430 ohms (all)

And here we have another strong indication of short between the pin TIA circuit that matches paddle 1 and ground.

The solution

With all my chips wagered in TIA, I tried to purchase a new TIA, as well as already described here in Skooter Blog.

Chip TIA do Atari 2600.

Chip TIA do Atari 2600.

With the new TIA at hand, I did the tests in Atari 2600 the Polivoks with American Board Rev. 17 problematic connected with it, and finally I got the correct values ​​for all paddles:

  • Paddle on the far left: 1M ohms (all)
  • Paddle on the far right: 442~443 ohms

And then came the moment of truth, connect the Atari 2600 the Polivoks with American Board Rev. 17 and see if paddles finally they would work properly. So come on: som drums, please… a little more… a little more… worked!!! \0/

Check out the video of the Atari 2600 the Polivoks with American Board Rev. 17 with the new TIA and the paddles working perfectly:

Before closing the pet spent shifted varnish welds where, I sanded a little rust metal shield with micro-grinding and spent varnish it also protects it better. I fixed it still one of the board support that was broken, using Araldite and a capful of little pot of Tic Tac. It would be better some black plastic, but found none available. Then I took the opportunity to make a good cleaning on the island.

Check out the photos:

Now it's time to finally have fun with the Atari games 2600 using the paddles. See the live broadcast I did in Skooter Blog channel on Youtube testing virtually all games that use paddle in Atari 2600:

Unfortunately Youtube cut much of the transmission in time to file, and this has left many games out. Other games were left out to have screen refresh rate problems (bad scheduled) and do not work well with Framemeister and OSSC. TV CRTs are more tolerant of this strange refresh rates, frequencies with non-standard or variants, and so many games have been released so.

Final Considerations

And so I fixed my Atari 2600 the Polivoks with American Board Rev. 17. TIA found that not only handles video signals, but also directly with the entry of the wheels of paddles, and the buttons joysticks. On the other hand, directional of joysticks (whose signals also act as buttons paddles) They are treated at Riot. What strange architecture of the Atari 2600…

I learned that the order of my tests was not the most correct. I should have started with the stress tests with and without the TIA, and the capacitance of the capacitor without the TIA. With this I would not have moved in the capacitors and direct leaving for the exchange of TIA.

But at least I still enjoy the capacitors acquired no outro Atari 2600 that is the problem in one of the paddles. In case it is best to start by changing the capacitor only has two terminals for the TIA is soldered directly on the board, and will give a reasonable handiwork remove it with all your 40 pins. Of course, if I have to do this I will enjoy for a socket and facilitate the next maintenance.

Also of course I could have left this issue for professionals resolve, but then I would not have learned new things. It's cool to know better this console that gives me joyful moments since 1987.

Financially also compensated, after all I spent about R $ 90 with the new TIA and suprimento vitalício de capacitores. Um profissional certamente cobraria mais que o dobro disso entre mão de obra e o custo do TIA, que teria que ser trocado de qualquer forma. E eu ainda gastaria com o Correios, pois não conheço ninguém especializado em consoles antigos na minha cidade. E correria o risco do Correios perder meu pacote.

É claro que eu excluí as minhas horas de trabalho dessa conta, senão ela não fecharia, mas como eu disse: vale pelo aprendizado. É frustrante quando não consigo resolver um problema. Mas quando consigo, todo o trabalho é compensado. E os testes do console após o conserto definitivamente podem ser categorizados como lazer. 🙂

5 1 vote
Article Rating

Permanent link to this article:

Sign up
Notify about
more new top rated
Inline Feedbacks
See all reviews

Tenho uma inveja do bem por Vossa Mercê.


It is very cool to read and see renovations and repairs of old consoles. I always try to solve the problems of my consoles, just because I like the process of learning (and save). Thanks by the teachings!

Iuri Fiedoruk

Atari with mod rgb is pure ostantação, heheh. Very beautiful image, cool the description of the whole problem and solution.

Rodrigo Saraiva

Parabens pelo compartilhamento…. Tbm sou fã do atari 2600 – meu segundo console (antes dele tive alguns telejogo)…
Coincidencia ontem estava restaurando um atari que comprei por R$ 50,00 – o coitado era pura sujeira mas a placa ate que estava bemdepois de trocar o regulador o bichinho funcionou r4edondinhonas pesquisas pela net em busca do esquema eletrico esbarrei na pinagem do DB9 e pensei em usar algum joysticks sucata que tenho pra fazer um paddle caseiro…. 🙂 mais que economianao…. nao sou rico 🙁o que me atrai é fazer meus dispositivos, aprender e vera coisanascerComecei esse processo de criar coisas fazendo mods em carrinhos de CR com meu filho.
Poderia me dizer o valor dos potenciometros? qual o esquema de ligacao neles?

TBM have a project to remake the Xevious in Batari with two shots fired by the player (of course we have to create a joystick with 2 buttons..

Hugs and congratulations again for sharing – and the report well written


Rodrigo Saraiva

Grateful for the kindness my dear

How would Dr. Spock: “live long and prosper”

We would like to know what you think, Leave your commentx
Enable Notifications    OK No thanks