Famicom AV Mod Situation in 2019

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

Moderator: Moderators

User avatar
Fisher
Posts: 1249
Joined: Sat Jul 04, 2015 9:58 am
Location: -29.794229 -55.795374

Re: Famicom AV Mod Situation in 2019

Post by Fisher »

Thanks for the tip.
Changing the capacitors made it way better, but still not perfect.
Interestingly the NROM and UNROM games I tested show no lines, while most games that use more complex mappers like MMC3 and VRC4 still show them after a while.

I made small resistor changes on the video and audio output to get a brighter image and louder sound, not sure if this can be causing the issues.
I must take a look at the schematics I found and write wich resistors I changed and post here to give an better idea of what was made.
I'll do that as soon as I can.
User avatar
Ben Boldt
Posts: 1149
Joined: Tue Mar 22, 2016 8:27 pm
Location: Minnesota, USA

Re: Famicom AV Mod Situation in 2019

Post by Ben Boldt »

Could you share a photo of the horizontal lines? Do you hear a humming in the audio? Maybe it could be a ground loop also.
User avatar
Fisher
Posts: 1249
Joined: Sat Jul 04, 2015 9:58 am
Location: -29.794229 -55.795374

Re: Famicom AV Mod Situation in 2019

Post by Fisher »

Yes, it has a hmmm on it.
It got a little better after the recap, but it's still there.
Here's a screenshot:
I think this shows the bar pretty well.
I think this shows the bar pretty well.
And here is the console:
Dynavision 3
Dynavision 3
His guts:
Guts 1
Guts 1
From another angle:
Guts 2
Guts 2
The problems disappear if I remove the board on the left, wich is comprised of a volume controller IC (Toshiba TC9153AP) and some transistors for audio amplification:
Audio comp.
Audio comp.
I couldn't add the other side of the PCB, sice 5 images is the limit.
Let me know if it's necessary.
I think I drawed this PCB's schematics some time ago, but I can't find it right now.
Probably it's a ground loop, wich I have no idea on how to solve.
Any help will be welcome.
Thanks in advance.
User avatar
Ben Boldt
Posts: 1149
Joined: Tue Mar 22, 2016 8:27 pm
Location: Minnesota, USA

Re: Famicom AV Mod Situation in 2019

Post by Ben Boldt »

That sure looks like a ground loop to me. Does the horizontal bar scroll or is it always at the same spot? (Not sure what that would exactly tell us but the console's video generation should not be perfectly synchronized to the AC line voltage.) You could try different/shorter A/V cables or cleaning the connectors. Maybe try a different power adapter.
lidnariq
Posts: 11432
Joined: Sun Apr 13, 2008 11:12 am

Re: Famicom AV Mod Situation in 2019

Post by lidnariq »

Frame rate on Brazilian Famiclones (ones that emit PAL-M natively) should be 60.03Hz, so should have a 31 second beat pattern (compared to the 17-second beat pattern that "real" NTSC has, the 10 second beat pattern than the 2C02 has, or the 143-second beat pattern that the 2C07 has)
User avatar
Fisher
Posts: 1249
Joined: Sat Jul 04, 2015 9:58 am
Location: -29.794229 -55.795374

Re: Famicom AV Mod Situation in 2019

Post by Fisher »

Yes, it scrolls down, when it reaches the botton it appears again at the top.

I've already tried a few cables and even other outlet.
The power supply is internal.
Maybe if I try to substitute the transformer with a switching power supply?
The interesting thing is that all this vanishes if I remove that small daughterboard.

I've found some schematics.
Here's the mainboard:
Esquema-Dynavision3-placa-logica.pdf
(78.01 KiB) Downloaded 306 times
Here's the power supply and the audio board:
Esquema-Dynavision3-placa-analogica-e-fone.pdf
(48.87 KiB) Downloaded 301 times
AFAIK, these schematics are from different PCB revisions from mine, I noticed:
- The motherboard is exactly the same on these schematics.
- The power supply board has some components numbered slightly different, probably due to the RF generator being changed to a RF module.
- The audio board is totally different than mine, not sure wich would be better.

I did some small changes to get better image and sound on my TV:
- I changed R12, that's on the base of the video out transistor, from 15K to 10K, with this the image that was too dark became "normal".
- I changed R28, that sits just before the audio output, from 5,5K to 1K, with this the audio became louder.

