Famicom av mod instructions?

[famifan]

well, i will try if i have the such high accurancy device. What is blocking you from answering some my question about your equipment? and how did you measure it.

e.g. if you use true 1% tolerance equipment and choose 0-10V gauge, your measurement accurancy step will be 10V*1% = 0.1V. In that case your device can't guerantee distinction between 4.90 and 4.95 even you've got it on the screen.  Should i continue feeding you? 

what is your device tolerance/accurancy?

facepalm.jpg
"1% tolerance" is about little bit different thing, but nice try I will try to explain in simple words.
Let's say you have 10bit ADC and measuring 0-10V range
What is minimum step of measuring? 1/1024th of the range, which consist of 1024 steps.
Is it accurate? Yes. What is the value of voltage if we read 55 from ADC? It's exactly 55/1024th of maximum range. But... what about precision voltage source for ADC? It's analogue, not discrete. And it might have "1% tolerance", so readout will be little bit off. But if this source not drifting during measurement, nothing wrong - you will get relative value of voltage and still can see 1/1024th steps on your measurement.

LOL, LOL, LOL again. Serious equipment has some certificates that proves it's accurancy (let's say for 1 year or 2). And it directly says, that this device has xx.xx% tolerance for DC measurements, yy.yy% tolerance for AC and so on.

and if you got two results from ADC
first is 55/1024th
and the last is 999/1024th. The last is better because it contains less ADC quantization errors.

So, my answer is "my device is accurate enough to measure changes more than 0.01V(0-10V range) on relative scale". Relative - this is keyword.

so, in that case you device need 0.05% basic occurancy or even better.
it seems that you have the certificated seriuos device that costs more then 1k$ ... Really  ?

http://en.wikipedia.org/wiki/Multimeter#Digital

For example, a meter reading 5½ stable digits may indicate that one nominally 100,000 ohm resistor is about 7 ohms greater than another, although the error of each measurement is 0.2% of reading plus 0.05% of full-scale value.

troll harder

[famifan]

don't care about me.

So, did you agree that you can't interpret your measurements correctly?

[Asprate]

I've done some testing with my famicoms. If i use the same famicom and tries Super Mario from a normal cart i get big jailbars, but if i play the same game from my FDS i get really weak jailbars  


Feels like there's something with the carts itself?  

[famifan]

So, did you agree that you can't interpret your measurements correctly?

Nope.

Ok, back on track. GPM-02 and 07 have similar layout for CHR Axx line, so this is not the case, but grounding. I will make one experiment, PowerPak related and update results.

do you have an oscilloscope?

i don't know what you've measured, but with my shitty analog oscilloscope i've seen that VCC is polluted by noise with the PPU/CPU clock frequencies. Yeah, the AC shunt capacitor around every IC reduces that noise, but it is still present anyway. And the Peak-to-Peak noise amplitude is high enough to cause the differences like 4.9 and 4.95.

also, with my unserious uncertificated equipment i can't see any notable VCC differences. Also i've checked the copper layer resistance and it is too small to make this huge VCC reducing.

i beleive you are smart enough and can give me your copper layer resistance approximation that can cause 0.5-0.10 voltage drop. ORLY?

[GohanX]

At the risk of interrupting the last couple of pages of fun, I just AV modded the Famicom I got from 80sFreak, and thought I would throw in my two cents.

I used the directions in the second post of this thread, using the Q1 transistor removed from the board, and used Da Bear's recommendation of using a cap between pins 20 and 22. The result was pretty good, but I had faint jailbars that were particularly noticeable with black screens, like the beginning of Gradius. I took it apart again, lifted pin 21, and put in some ghetto RF shielding around the video amp (probably didn't do anything, but it can't hurt) and the jailbars are completely gone.

One thing worth noting, the video is pretty clear, but it doesn't look quite as bright and colorful as my AV modded NES 2. I'm going to check my notes to see what resistor/cap values I used on it, as I'm not sure if it has something to do with the video amplification or if it's simply a hardware difference. I may duplicate the NES2's amp and check to see if there is any difference.

[GohanX]

*shrug* I just followed the directions, and they worked. Unsure if the cap did anything, I didn't try it without one.

[famifan]

i beleive you are smart enough and can give me your copper layer resistance approximation that can cause 0.5-0.10 voltage drop. ORLY?

Yeah, no problem, just tell me how big is current. Ohm manipulated with three parameters
By the way, you should check noise on the ground lines.

afair, total famicom current consumption measured by myself some months ago is laying around 0.5A +-100mA (the fluctuation depends of the ROM ICes that's used in cart. Old TTL roms eat a lot of current). I'll made more accurate measurement later if you need.
Post Merge: June 22, 2012, 06:20:01 am

Ummm pin 22 is /SYNC, but running in parallel with pin 21. And they are pretty close inside PPU case.

