NESRGB Jumper 8 - TTL vs 75ohm CSYNC w/ OSSC

[jsteel]

Hi,

TLDR; NESRGB + OSSC - Jumper 8 open or closed?

I bought a Famicom AV with an NESRGB board in it. I've been using it perfectly fine with my OSSC but I've recently been hearing that TTL CSYNC could put stress on (or could damage) devices. My jumper 8 is open so I'm on TTL but from what I can find out, it's not really TTL; it is "close enough" to 75ohm. I just wanted to get some advice if I should look to solder jumper 8 or not; does the OSSC care? I'm not really an expert with a soldering iron and I'd hate to put a blob of solder on there, find it causes an issue with my OSSC (as it has so far been working perfectly well) and then have to try and de-solder the blob and potentially make a mess of it.

My SCART cable doesn't have anything on the CSYNC line. If the OSSC wants 75ohm, would another option be do add something to the cable? I'd feel more comfortable modifying the SCART cable vs the (expensive) NESRGB mod.

Thanks

[nmalinoski]

Since you're going into AV1 with SCART, sync needs to be 1V 75Ohm; 5V TTL is way too much; only AV3 is able to take TTL.

As for the configuration of the RGB mod, I think the most correct approach is TTL for clean composite sync (What the SNES uses), which means a 470Ohm (or whatever it is) resistor on the sync line in the cable if the cable is wired to pull composite sync from pin 3 on the AV multi-out (Which makes it incompatible with PAL, because pin 3 there is 12V).

If, however, your cable is wired for sync-on-composite, and the AV multi-out wasn't rewired to put TTL composite sync on the composite video line, then no components are necessary.

[Kez]

Refer to this post from Tim on the subject, the "TTL" output is nowhere near 5V and I am sure will be fine with OSSC. If everything is working now, you're probably fine to leave it as is.

That said, for the most correct output.. closing J8 would be the way forward and is a very simple thing to do.. but it is up to you to decide whether you are comfortable with it or not.

I use a NESRGB with my OSSC and J8 closed with no issues. But of course YMMV.

[jsteel]

OK that's good to know someone else uses an OSSC and has J8 closed. I think I'll try that then to be compliant. Worst case I'll have to find some time to carefully remove the solder if it causes any issues. Thanks for the help.

[ldeveraux]

I just dejitter modded my NES and the instructions said to open JP8. With that in consideration, is there any reason for or against?

[jsteel]

Do you mean these instructions? I don't see any reference to Jumper 8 here. I have a version 2 board with dejitter built-in.

https://etim.net.au/nesrgb/background_f ... itter.html

[ldeveraux]

Do you mean these instructions? I don't see any reference to Jumper 8 here. I have a version 2 board with dejitter built-in.

https://etim.net.au/nesrgb/background_f ... itter.html

No I followed the RetroMods tutorial: https://retromods.game.blog/nes-famicom ... d-install/

I installed my RGB mod before the dejitter was combined and I didn't want to lose my palette selector.

[jsteel]

Sorry I'm not familiar with that mod but maybe someone else can chime in.

After further research I've had a second thought and wonder if instead of shorting J8, if I should leave it open and add a 330ohm resistor on the cable's csync line, so I could use it with an NTSC SNES/SFC as well, as I believe that outputs TTL only. But if the NESRGB outputs a different level of TTL compared to a SNES/SFC is this going to cause an issue?

[leonk]

Hi,
TLDR; NESRGB + OSSC - Jumper 8 open or closed?

TLDR; The correct answer - it depends.

If your NESRGB install uses Tim's 8 pin miniDIN then your NESRGB should be configured for 75ohm. No components in cable.

If your NESRGB install uses a Nintendo MultiAV port (taken from another broken Nintendo console or 3D printed) than your NESRGB should be configured for TTL and a 470 ohm resistor should exist in the SCART cable. The technical reason for this is because all Nintendo made consoles that have RGB out and use this port are TTL CSYNC. So if you want a single cable to use on multiple Nintendo consoles, you need to do this.

[jsteel]

If your NESRGB install uses Tim's 8 pin miniDIN then your NESRGB should be configured for 75ohm. No components in cable.

If your NESRGB install uses a Nintendo MultiAV port (taken from another broken Nintendo console or 3D printed) than your NESRGB should be configured for TTL and a 470 ohm resistor should exist in the SCART cable. The technical reason for this is because all Nintendo made consoles that have RGB out and use this port are TTL CSYNC. So if you want a single cable to use on multiple Nintendo consoles, you need to do this.

