"NES Super 8" project

[low_budget]

The FCC does RF interference testing (in the USA anyway.)

Testing to ensure the electronics both A: do not emit RF interference and B: are not susceptible to RF interference.

 

The large steel shielding present on retail video game systems and electronics is required by FCC regulations, it isn't needed for good quality picture. I've never seen any difference in picture from removing this shielding from a game console, except in the Vectrex where high voltages are present.

 

Any testing Nintendo would do is to ensure their electronics pass and are therefore legal to sell.

Naturally there would be quality control testing by Nintendo to ensure good picture with the available TVs at the time, but this would be probably be done before the FCC testing.

 

Nintendo released a variety of NES PCB revisions with the picture quality varying considerably over the years.

The front loader has generally good composite video and RF.

The 1993 Nintendo NES top loader has bad jail bars, so Nintendo must have rushed quality testing on these PCBs. It is only the rare PCB redesign of these systems that don't have jail bars. (and the AV famicom.)

 

The RGB PPU used in the Nintendo arcade systems inherently has some jail bars, but I think this is due to the chip never being designed for high resolution displays. When played on a CRT monitor the picture is perfect. The jailbars here cannot be eliminated entirely, only covered up. I've seen many solutions trying to fix this and they either don't help or have negligible improvement.

I don't think solutions I've read like covering a PPU chip in copper foil would offer any improvement, I have lots of ground plane coverage under the PPU that accomplish the same thing.

 

I will do a re-test with my GBS-8200 VGA converter, as this seemed to amplify any jailbar presence. The s-video converter didn't show any jailbars.

 

RF interference is only an issue if the PCB design is really bad (like the vectrex.) The major cause of bad jailbars are PCB traces that cross important video signals and/or poor grounding. That's why the NES top loader looks like crap compared to other NES systems.

 

The issues with the RGB picture quality on the Super 8 that I thought were issues were instead caused by the Powerpak.

The composite PPU from a NES is very sensitive to timing (clock signal and data), but the RGB PPU is not.

 

In the process of designing the Super 8, I've analyzed different NES, Famicom, and arcade PCBs and schematics and compared the picture of each. I'm on my 4th revision of the NES Super 8 and the picture has kept getting better with each version, so I must be doing something right, LOL.

Edited June 24, 2013 by low_budget

[+grips03]

The difficulty with adding a RGB converter to the Super 8 on-board is room. The RGB amp and sync cleaner I am currently using only require a few support components, while the CXA 2075 needs about twice as much room as what's currently available in this area of the PCB.

The only ways I could think of to accomplish this are:

I could make the PCB bigger, but I currently have everything laid out in a nice compact rectangle and this would give the board design a tumor. It would no longer fit in the AG-85 case.

I could make a module that plugs into header pin sockets on the Super 8 PCB, but this would add cost.

I could use all surface mount resistors and caps, but that makes it hard to assemble or modify the system.

 

I do not think I can incorporate the CXA2075 or similar RGB converter into the main PCB for v2.0.

 

I often find that adding one feature leads to compromises somewhere else. I've had lots of good suggestions for extra features and whatnot, but it almost always involves a compromise elsewhere, usually increased PCB size and/or extra cost.

 

 

The v1.8 and later boards have:

The RGB amp's input about as close to the PPU output pins as possible.

The sync line is short, doesn't cross any data lines, and is filtered by the LM1881.

The ground plane covers a large area to shield key data and a/v signals. Lots of shielding under the CPU and PPU.

 

I wish I knew earlier that the powerpak will cause interference in a RGB NES, it would have saved me a lot of time. I thought I had video issues to troubleshoot, but the picture quality is excellent (as long as you play the original carts or the new Krizz flash cart.)

 

I was so dedicated to the RGB Super 8, I haven't been doing much testing with the composite PPU Super 8 PCB. I want to do more testing with 74LS 39 and 74LS373 instead of the 74HC139 and 74HC373 I was using in the RGB system. I was also going to make some grounding changes to this version. Including the CPU and PPU chips, this system is about half the cost of the RGB version, so there maybe could be some interest?

