smbaker controller board and joystick positioning issue

[barrelsAndRivets]

I assembled one of Dr. Baker's Atari 5200 controllers as documented here:
http://www.smbaker.com/homemade-atari-5200-analog-controller

 

Overall I'm confident in the build. However, even before trying to use it, I thought to run it through Pete's Diagnostic Cart to make sure everything was mapped correctly, etc.

 

The main thing I noticed involves the positioning of the analog control. When centered, my stock (rebuilt) CX52 registers at position 100,100 and the blue dot appears almost exactly centered in the blue box. I've previously adjusted the pokey so I'm confident that the system itself is in good shape in this sense.

 

However, with the smbaker build, the joystick registers at position 50,50 as pictured here:

 

It appears to the top-left of the blue box as pictured here:

 

This goes well beyond my understanding of the schematic and the digital -> analog potentiometer solution and I'm wondering what could be off and/or what I can adjust. Again, I have not tested any games yet but can only guess this is problematic (?)

 

Thanks in advance!

Edited October 29, 2020 by barrelsAndRivets

[+-^CrossBow^-]

And to make it more confusing, the actual adjustment per the service manual is actually around 112+/- on the HOR reading for a properly calibrated console with the loopback board installed. Actually it comes out to about 112 HOR 116 VRT on most consoles I've calibrated.

 

 

 

[Danjovic]

your digital pots may be 50k instead of 100k or your caps in pot x poty lines may have a lower value.

Which are the limits you can get at full up-right and full bottom-left?

Edited October 29, 2020 by Danjovic

[Swami]

1 hour ago, barrelsAndRivets said:

I assembled one of Dr. Baker's Atari 5200 controllers as documented here:
http://www.smbaker.com/homemade-atari-5200-analog-controller

 

Overall I'm confident in the build. However, even before trying to use it, I thought to run it through Pete's Diagnostic Cart to make sure everything was mapped correctly, etc.

 

The main thing I noticed involves the positioning of the analog control. When centered, my stock (rebuilt) CX52 registers at position 100,100 and the blue dot appears almost exactly centered in the blue box. I've previously adjusted the pokey so I'm confident that the system itself is in good shape in this sense.

 

However, with the smbaker build, the joystick registers at position 50,50 as pictured here:

 

It appears to the top-left of the blue box as pictured here:

 

This goes well beyond my understanding of the schematic and the digital -> analog potentiometer solution and I'm wondering what could be off and/or what I can adjust. Again, I have not tested any games yet but can only guess this is problematic (?)

 

Thanks in advance!

Yeah, see if you can reach 150 by pushing the stick all the way horiz and vert. 

[barrelsAndRivets]

I just took another look at this and I may have an explanation. Here's the board:

 

The yellow caps are 399-4264-ND‎, which are 0.1UF. The blue caps are 490-7533-1-ND‎, which are 0.22UF and are meant for the potentiometer step-up.

 

The "blues" were supposed to go in C1 and C2, and the "yellows" in C3 and C4, but those aren't labeled on the board. I'm going to take a wild guess that I inadvertently swapped the relative positions (!) especially when comparing to pics of others' builds.

 

I'll plug it back in later this evening and see what my minimum/maximum values are as it sits, but could this be the culprit?
 

[+DrVenkman]

The wrong value of caps will definitely give incorrect readings on the potentiometer. The other (far too common) problem I have run into building these is a digi-pot chip out of tolerance on one or both axes. Usually it's just one, and usually, for some reason, it's the X-axis. I have probably 15 of them around here that read 10 - 20 points low on one or both axes when measured with Pete's. That usually manifests itself as a game that does not work properly in maze games but it depends on the game. The best test of a functional controller is Gyruss, interestingly enough. If the controller isn't working right, you typically can rotate left and up to the 12 o'clock position, but not back down and right the rest of the way clockwise, or some variation of that. Basically, if you can play Gyruss properly with smooth controlled rotation around the tube in both directions, the controller and POKEY resistor combo is working fine.

[barrelsAndRivets]

4 hours ago, DrVenkman said:

The wrong value of caps will definitely give incorrect readings on the potentiometer

Can you confirm that where I have the yellow caps in on my board, right above the joystick, that the 0.22UF caps are actually supposed to go there (C1/C2)? Thanks!

[+DrVenkman]

7 hours ago, barrelsAndRivets said:

Can you confirm that where I have the yellow caps in on my board, right above the joystick, that the 0.22UF caps are actually supposed to go there (C1/C2)? Thanks!

C1 and C2 are supposed to be 0.22 uF caps. Color may vary depending on the manufacturer of the capacitor of course. 

[barrelsAndRivets]

Thanks to all that replied -- parts arrived, and after swapping the capacitor locations I'm in business