Damaged PPU on top loader

Discuss hardware-related topics, such as development cartridges, CopyNES, PowerPak, EPROMs, or whatever.

Moderator: Moderators

alfredocalza
Posts: 111
Joined: Sat Jun 22, 2019 1:06 pm
Location: Madrid, Spain

Re: Damaged PPU on top loader

Post by alfredocalza »

Ok, I have a new update. I tried placing tantalum capacitors between 5v and gnd of the ppu. I tried 22uf and 47uf. It made no difference whatsoever. Is there a permanent solution to this? Could I just cut the original trace of the video line that goes out of the ppu through pin 21? If I get rid of the trace would that eliminate the interference? I am not using that trace since I have the nesrgb installed. However, the socket is still soldered to that trace on the motherboard. Could that be the source of the problem? Which components are the ones that create noise in the top loader board?
User avatar
Quietust
Posts: 1920
Joined: Sun Sep 19, 2004 10:59 pm
Contact:

Re: Damaged PPU on top loader

Post by Quietust »

alfredocalza wrote: Fri Apr 22, 2022 8:09 pm Could I just cut the original trace of the video line that goes out of the ppu through pin 21? If I get rid of the trace would that eliminate the interference? I am not using that trace since I have the nesrgb installed.
From what I understand, NESRGB effectively uses the original PPU's analog video output line as an "EXT4" output so that it can distinguish sprite pixels from background pixels, so if you disconnect that, all sprites will be displayed with colors from the corresponding background palettes.
Quietust, QMT Productions
P.S. If you don't get this note, let me know and I'll write you another.
alfredocalza
Posts: 111
Joined: Sat Jun 22, 2019 1:06 pm
Location: Madrid, Spain

Re: Damaged PPU on top loader

Post by alfredocalza »

Well I wasn't really thinking about disconnecting it in the nesrgb board. What I had in mind was to just cut the video line on the motherboard, which is not being used. Could that line be the source of the noise, or is it the 5v line from the ppu more likely to be the culprit? Which components are the ones that cause noise in the nes top loader board?
User avatar
Ben Boldt
Posts: 1149
Joined: Tue Mar 22, 2016 8:27 pm
Location: Minnesota, USA

Re: Damaged PPU on top loader

Post by Ben Boldt »

I don’t really know what is going on here but I think you should start as non-destructive as possible. I think it is worth coming up with a better theory before cutting traces.

The jail bars are a visual form of electromagnetic interference, aka EMI. EMI can be transmitted in 2 ways: radiated and conducted. Radiated EMI goes through the air like a radio station and conducted EMI goes through wires.

For all intents and purposes, I believe that your jailbars originate from the ALE signal, since it is synchronized and the frequency corresponds. I think we can say that for pretty sure that is the single point of origin.

An example how ALE could be radiated: the PPU puts out the high-frequency ALE signal. This goes through a long trace which acts as a transmitting antenna. Something in the video path of the NESRGB then also has a long trace that acts as a receiving antenna, which is shown as jailbars.

An example how this can be conducted: the PPU puts out the high-frequency ALE signal, which drives into some impedance, possibly even internally. This impedance draws alternating current from the 5V rail powering the PPU. The 5V power rail voltage may then ripple at this frequency. If the same 5V rail runs the video amplifier in the NESRGB, that could be conducted into the video, shown as jailbars.

An example how it may be some of each: ALE drives into some impedance again, PPU draws current from 5V rail again. The alternating current on the traces of the 5V rail radiate the ALE signal. A long trace in the NES RGB’s video path picks it up again.

We do not know yet if it is radiated or conducted because you haven’t found anything that affects it yet. Conducted EMI can be affected with capacitors. Capacitors can make it better or worse. Worse is OK too; that’s still useful information if it can make it worse, as that may lead you to reduce capacitors or add inductors, etc. The clue that it got worse is what starts you down that path. Try different size capacitors here and there. Small ceramic up to maybe 10uF may be appropriate for this. Larger capacitors are basically too sluggish to help at these high frequencies. (Bigger isn’t better here in other words.)

