Page 8 of 10

Re: Help fix a NES Clone (Phantom System)

Posted: Thu Oct 19, 2017 5:15 pm
by Fisher
Finally could flash the rom and test.
And the results: No joystick connected!!
Damm!! I just don't know what can be causing this bug and I'm out of ideas again!!
Any suggestion are welcome!!
l_oliveira wrote: opaque 3M tape
I saw some of these on the component side, but what I'm referring is a thing that seems like dust, is kind of white and opaque and appears on both sides.
The picture just don't do justice showing this thing, I'll try to take a better one tomorrow.

Re: Help fix a NES Clone (Phantom System)

Posted: Thu Oct 19, 2017 6:16 pm
by l_oliveira
Phantom System has some silly crap on it's design like some control lines of the CPU connected to the cart slot which aren't supposed to...

Audio and Video on the cart slot? Only to make the signals gather interference from the bus signals, right?

Why a cartridge would need/want INP0/INP1 pins or OUT0? (OUT0 may cause interference on the controller ports).

I suggest you cut these traces so they no longer go through the cart slot.

Re: Help fix a NES Clone (Phantom System)

Posted: Thu Oct 19, 2017 6:39 pm
by lidnariq
Had OUT0..OUT2 been connected to the cart slot, it could have made small discrete logic mappers cheaper (obviating 3/4 of a 74'161)

Of course, that supposes the existence of developers that can write or port software to use that interface instead.

Re: Help fix a NES Clone (Phantom System)

Posted: Fri Oct 20, 2017 1:45 pm
by Fisher
I cutted /INP0, /INP1 OUT0 and Video out from the cart connector.
I didn't cutted audio because maybe I can do an adapter to FC carts + audio.
Got the same results!! :evil:

Fortunatelly, after puting sockets on the 74'139 and in the buffers and put them back in, I ran Tepple's program and got this:
20171020_183115.jpg
This should be normal, but only the 1st controller was connected.
The controller is like this one:
Image
Internally it's only a epoxy blob.
I also got the Low-level probing results:
Low-level probing results
Low-level probing results
Hope this helps to find some clues on how this can possibly be fixed.
Again, thanks in advance.

Re: Help fix a NES Clone (Phantom System)

Posted: Fri Oct 20, 2017 6:25 pm
by tepples
...00.SS means bits 4 and 3 are driven to zero, bit 2 is not connected, and bits 1 and 0 are both serial. Based on which bits are driven to zero vs. not connected, it thinks you have a top-loading NES. (It isn't built to recognize famiclones.) And if you have a top-loading NES, it assumes the controllers in ports 1 and 2 are dogbone controllers.

"1P D1" refers to player 3's controller in the Famicom DA15 expansion port, which normally doesn't exist on a top-loading NES unless it's hacked in.

Re: Help fix a NES Clone (Phantom System)

Posted: Sat Oct 21, 2017 4:41 am
by Fisher
I unplugged Out1 and Out2.
Tested with Gradius and got the same results.
Could this be some problem with the CPU's RAM or the clock generator?
That and the PPU (PPU + latch + RAM) are the only parts I haven't messed around yet.

Re: Help fix a NES Clone (Phantom System)

Posted: Sat Oct 21, 2017 6:38 am
by krzysiobal
Why not connecting logic analyzer to CLK/STRB/DATA to see what happens?
I meet tons of broken 74368 but never working CPU with STROBE failure.

Re: Help fix a NES Clone (Phantom System)

Posted: Sat Oct 21, 2017 10:55 am
by Fisher
Unfortunatelly I don't have one... :cry: Maybe I can build it??
I tried to substitute it's CPU with another from a working clone and got the same results.
That's really bizarre!! :shock:
Maybe it's some resistor or capacitor going bad?
Again, I have no idea of what to try next.

Interestingly, I played Rockman 3 on this clone for hours, and some other games like Super C, SMB3 and SCAT too.
Maybe the clone just doesn't "like" Rockman 2 and Gradius.
And I tought machines had no feelings... :lol:

Re: Help fix a NES Clone (Phantom System)

Posted: Sat Oct 21, 2017 12:48 pm
by lidnariq
OUT1 and OUT2 shouldn't be relevant.

You can get a cheap logic analyzer by getting one of these and some test clips.

Re: Help fix a NES Clone (Phantom System)

Posted: Sat Oct 21, 2017 1:57 pm
by Fisher
That seems very nice!! I'm putting in my wanted list!!
After getting a logic analyser I should search for what?
I already did tests without both the 74'368 and the resistors and got the same results.
Again, this is bizarre!!

Re: Help fix a NES Clone (Phantom System)

Posted: Sat Oct 21, 2017 3:54 pm
by l_oliveira
You should be testing for broken traces maybe.

Also have you tested a different CPU IC on that system?

Re: Help fix a NES Clone (Phantom System)

Posted: Sat Oct 21, 2017 5:40 pm
by Fisher
Yes. I tried other CPU and got the same results.
I tested all traces before putting back the demux and the buffers.
Now I'm using sockets on them.

I also checked the TST pin (on the schematic it's marked as /RDY) and it's OK.
What else should I look for?
Maybe an exorcist? lol!

Re: Help fix a NES Clone (Phantom System)

Posted: Mon Oct 23, 2017 5:49 am
by Fisher
I did a memory test with NES test rom (or whatever name it is).
Altough I didn't see properly the graphics (I only have an UNROM testboard) the sound it made is quite similar to what I hear on an emulator.
With this in mind I'm assuming this test went fine.

Another interesting thing I noted is that it becomes worse if I try using the GameGenie with it.
Rockman 2 reboots as soon as the screen fades after the helmet appears on Rock's head.
With Gradius I got the same results and without the Genie: I just can't start the game at all.

This is what the Rockman's repro looks like on the component side and the solder side.
Just in case someome wants to take a look.
The only thing I removed was the resistor, since it seemed to be redundant.

Could a small alteration on the clock generator result on this?

I've tested with a few more games: Battletoads, Q*Bert, Gyruss, Kings of the Beach, Fester's Quest Journey to Silius, Metroid, Bionic Commando, Life Force and Gradius 2. All played just fine. If someone suggests a specific game to test, I'll try to get the game make it.
It would be nice if I could get anoher Gradius and Rockman2 to test, but unfortunatelly I couldn't find anyone around who had it.

I'll post again if I have any other news.
I think I'm probably missing some detail on this weird defect...
For now I'm out of ideas again. Any suggestions are welcome!

Re: Help fix a NES Clone (Phantom System)

Posted: Mon Oct 23, 2017 7:15 am
by krzysiobal
Can you make photos of entire PCB of this console from both sides?
Are the resistors pulling up CK/D0/NMI/IRQ etc discrete ones or common resistor ladder?
I had one time funny problem with internal short-circuit in such ladder.

Re: Help fix a NES Clone (Phantom System)

Posted: Tue Oct 24, 2017 2:02 pm
by Fisher
The resistors are all common ones.
Here are some photos of the mainboard.
I didn't put the 2nd player's buffer back yet.
Hope this helps.
Component side
Component side
Solder side
Solder side