PPU's pin 22 is connected to VCC ( /SYNC: When low, forces internal state to reset/blank).

[li Arc]

Hello all, on a recent trip to Japan, I found a shop selling used Famicoms that were tested and cleaned, so I bought one to bring home.  Having had a Famicom since the early 90's, I found my old FDS and plugged it in.  Needless to say, it started up fine, but the video signal was horrible on a modern plasma.  I went searching and found the AV mod, which I performed (used a 1/2W 300 ohm resistor, a 1/4W 100 ohm resistor, a 2N3906, and a couple 16V 220uF coaxial caps).  When I plugged it in, I would get a clear video signal, only it lasted a few milliseconds, and flashed periodically (showing the updated picture) every 1s or so.

At first I thought it might have been the parts I used, especially the PNP, so I tried bypassing the circuitry and hooked the TV up directly to the video pin, and though much attenuated compared to using the transistor, the video signal did the exact same thing.  Not having removed Q1, I plugged the RF back in, and the picture was as it always has been, no flashing frame skipping.  I plugged it into my receiver, thinking maybe it's the TV, but the receiver won't even acknowledge the signal and showed nothing.  Not only this, but no sound was being generated (I couldn't tell if RF generated sound because my plasma doesn't have speakers).  I assume if it's generating a good signal for 20ms every 1000ms, it's probably not enough for the receiver to recognize it (unless my odd Dell TV really doesn't like it somehow).  I double checked the voltage at the 7805, which is steady (not periodic) at 4.92V, and tried moving the coax RCA cable ground to the same ground as the 7805 is using (in case the ground plane was being affected by something periodic), but no change.  My next step is to take it as is and try it on another TV, but I'm hoping that's not it.

Ideas? 

[famifan]

All hail to GohanX  

Jailbars are gone after PPU has been shielded with a piece of copper foil!

[famifan]

2 famifan well done, just no need shielding behind the crystal in the chip. By the way, did you shielded bottom of PPU as well?

PPU had been wrapped by coil, so the bottom was shielded too. As you can see i've put the hit-resistant insulating insertion between the coil and the PCB

[GohanX]

Famifan: Well, that's not exactly what I did, but it works! I built my amp on a tiny piece of breadboard, wrapped it in foil and tucked it in the side of the case where the controller cords run up the side of the system.  

What do you have going on there where the RF board normally goes? Did you yank it and build a standalone power board? I had been thinking of doing that so that I would have room to mount proper AV jacks to the back of the system.

[famifan]

Famifan: Well, that's not exactly what I did, but it works! I built my amp on a tiny piece of breadboard, wrapped it in foil and tucked it in the side of the case where the controller cords run up the side of the system.  

What do you have going on there where the RF board normally goes? Did you yank it and build a standalone power board? I had been thinking of doing that so that I would have room to mount proper AV jacks to the back of the system.

yeah, needless RF circuit had been completely removed.

i've just made some improvements such as:
1) adding good heatsink for 7805 (afair yanked it from old 66/100MHz cpu). It is also the bigget heatsink i've found that fits into tiny famicom case.
2) adding some electrolytic capacitors for 7805
3) connecting the turn on switch to the 7805 input

av jacks.... i have 1.5m long hardwired cable with jacks that goes out of famicom.

[GohanX]

I may duplicate that sometime. I've actually got some 7805 heat sinks laying around I pulled from dead Genesis model 2's.

[li Arc]

Any chance to try with another TV?

Yep, my guess was right...tried another TV and it worked.  Not sure why the comp output won't work on my both my TV and my amp, but it looks like I'll have to live with RF for a while longer, until I get another TV.  Sound seems fine as well, and was able to push that through the amp.

li Arc

[GohanX]

I tinkered with my Famicom again today, and removed the power board. I relocated the video amp to the back of the Fami and installed a proper video jack in the hole where the RF jack used to be. I was running short on time so I didn't cut out any holes for the audio, and I just ran the RCA cable out one of the TV switch holes. I also put on some foil to cover the PPU instead of the video amp, although I didn't quite go as hardcore as 80's freak and put it on the underside.

The result: Perfect video out of the Fami. No jailbars, and none of the slight interference I was getting from using a crappy video cable. The video is on par with my Toaster and modded NES 2 (both use the same video amp) but the brightness is a shade darker, which actually makes the colors pop a bit more. For composite video, I'm quite impressed.

Maybe next weekend I'll properly install the audio jacks, I was thinking of filing away the channel switch holes and putting them there, although the three holes (dual mono audio) would have weird spacing. Still, I think that's the neatest way of doing it.

Also, when I removed the power jack, I chipped the plastic a bit. It doesn't interfere with operation, but it's an eyesore. I don't suppose anyone would have a spare power jack from a parts system they'd be willing to sell? I know I can get them at Radio Shack, but I fear they probably wouldn't fill the hole in the case properly.