 

If you do change your mind here is easy to use RGB to component circuit that works well.

http://circuits.linear.com/158

[+grips03]

Do you have v1.8 with composite for sale. Looking for complete system with case.

Edited June 27, 2013 by grips03

[Game-Tech.US]

V1.8 initial testing and rgb and YPbPr testing YT

, lots of criticisms for Low_Budget, but he asked for it!

[+grips03]

V1.8 initial testing and rgb and YPbPr testing YT

, lots of criticisms for Low_Budget, but he asked for it!

 

thanks for posting. You lab is so much cleaner than mine. Good recommendations on the NES Super 8.

 

Is that a suppression diode wired on the DC input or something else?

[Drakon]

The nes top loader jailbars are completely because of pcb layout. Years ago I composite video modded a top loader. I had the unamped composite video wire loose. I had the system running and moved the unamped composite wire closer and further from the pcb. I watched on my tv as the jailbars appeared and disappeared as I moved the wire closer and further from the board. Unamped video / audio signals + noisy pcbs don't get along very well. The problem with unamped analog signals is they pick up unterference easily, this is why moving the amp as close as possible really makes a huge difference. Once a signal is amped it's much harder for it to pick up noise.

 

 

 

This is the nes-cpu-11 revision pcb, one of the worst in terms of jailbars when rgb modded. I point blank range soldered in the rgb amp and it's very clean looking. You can further reduce jailbars by improving pcb layout. My av famicoms / twin famicom have even less jailbars than this thing, it's only a small difference though.

Edited July 19, 2013 by Drakon

[low_budget]

Clarifying some things mentioned in the review video.....

 

I plan on offering the Super 8 assembled, not as a kit. I hate bagging lots of small parts and making sure they're all there. Plus I don't like to do tech support if somebody has issues with assembly.

I normally use precision IC sockets for the PPU.

All vias on the PCB must be the same size, so I opted for large ones. Tenting the vias cost extra and for this type of PCB I found it unnecessary.

The controller port PCB mounting holes are correct on v2.0. This is due to the v1.8 design being rushed and changing the controller port PCB after the order was placed.

The "aux" adjustment pot is for expansion audio. Pin 54 goes through this pot and is mixed with audio channel 1. Expansion audio should work as long as pin 54 of the cart slot is receiving audio.

The reset button can be mounted to the motherboard or the controller port PCB, I wouldn't use both. A jumper wire is needed, I could have used a header pin between the controller port headers, but didn't.

The 5v regulator's voltage is set by the 240 ohm resistor next to it. It's about 5.2v, but can be set to 5v with a 270 ohm resistor.

In my opinion, the PTH08080 is a great regulator and not the source of any issues.

The clock circuit has been simplified on v1.8 and v2.0, this circuit is based on the NES top loader while earlier circuits were based on the front loader. Again, I am thinking the clock circuit might need some tweaking.

I like to isolate the grounds of video, audio, input power. It may not be necessary, but it certainly can't hurt. Isolating the video ground improved the picture quality (for me.)

I agree there probably was no benefit to the separate audio ground, but yeah, it can't hurt.

Without being in a case, the game cartridge may lean front or back and if moved the game may glitch. Games need to be clean to work reliably.

 

Thanks again for the review Akaviolence!

[low_budget]

Haven't updated this post in a while, but I've been busy.

 

I have the v2.0 PCBs available on Tindie.com

 

It has the I/O circuitry redone to allow wiring of a Famicom expansion port, improved audio circuit, and better trace routing overall.

 

I also made a separate CXA2075 based RGB to s-video / composite converter board that works great with the Super 8. I'm doing testing to see if I should just eliminate the THS7314 and tap RGB for it directly off the PPU.

 

The picture with the converter is good but could be better, so I'll be working on this in the next month.

[Hyperboy]

This Rocks \m/

[Game-Tech.US]

I also made a separate CXA2075 based RGB to s-video / composite converter board that works great with the Super 8. I'm doing testing to see if I should just eliminate the THS7314 and tap RGB for it directly off the PPU.