Radiated EMI can be affected with shields. Adding or removing shields may make it better or worse. Do keep in mind, as you have actually already run into, shields are conductive. For your experiments, you can take a piece of aluminum foil and completely tape it on both sides. Then just try moving it all around, in different crevices, roll it around wires, etc. Just look for anything that affects the jailbars, better or worse. The foil does not need to be electrically connected for these experiments. Heck, try foil around the cartridge too; be creative to find your clues.
alfredocalza
Posts: 111
Joined: Sat Jun 22, 2019 1:06 pm
Location: Madrid, Spain

Re: Damaged PPU on top loader

Post by alfredocalza »

Ben, thank you so much for the info. This is very helpful. I took a 4.7uf ceramic capacitor and put it on the 5v pin and the nearest ground. Some jailbars disappeared. However, the ones that remained got much stronger. I will put a larger capacitor again between ground and 5v and see what happens. Also will try with 0.47uf and 1uf ceramic. I'll get back with the results.

Question, why put a capacitor between the 5v pin and the gnd pin? Why not use another gnd source which is closer to the 5v pin?
User avatar
Ben Boldt
Posts: 1149
Joined: Tue Mar 22, 2016 8:27 pm
Location: Minnesota, USA

Re: Damaged PPU on top loader

Post by Ben Boldt »

alfredocalza wrote: Sat Apr 23, 2022 6:50 pm Question, why put a capacitor between the 5v pin and the gnd pin? Why not use another gnd source which is closer to the 5v pin?
It probably doesn’t make much difference. When using the original composite output, the PPU “doing stuff” internally might benefit from a cap closest to its power pins, giving it the most stable possible input voltage. But if what you are seeing with RGB is what gets conducted out of the power pins of the PPU, I wouldn’t say that you necessarily benefit going straight to the pins like that anymore. It is all just kind of ghosts and magic though. You never know what will work best.
alfredocalza
Posts: 111
Joined: Sat Jun 22, 2019 1:06 pm
Location: Madrid, Spain

Re: Damaged PPU on top loader

Post by alfredocalza »

I was able to reduce the jailbars substantially, but wasn't able to fully remove them. I put two 4.7 uf caps in PPU 5v (total 9.4uf). I also put another 4.7uf in CPU 5v. Also fitted the voltage regulator with a 470uf electrolytic, and put a 0.1uf ceramic between pins 22 and 20 on the PPU. I have less jailbars, but now instead of them being evenly spaced, they appear in sets, followed by a part of the screen without any jailbars and then followed again by a set of jailbars.

I can live with it as it is right now. However, I now kind of feel a little stupid because in all of this time I forgot that the nesrgb has an option to be fed with an external voltage regulator. In order to do this I would just need to hook it up and open jumper J3 to cut the current coming in from the ppu. Do you think that this would completely remove the jailbars?
User avatar
Ben Boldt
Posts: 1149
Joined: Tue Mar 22, 2016 8:27 pm
Location: Minnesota, USA

Re: Damaged PPU on top loader

Post by Ben Boldt »

Did you try the capacitors one at a time? Which ones made it better or worse?

I am not sure if pin 22 is 5V on your model. The capacitor there was meant for original Famicom, but wouldn’t work on front-load NES. I am not sure about your model. This is a top-load 72-pin NTSC, is that correct?
alfredocalza
Posts: 111
Joined: Sat Jun 22, 2019 1:06 pm
Location: Madrid, Spain

Re: Damaged PPU on top loader

Post by alfredocalza »

Yes, I tried everything at once. I wasn't sure if some of them made a difference at all, but at the end I put them all. The ones that really did make a difference where the two 4.7uf ceramics that I put on the 5v PPU. I also think that the 4.7uf that I put on 5v CPU softened the jailbars just a tiny bit (it was my impression, maybe I'm wrong). Please let me know if I should remove the capacitor on pin 22 of the PPU. I do not have a famicom; I have a top loader. Also, let me know if there would be a benefit in fitting an external voltage regulator. I already figured out where to put it. Here is a picture of the things that I tried yesterday:
20220424_113844.jpg
User avatar
Ben Boldt
Posts: 1149
Joined: Tue Mar 22, 2016 8:27 pm
Location: Minnesota, USA

Re: Damaged PPU on top loader

Post by Ben Boldt »

An external voltage regulator may help, yes. I am not sure why they offer an external regulator; maybe for this very reason!

You should test each cap individually because some caps might make it worse. You said you had some new, smaller jailbars show up. It might be one of the caps that added the small jailbars. Probably not, but possible. If you have the time and patience, it would be interesting to know exactly what does and doesn't help one at a time, especially if other people come here with the same problem, then it helps them too.

Pin 22 is 5V on a normal Famicom, so it is convenient to put a capacitor from there to pin 20 (GND). However, a front-load NES uses pin 22 as a reset signal from the CIC chip. I do not know what the top-load NES does with pin 22, so I can't recommend putting that there unless someone knows that.