Could these changes be causing this trouble?
User avatar
Ben Boldt
Posts: 1149
Joined: Tue Mar 22, 2016 8:27 pm
Location: Minnesota, USA

Re: Famicom AV Mod Situation in 2019

Post by Ben Boldt »

Looking at connector CN2, presumably this comes off of a transformer, with the center tap connected to GND. It then rectifies the AC from the transformer using D1 and D2, and peak charges C6, which is the large 2200uF capacitor. With this type of power supply, there is always a 120Hz ripple which increases with more load. If C6 is going bad, the ripple will get worse, and will improve by removing load, i.e. removing the audio board. Based on your description of 2 large horizontal bars that scroll slowly, it makes sense.

The NES and Famicom have a similar large 2200uF 25V capacitor. With all of these old systems, I highly recommend replacing this capacitor even if it isn't bad yet or doesn't solve a problem anyway. I think there is a chance that this is what is wrong with yours. I could only see 1 horizontal bar in your photo, and I would expect 2 if C6 is bad but it is just going to be a good idea to replace that old capacitor anyway, worth a shot.
User avatar
Fisher
Posts: 1249
Joined: Sat Jul 04, 2015 9:58 am
Location: -29.794229 -55.795374

Re: Famicom AV Mod Situation in 2019

Post by Fisher »

Well, when I recapped it I already changed this capacitor.
Should I put a bigger one?
Strangelly I remember the 2 bars, but also only saw only 1 this time.
Maybe it was before the recap.
It was a while ago, I was with very little free time to get info and look for help. :-(

I've found the draw I did from the audio board:
Audio board.png
Not the best draw and probably has some mistakes, but it's what I have for now...
I'll try to get some "naked" scans for that board.
Maybe it's not needed since it's a single face PCB, 2 good pictures probably will do the same job.

Interestingly there's a V- signal that powers a flipflop.
There's also a smaller rectifier circuit with a BC549 and a zener wich gives the VDD.
AFAIK the VDD is there to make the volume IC "remember" the last volume level.
I tought this could be the cause and tried to unplug this pin on the audio PCB with no change.
User avatar
Ben Boldt
Posts: 1149
Joined: Tue Mar 22, 2016 8:27 pm
Location: Minnesota, USA

Re: Famicom AV Mod Situation in 2019

Post by Ben Boldt »

Fisher wrote:Well, when I recapped it I already changed this capacitor.
Should I put a bigger one?
Strangelly I remember the 2 bars, but also only saw only 1 this time.
Maybe it was before the recap.
Bummer. You could experiment by taking an additional cap and attaching it temporarily. You could also try that across the negative voltage rail (i.e. C20, 100uF). But I am thinking that now it pretty much points back to some sort of a ground loop.

I am not sure if I have understood something quite right. I know when you remove the audio board the problem goes away. Did I hear you say that also if you disconnect the external audio cables (with the board still installed), it also makes the problem go away? If so, maybe that is a clue.
User avatar
Fisher
Posts: 1249
Joined: Sat Jul 04, 2015 9:58 am
Location: -29.794229 -55.795374

Re: Famicom AV Mod Situation in 2019

Post by Fisher »

I tried to unplug one or other of the AV cables without success.
Them I tried to add a 1000mfd x 16v together with C6 and the bar got softer.
Since the next bigger one I had around was a 4700 x 25v I added it and the bar vanished.
I think it's a bit overkill, but seems to have solved it.
Still needs more extensive testing but looks OK.

Another doubt I have is how can I add the extra Famicom audio to the headphones.
I tried to use some resistors, but they either muted the extra audio or made the whole audio distorted.
maybe I should use a transistor?
lidnariq wrote:Two horizontal lines? That's almost certainly a power supply filter issue, and you probably just need to replace the capacitors in there.

What Ben Boldt and Ultron were investigating was a pattern of vertical lines, like what Ben Boldt showed here.
I totally missed that post, sorry :-(
I'm aware that I'm going pretty much off topic here, and even intended to open a new thread. But since we're already discussing it here, I let it go.
Your idea about the capacitor was correct. Not sure why the full recap didn't solve everything at the beginning.
Not very good capacitors maybe?
User avatar
Ben Boldt
Posts: 1149
Joined: Tue Mar 22, 2016 8:27 pm
Location: Minnesota, USA

Re: Famicom AV Mod Situation in 2019

Post by Ben Boldt »

The voltage at C6 might be more than 16V. You can't get away with a 16V cap in a real NES, I tried that once and it did vent after a while. Not sure about this console though. Increasing this cap in uF and voltage sounds like a good upgrade to me.
User avatar
Fisher
Posts: 1249
Joined: Sat Jul 04, 2015 9:58 am
Location: -29.794229 -55.795374

Re: Famicom AV Mod Situation in 2019

Post by Fisher »

Indeed a great upgrade!
Even the jailbars are very soft now! :-D
Fester's jailbars
Fester's jailbars
It was great to finally finish the fixing of this console.
I bought it as non-working at the beginig of the year.
It had many problems: burnt CPU, bad VRAM and a couple bad transistors.