Keep in mind the 2075 draws 100mA while the ths 20mA...

[Donking2000]

This is a very nice project. I wait for my V2.0

[retroillucid]

It looks like I completely missed this thread

 

Want one!

[baby]

Great works awesome

[low_budget]

I've found that the CXA2075 circuit I'm using outputs really good s-video (not the best, but yes really good.)

The composite video isn't as good as I hoped, however.

Also the RGB output with the chip isn't as good as the THS7314.

I've been trying many fixes and had only limited success

 

The schematic for the CXA2075 circuit is based on the NTSC application note available from here:

http://www.videoi.com/~pietro/cxa2075/

 

I have tried the CXA2074 circuit with and without 220uF caps in the Y, C, and Composite outputs, seems to be slightly better without them. The RGB converters I've tried didn't work when using the caps in the RGB outputs so I left those out.

I have the 47uF and .01uF caps very close to the power pins of the CXA2075 and some distance between the two power pins. I may need to adjust these values.

I'm using a 3.57....whatever MHz oscillator for the color subcarrier frequency.

I tried connecting the CXA2074 circuit to the PPU's RGB outputs, bypassing the THS7314. The picture was darker but had more contrast.

 

 

I made some more videos using my low budget production values:

I really need to get a better camera and/or tripod.

 

NES Super 8 v2.0 with RGB converted to upscaled VGA output:

 

Issues with CXA2075:

http://www.youtube.com/watch?v=lZiAAwVRGJk

 

 

 

[baby]

@Low_budget

have you see this project :

http://universalppu.com/

i think is it a good altenrative at the RP2C03B

[low_budget]

@Low_budget

 

have you see this project :

 

http://universalppu.com/

 

i think is it a good altenrative at the RP2C03B

 

Yes I have been following the universal PPU project. It looks very promising.

I believe it is in the prototype phase at the moment.

 

I could have many advantages over the original Nintendo PPU like sharper picture, fewer glitches, and exact same color palette as NES.

It's quite a bit larger than the RP2C03B, so I may need to make some changes to make it compatible with the Super 8 when it's released.

It is just one guy working on it in his spare time, so it may be a while before finished the finished PCBs are ready.

[Asbrandt]

Also worth mentioning is this, which was just revealed and is surprising in implimentation... Well, I'll leave it to the reader.

shmups.system11.org/viewtopic.php?f=6&t=47617

[HARMIK]

That looks very interesting.

[+grips03]

Also worth mentioning is this, which was just revealed and is surprising in implimentation... Well, I'll leave it to the reader.

shmups.system11.org/viewtopic.php?f=6&t=47617

 

looks very good.

[low_budget]

Wow I just heard of the NESRGB and it already has a release date in less than 2 weeks?

 

That's really cool.

 

It looks like it will work with the NES Super 8 too!

 

 

 

Thanks for finding the article Asbrandt

[Asbrandt]

It seems viletim kept it under wraps until it was ready and working and considering how unbelievable it is, I can't blame him.

 

Be interesting to see what a board layout specificly designed around this thing would look like, though, since you'd be able to drop some parts off the bill.

(Not saying I expect such a board, just saying I'd be curious.)

[+grips03]

it would be a composite NES Super-8, right? as Tim's board does RGB, etc.

[NES_Pal]

Low_Budget This is AMAZING!!!

 

I've only recently got the bug for older game consoles from my youth and this is a Wicked mod/ Remake.

 

Anyway, I'm in Australia and the systems are PAL will I be able to make the console work on our chips? (do I need anything else?)

Not that technical but there are heaps of wonderful helpful people online :-)

 

I know that you have put up heaps of info regarding all the versions of the console but if it were possible to consolidate "what you get and what you still need to get to run a NES Super 8" that would be sweet for us dummies :-) (ie. Board, Chips, power, AV cables required ect) Version 2 that is

[bombidragon]

So when are you going to collaborate with the NESRGB guy?

[Drakon]

Low budget is mailing me a super 8 and I already have my hands on two early nesrgb kits. You can use your imaginations with what will happen.

Edited October 18, 2013 by Drakon