Bad control port ? on a 250425 C64

[Vipersan]

Most of the C64s I've worked on are 250407 boards..
I aquired recently a 250425 board so hooked up my test harness and diag cart ..
It passes everything ok except a BAD Control Port is highlighted.
Going off previous C64 encounters this kind of error can usually be traced to U1 6526 keyboard CIA .... U28 4066 analogue switch or even the SID (paddles) ?
So I pulled an older 250407 known to be good and working..
Tested the harness and diag cart on this board.
Just to be sure there wasn't a fault with the test procedure/harness.
It passed 100%
This board is fully socketed which makes for easy chip swapping.
I then pulled U1 U28 and U18 (SID)..
Socketed U28 on the suspect 250425 ...the keyboard CIA and SID were already socketed.
I was now able to swap out the 3 chips one at a time...
All three now swapped but the Bad controller error persists ?
Now I'm confused.
I guess it could be a dry joint or bad trace...so I guess I will spend a few hours checking the traces for continuity....but before I do this I thought I'd ask if there are known compatibility issues using a test harness on this 1984 pcb...
..or if there are any other chips that could cause this control port error ?
I am assuming the error is real and I'm not chasing a phantom.
suggestions welcome.

 

[Vipersan]

So ...there may well be an issue which is paddle related.
Using GCT.prg on my kung fu flash I first connected an atari style joystick on game port 1 on the fully working 250425 board.
It tested fine
Then on to Game port 2 with the same joystick.
It too worked fine.
I repeated this whole process on the suspect 250407 board ..
again both ports responded just fine to the joystick in ports 1 and 2

Ok paddle time.
First the 250425 with my atari paddles (the one with a tennis racket logo) using game port 1
The pots were a little noisy but responded well in both X and Y axis...linear transition from 0 to 256.
On to game port 2.
The Y axis fairly responsive with linear transition from 0 to 256.
but the X axis ..instant transition from 0 to 255 at the slightest rotation of the pot.
Acting more like a microswitch on a joystick.
Conclusion..
could be an issue with the SID which works perfectly fine for audio...but is resonsible for paddle movement interpretation...
I'm hoping it isn't SID....and will swap once again the SIDS between 250407 and 250425 boards later.
All tests will then be repeated.
It's not like this is a deal breaker but annoying just the same.

 

[Vipersan]

So ..it looks like SID is fine after all..
Borrowed the one from the working 250407 board.
popped it into the 250425 board..ran the paddle test again..
Exactly the same problem ???
I guess I'll have to search elsewhere on the board for a solution.
As I said at the start ..I'm not familiar with 250425 board..
Is there a revision change that changes the paddle action on this board and thus prompts a BAD control port from diag cart ...that really doesn't exist ?
..or is there a genuine fault ?
At this point I really dont know.

 

[Vipersan]

Checked some voltages on U28 with the chip removed...
Pin 1 of the 4066 was highly suspicious showing around 3.2v constantly.
closer examination under the microscope showd a possible trace from pin 1 too close to another trace running parallel to it.
ie it had moved.
out came the socket I had fitted...using low melt as the board had already suffered enough I figured.
sadly the suspect trace to pin one came with it...so I'll bodge this tace before fitting a new socket...and stick the fine wire down with green nail varnish first.
will it fix the issue..?
who knows ...but It has to be done.

 

[Vipersan]

Looks like it paid off..
..a clean sweep with the diag cart now.

 

[VintageVic]

Congrats, another commodore healed!

 

[cinnober]

Hmmm the 586220 diagnostic manual says this..
I have a 250425 myself with EXACT same error using harness/diagnostic cartridge...but I dont have the 5-PIN video out..but have the mentioned capacitors .