The knowledge I used to fix it was almost all here at the forums.
In fact, I asked only about that last one because I was afraid it was a ground loop, wich I had no idea on how to start looking.

It still has some noise at the audio, but I think it's due to the design.
Unfortunatelly I have not much time to search for a "cure", so let it stay this way.
So thanks to you guys it now works perfectly. :beer:
Anomunus
Posts: 4
Joined: Thu Sep 26, 2019 1:34 am

Re: Famicom AV Mod Situation in 2019

Post by Anomunus »

Ben Boldt wrote:If you want me to tell you a specific value, go for 22uF 16V. But really I could tell no difference between any of those I listed above, so if you can find one from the list on an old motherboard or hard drive or LCD or something, it will be just as good. Special note: unlike electrolytics, the stripe means (+) on tantalums. It is very important not to get it backwards.
Hey just returning to share my results of building your circuit (as you previously showed here) with the tantalum you specified. My original intention of doing the AV mod was to see if I could fix my Famicom's being unable to show a colour image when I was using its RF output. Unfortunately it's still black and white (which means something else is wrong with it, and I don't know what), but the resulting image from this design is definitely sharper than the RF output. Also, can't tell if there's any jailbars honestly.

My implementation allows the board to fit back into the case perfectly; combined with my unacceptably poor soldering skill this meant having to lengthen the circuit a bit in order to make the components fit into place, as well as changing one key component.
I used the same resistors and 560pF ceramic specified. For the 2SA937 transistor that is supposed to be re-used, I failed to get it off the board properly (ended up accidentally tearing it off at all the legs). I replaced it with a NTE290A (equivalent to 2SA1015 which is apparently used in the Twin Famicom and has been used in this mod found here). The electrolytic capacitor is 220uF 35V. The tantalum is 22uF 25V.
The Hong Kong Famicom lacks a channel switch, so I am able to wire the RCA plugs through that hole without cutting the case in any way!
The Hong Kong Famicom lacks a channel switch, so I am able to wire the RCA plugs through that hole without cutting the case in any way!
Bottom of the board, where the circuit is attached to the PPU.
Bottom of the board, where the circuit is attached to the PPU.
I had trouble soldering the components to the pins at the top lol, so I just added wire in order to solder to the bottom. Notice how the electrolytic capacitors for both video and audio are tucked into negative space just out of clearance of the cartridge eject mechanism. As well, there cannot be anything along the long part of the cartridge eject because it sits on an extruding part of the Famicom's case bottom. The tantalum is also placed into negative space as it won't fit directly on the bottom of the board.
Front of the board.
Front of the board.
Here's two games I tested with, using a 28" Sharp flatscreen LCD.
Tetris Level 7. Looks fine.
Tetris Level 7. Looks fine.
Whenever I got to this level in Tetris while I was using the RF, many of those darker pieces would just blend with the background because of the quality of that output, but on the AV I can actually tell those dark pieces apart (and therefore not have major anxiety playing that part lol)
Super Mario USA (aka Super Mario 2)
Super Mario USA (aka Super Mario 2)
Honestly I don't know what's up with those diagonal lines. Probably something to do with the fact that my Famicom is in monochrome?

I know it's a bit off-topic, but if anyone has suggestions on what I can do to fix the monochrome please help me :(
Last edited by Anomunus on Fri Oct 04, 2019 9:22 pm, edited 2 times in total.
lidnariq
Posts: 11432
Joined: Sun Apr 13, 2008 11:12 am

Re: Famicom AV Mod Situation in 2019

Post by lidnariq »

Adjust the clock crystal for the famicom.

... Why is the trim cap missing?
crop_IMG_3684.jpg
crop_IMG_3684.jpg (50.52 KiB) Viewed 12017 times
TVs won't decode NTSC color unless the frequency is within 3ppm of correct (officially. Real TVs are probably more tolerant); your Famicom is probably running several hundred ppm too fast due to the missing trimcap.
Anomunus
Posts: 4
Joined: Thu Sep 26, 2019 1:34 am

Re: Famicom AV Mod Situation in 2019

Post by Anomunus »

lidnariq wrote:Adjust the clock crystal for the famicom.

... Why is the trim cap missing?
crop_IMG_3684.jpg
TVs won't decode NTSC color unless the frequency is within 3ppm of correct (officially. Real TVs are probably more tolerant); your Famicom is probably running several hundred ppm too fast due to the missing trimcap.
I'm going to start a new thread, so that this thread doesn't get derailed, because my post will be more about Hong Kong Famicoms than about AV modding (now that I've completed this circuit successfully). Let's continue the discussion there.
Post Reply