Jailbars on a front loader - help required

[lidnariq]

I'd try something like "5V pin on PPU -- 100µF -- composite center"

[tepples]

I think lidnariq is talking about a "bypass capacitor" or "decoupling capacitor". Those are usually 100 nF ceramic and placed as close as possible to the +5V and GND pins of the PPU.

[lidnariq]

Not in this case. Here I'm actually talking about a DC blocker rather than a bypass cap.

Oh, hey, actually, romibraman could use an AM radio to check for the signal strength of the signal coupling in. In a PAL NES , PPU A13 will oscillate at 655-665kHz (661-671kHz NTSC), so one should be able to get a measurement of signal strength by passing a loop of wire near an AM radio antenna set to 657 or 666kHz. Don't directly connect things, just rely on capacitive coupling.

[romibraman]

Ok so replacing the 74HCU04 had no effect...

And I don’t have an AM radio

I guess at this point I should try to find an oscilloscope.

By the way if I were to find one and I could confirm there’s indeed a signal from PPU a13 coupling in (and im’pretty Sure you’re right because the only time I was able to remove the bars was after desoldering ppu a13). What would be the fix ? Would there even be one ?

[lidnariq]

As I said, there's basically two paths and one solution for each.

If it is the 5V supply being insufficiently regulated, you "firm up" the regulation by adding more capacitance, and/or decreasing the ESR of the capacitance that is there. (This is why Fisher's pictures show the extra wires soldered in: to decrease the ESR of the traces on the board). In real extremes, you can add multiple 5V regulators for the different parts around the analog path to be able to reduce the amount of variation on load on each regulator. But your board is an OEM one, so it "shouldn't" need major rework, instead parts "should" have aged out.

If it isn't the 5V supply, then the only other real option is capacitive coupling; this is fixed by rebuilding the video pathway and isolating it from all the other parts.

[Memblers]

I have heard of a "wavy lines" problem being fixed by adding a 0.1uF (100nF) ceramic capacitor in parallel with C15 (near that blue connector). Not the same symptom, but it wouldn't hurt to try.

[romibraman]

As I said, there's basically two paths and one solution for each.

If it is the 5V supply being insufficiently regulated, you "firm up" the regulation by adding more capacitance, and/or decreasing the ESR of the capacitance that is there. (This is why Fisher's pictures show the extra wires soldered in: to decrease the ESR of the traces on the board). In real extremes, you can add multiple 5V regulators for the different parts around the analog path to be able to reduce the amount of variation on load on each regulator. But your board is an OEM one, so it "shouldn't" need major rework, instead parts "should" have aged out.

If it isn't the 5V supply, then the only other real option is capacitive coupling; this is fixed by rebuilding the video pathway and isolating it from all the other parts.

Ok here’s what I’m going to try next, tell me if it makes sense.

Pull pin 21
Run a wire to a bread board
Recreate the composite signal there
Test it on a tv.

This way I should no for sure if the signal is corrupted after the PPU right ?

[lidnariq]

Yeah.

You can use a super simple amplifier: https://forums.nesdev.com/viewtopic.php?t=13902 , or a slightly more complicated one: https://forums.nesdev.com/viewtopic.php?t=10554

Coax or twisted-pair might be clever at first, because your wires to your breadboard are probably not going to be particularly short and they'll pick up noise.

[romibraman]

Yeah.

You can use a super simple amplifier: https://forums.nesdev.com/viewtopic.php?t=13902 , or a slightly more complicated one: https://forums.nesdev.com/viewtopic.php?t=10554

Well to compare apple with apple, il going to use a transistor from another working Nes .

Should be fun. I’m traveling all day today so cyu tomorrow for the next episode of :
« The Rabbit hole to hell »

[Zonomi]

Which cable are you using ?

[romibraman]

Which cable are you using ?

Nes official scart

Works well with all my other FRE NES

[Zonomi]

So you're using the ugly composite to RGB box.
Try to cut the trace for the RGB lines, and connect directly the composite to the output.

[romibraman]

So you're using the ugly composite to RGB box.
Try to cut the trace for the RGB lines, and connect directly the composite to the output.

I totally would be except the composite signal I have before the RGB box already has the jailbars. So RGB in this case is not the issue

[romibraman]

Yeah.

You can use a super simple amplifier: https://forums.nesdev.com/viewtopic.php?t=13902 , or a slightly more complicated one: https://forums.nesdev.com/viewtopic.php?t=10554

Coax or twisted-pair might be clever at first, because your wires to your breadboard are probably not going to be particularly short and they'll pick up noise.

Ok so short update for me.

My attempt to recreate the composite signal On a breadboard has not been successful yet. I can’t get any signal on my TV. I m missing the right resistance value so that may be why.

The amplifier diagram above indicate I need at least two resistors of 33r and 220r and I don’t have either. I ve tried with 1 100 r between the E pin of the transistor and ground and no signal at all. I ve tried the same 100r in série between E and my composite cable and no signal at all.

Do you think it’s normal I have nothing on screen ?

I know my transistor is doing the job because if I run a wire between my beadboard E and the NES PCB transistor trace, I get the signal back on the NES RGB out. (Just to be clear I did remove the transistor from the NES PCB)

Another question all my NES and Famicom have this piece called FC2 at the very end of the composite trace. I’m not familiar with it. It looks like a small cylinder. What is it and do I need to add one to my breadboard ?

Thanks,

Romi

[lidnariq]

The amplifier diagram above indicate I need at least two resistors of 33r and 220r and I don’t have either. I ve tried with 1 100 r between the E pin of the transistor and ground and no signal at all. I ve tried the same 100r in série between E and my composite cable and no signal at all.

The 2N4401 (in this schematic) is an NPN transistor. The 2SA937 is a PNP transistor. They are not interchangeable.

The schematic on the NES-001 mainboard instead is this:

Code: Select all

               +5                                         +5             
                |                                          |           +5
               510Ω                                     5600Ω           |    
                |                                          +--C?---+  |/ 
                +--FB--+--- to RF modulator ---10uF--330Ω--+-3.3uH-+--|  
              |↙       |                                   |          |↘ 
Video In -----|       330pF                               330Ω          +--68Ω--+-- to RCA to TV
              |\       |                                   |           560Ω     |                 
                |      |                                   |            |       C? 
               gnd    gnd                                 gnd          gnd      gnd

You can flip the super-simple schematic over to use a PNP transistor:

Code: Select all

             +5V
              |
            220Ω
              |
              +--- Video Out
              |
             33Ω
              |
              +
            |↙
Video In ---|
            |\
              |
             gnd

but note that it'll add a large DC bias. Your TV may or may not object.

Another question all my NES and Famicom have this piece called FC2 at the very end of the composite trace. I’m not familiar with it. It looks like a small cylinder. What is it and do I need to add one to my breadboard ?

That's a ferrite bead. It's a tiny inductor. You shouldn't need it for testing.