SNES vertical bar: Quick and easy fix.

You're right about (dis)charging, I was talking about the steady-state filtering of "noise" on the signal; either one could be a limiting factor depending on R, which isn't known afaik.

 

Hi!

I got a SNES Jr, and I also got the vertical "bars", one in the middle and one on the right side.
Tried with a 220 uf cap on the 7805, with minor differences.
But where do I put the other one? Can't find the S-ENC on the pcb...

 

Hello, so far not all console versions respond the same to this fix. It works great on my CPU-GPM-02 but mostly useless on my old SHVC model.
If you want to try it on your jr, the second cap would go on the video encoder but It will be a different one on your console. You would need to find the encoder and its datasheet and then put the cap positive leg to Vcc and the negative to ground.

 

Hi, it's me again. I finally managed to fix my 1-Chip-01.
What did I do different this time? I used the right brand of capacitors! Apparently the SNES just loooooves Panasonic caps and refuses to get fixed with anything else...

Anyway here the final results:

Before:


After:


I used 3 no-name 0.1uF ceramic capacitors, 3 Panasonic 220uF and one Panasonic 2200uF cap:




I actually replaced the 1000uF cap that came with my SFC with a 2200uF since I noticed that a PAL SNES also got a 2200uF and here in Germany we got 220V mains and not 100V like in Japan.
The 3 220uF caps I placed on the output of the 7805 voltage regulator and on the two 5V supply pins of the S-RGB.
I also threw in 0.1uF caps for good meassure but they didn't do anything because I had them in before too and still had the vertical line.

So if you have tried this fix and it didn't help get some high quality Panasonic capacitors. Before I had Philips and they didn't do anything at all.

Caps I used:
http://www.conrad.de/ce/de/product/...-x-L-63-mm-x-112-mm-Panasonic-ECA1CM221I-1-St
http://www.conrad.de/ce/de/product/...-x-L-125-mm-x-20-mm-Panasonic-ECA1CM222B-1-St

 

Mains shouldn't matter, both the 100v and 200v Psus still output 10v or so.

But this certainly just looks like a power issue. It also confirms the issues I saw when designing my PCB on my scope.

 

Interesting that changing the brand of the cap worked. I usually stick to nichicon and panasonic but I often get to use no-names as this is what the local stores tend to keep...
Glad you got it working, the pictures speak by themselves!

 

I doubt it's the brand, but the esr rating etc.

A cap isn't just a cap, there are other characteristics that are often overlooked

 

Caps I used before that didn't do the trick: http://www.bg-electronics.de/datenblaetter/Kondensatoren/RSP 036.pdf
Magic Panasonic caps that worked like a charm: http://www.produktinfo.conrad.com/d...-01-en-ELKO_ECA1CM221I_M_R_220myF_16V_85C.pdf

I don't know where exactly to look which one has better ESR and so on, both are the 220uF 16V variants.

 

I think you are right, it would be interesting to find what specific property of the cap is at work here.
Both caps seem pretty similar in specs but I cannot find ESR rating on the Panasonic.
The ripple rating is the same for both.

 

I've just had a peek and the ones I used were from the Panasonic FR series. Datasheet here.
Probably had them left over from an order in multiples of five. That would make sense!

 

would the capacitor on 7805 affect in some way the 50/60 mod, having take +5v and gnd for the switch?

 

It shouldn't - I've installed them on two SNES' with SuperCICs (one a standard SNES, the other was a 1-chip) and no troubles with switching refresh rates.

 

The S-ENC, S-ENC B, and BA6952 all have identical pinouts (and are component-moddable, oh joy!). The fix for these encoders is to attach a suitably-sized cap (I personally use 220uf, though 470uf caps are fine, as well) between VCC and GND, with the negative leg of the cap heading towards ground. I find it easier to use the ground on the encoder chip itself rather than elsewhere. However, on the RGB and RGB-A, the pinout is VERY different, so in order to use this mod on these chips, you'll need to attach the positive leg of your cap to Pin 19, and the negative leg to either Pin 13 or 16 - I prefer 13, as I don't need to worry as much about accidental solder bridges, thanks to it being an end-pin.

 

Just did the mod on my 1-CHIP-01 SFC yesterday.
One 16V 220uF cap between the G and O pins of the 7805 and one between pins 6 and 11 of the S-RGB encoder chip.
Worked like a charm, even though I didn't notice the bar in many games, I got rid of it in the FF6/3 intro.

 

The ghosting can be reduced or eliminated via a Low Pass Filter.

The person who posted this thread states that he tested the DNR on Pioneer Plasma with a 1CHIP-03 SNES, and set the DNR to the "High" thus reducing the ghosting by 70%. This leads me to believe that someone with good tech knowledge (and good soldering skills) could easily pick up the right Low-Pass Filter (from eBay perhaps) and install it to fix all issues (including the vertical bar). Afterall, those two issues are noise related, and a LPF could kill two birds with one stone.

 

That's right. Using the modes with stronger filtering on my XRGB-mini does help a lot. I would still like to get rid of this problem at the source though

 

Question, this fix is based on adding a 2 capacitors to filter the 5V which feeds the chip.

Wouldn't it be possible to remove the 7805 and feed the console using an external powersupply? This should offer more stable power for the videocircuit.

Another possibility would be replacing the 7805 by another part, i found the following upgrade mainly intended for cdplayers/dacs/...

http://www.newclassd.com/index.php?page=70

 

I suspect the interference is generated inside the SNES rather than coming from the outside or the regulator.
A 7805 is not overly noisy but the SNES did not have much in the way of 5v rail filtering.

I did try to power an SNES with a mini switcher similar to what you linked. Switcher are usually very noisy but using an inductor coil on them seems to help a lot. Without the coil I had scrolling jailbars on the picture.
I had pretty good result with the switcher + coil. The bar was still visible but I felt it had a more purplish taint. Weird uh?
http://www.sega-16.com/forum/showth...-Buck-converter-step-down-switching-regulator

 

Nice, someone who has actually tried this! Too bad it's not a fix, i will just add capacitors to my Superfamicom once i get it.

Would have been an easy fix for the people like me, i can solder but have no clue about designing circuits or how they actually work .

 

Just be aware it does not seem to works as good on every console revision and in any case, it does not remove 100% of the bar.
But in the best circumstances it can reduce it to near invisibility.