Your work looks good, nice job.
alfredocalza
Posts: 111
Joined: Sat Jun 22, 2019 1:06 pm
Location: Madrid, Spain

Re: Damaged PPU on top loader

Post by alfredocalza »

Bad news. Installed the 2nd voltage regulator, fed it with 9 volts from the 1st voltage regulator, connected the middle pin to ground and the outer pin to the nesrgb board. Drilled a hole in the original heat sink and places it there. The installation took me some time and effort. However, I am back to day 1 with the EXACT same jailbars that I had before I installed any of the capacitors. This leads me to believe that the noise is rather coming from the 9 volts line. Have you seen how this 9v line is wired? It looks like a complete mess. The large 1500 uf capacitor is placed before the on/off switch. After the capacitor the line is split into many different lines going all the way across the noisy part of the board, until it reaches the on/off switch. Then it comes out of the on/off switch and it is split up again, goes back through the noisy part of the board (where the crystal oscillator is) and then into the original voltage regulator. You can refer to my previous picture and see the detail of the 9v line trace there. Could this be the source of the jailbars problem in the nes top loader? Should I just cut those stupid 9 volts traces and re-wire the whole thing??
alfredocalza
Posts: 111
Joined: Sat Jun 22, 2019 1:06 pm
Location: Madrid, Spain

Re: Damaged PPU on top loader

Post by alfredocalza »

Ben, to answer your question regarding the caps. I did test them separately, and the ONLY capacitors that helped reduce the jailbars, but not completely remove it, were the two 4.7uf that I put between the 5v pin of the PPU and the nearest ground (refer to my previous picture). The first one definitely helped, but when I placed the second one in parallel, it only made a very slight improvement. However, these are not doing anything right now on my board because I installed the 2nd voltage regulator which is now powering the PPU and nesrgb board. Here is a picture:
20220424_185958.jpg
alfredocalza
Posts: 111
Joined: Sat Jun 22, 2019 1:06 pm
Location: Madrid, Spain

Re: Damaged PPU on top loader

Post by alfredocalza »

alfredocalza wrote: Sun Apr 24, 2022 4:37 pm Bad news. Installed the 2nd voltage regulator, fed it with 9 volts from the 1st voltage regulator, connected the middle pin to ground and the outer pin to the nesrgb board. Drilled a hole in the original heat sink and places it there. The installation took me some time and effort. However, I am back to day 1 with the EXACT same jailbars that I had before I installed any of the capacitors. This leads me to believe that the noise is rather coming from the 9 volts line. Have you seen how this 9v line is wired? It looks like a complete mess. The large 1500 uf capacitor is placed before the on/off switch. After the capacitor the line is split into many different lines going all the way across the noisy part of the board, until it reaches the on/off switch. Then it comes out of the on/off switch and it is split up again, goes back through the noisy part of the board (where the crystal oscillator is) and then into the original voltage regulator. You can refer to my previous picture and see the detail of the 9v line trace there. Could this be the source of the jailbars problem in the nes top loader? Should I just cut those stupid 9 volts traces and re-wire the whole thing??
I'm thinking about this and re-reading Ben's post regarding the ALE signal. I followed the trace that carries composite video out of the PPU, and there is about a 1/4 of an inch long part of this trace that gets very close to the ALE signal before it enters a transistor to amplify the video. Would this be the antenna capturing the noise? Can this interference be brought up to the nesrgb through the video out trace of the nes motherboard?
User avatar
Jarhmander
Formerly ~J-@D!~
Posts: 569
Joined: Sun Mar 12, 2006 12:36 am
Location: Rive nord de Montréal

Re: Damaged PPU on top loader

Post by Jarhmander »

That may sound stupid, but have you tried using a small ceramic capacitor (100nF) in parallel with your 4.7μF?
((λ (x) (x x)) (λ (x) (x x)))
alfredocalza
Posts: 111
Joined: Sat Jun 22, 2019 1:06 pm
Location: Madrid, Spain

Re: Damaged PPU on top loader

Post by alfredocalza »

Jarhmander wrote: Mon Apr 25, 2022 3:37 am That may sound stupid, but have you tried using a small ceramic capacitor (100nF) in parallel with your 4.7μF?
Is that because the smaller capacitor might be faster? I will try this and get back. Also, if ALE is causing the alternating draw from 5v, wouldn't this mean that the PPU is damaged after all?

I will try to source another PPU maybe from ebay. A broken console or something like that.
Post Reply