Thanks for the reply. Mine is the FCAV so has the MultiAV. Yes something between 330ohm and 470ohm seems to be the recommendation for the SNES/SFC but I have seen reports that the NESRGB doesn't output the same level TTL as the SNES/SFC so are you sure the same resistor value applies to the NESRGB? I guess I could try it but would like to be certain that I'm not going to stress my OSSC by feeding it a too low level of CSYNC, if that's even an issue. Thanks

[leonk]

For the FCAV I would cut the V trace to the multiout (pin 9) and then run R G B CS Y C V wires from nesrgb to multiout. Use TTL and use standard Snes scart cable with resistor in-line with csync and you’re done.

[jsteel]

Is your comment in reply to my concern about different TTL levels between the consoles? If so, I haven't understood.

[leonk]

We are talking about csync here. A resistor between 300 and 470 ohm will work just fine because the IC in the OSSC that handles sync will work just fine in that range (we are talking analog to digital conversion here with a wide range of acceptable input). What’s not good is if you send TTL into it without resistor because then you’re sending a signal that’s too hot and eventually result in IC failing.

[jsteel]

Tim writes here that the NESRGB was designed for straight-wire cables, and "The CS# signal on the NESRGB is not really a TTL output. It's a TTL signal with a series resistor. This way it works reasonably well with both TTL and 75 ohm input when directly connected."

viewtopic.php?f=6&t=47617&start=3930

This is where my confusion is coming from comparing it to the SNES TTL; it seems they are not the same so can a cable really be used for both? I'm thinking it would be best if I close J8 so instead of "not really TTL" it is 75ohm (no confusion) and then stick a 330 to 470ohm resistor on a different cable for my SNES, just to be sure everything is at spec.

[makar1]

My AV FC with the NESRGB works fine at 4x using either a SNES NTSC C-SYNC or Sync-on-Luma cable from Retrogamingcables UK. You have to wire the Luma from the NESRGB board to use it as sync though.

[jsteel]

I don't see an NTSC sync-on-luma cable from them but sure, luma is an option. I believe luma is a safe level on all consoles (no resistor needed); I use luma as sync on my PAL SNES and PS2. Their SNES CSYNC cable has a resistor on the sync line, and from what I've gathered here (Tim's post) you don't need one. So maybe it works fine but probably it's not at spec for the NESRGB; at least that's the assumption I've made here.

I've now closed jumper 8 and everything seems normal. While I wasn't looking for a change I just want to follow best practice and avoid issues down the line. I think this is my best option and I'll grab a different cable for my SFC.

[leonk]

I figured I’d quickly connect the NESRGB to the oscilloscope so that there won’t be any room for miss interpretation.

What you see here CS# output at TTL level captured from NESRGB v2 while connected to a PVM (typically voltage will be even higher under no load)



Sorry for the crap iPhone quick and dirty pic of the oscilloscope. But you can see the 15.7kHz frequency (about 240p NTSC equivalent) but what’s more important is the peak to peak voltage. It’s almost 2V! (Under no load it goes up to 3.3V)

Yes. It’s not 5Vpp (true definition of TTL) but it’s not 75 ohm either (300mVpp or 0.3V)

What this all means:

- if using PVM. Who cares. It can sync to a shoe string and has circuit to normalize sync (reduce to 300mV from TTL or amplify to 300mV if you got a weak signal)

- modern scalars are more sensitive. There are 2 consequences from being out of spec:
1) Under 300mVpp (eg use scart cable with resistor and 75 ohm setting) varies from working ok to occasional drop video to no video at all (especially true on xrgb)

2) over 300mVpp (by a lot; TTL setting with no resistor) you are putting a lot of stress on the scalar sync circuit by running it hot / out of spec. Hot air chip replacement is in your future (see arcade PCBs with 5V TTL csync blowing up OSSC and GSCARTSW)

[jsteel]

Thanks for that, and this is with J8 open on the V2? So with it closed it is true 75ohm and no issue? If so that's the option I've taken. Knowing the TTL is off-spec makes me still think that the SNES (assuming it is on-spec?) can't use the same cable as surely the resistors would need to be different values for each console, unless the TTL level is actually the same on both?

[leonk]

You have to remember that csync is not the same as the R G B video lines

If voltage on RGB is 50% than the brightness would be 50%. If voltage on csync is 50% well. You won’t see anything. Csync is just a pulse that resets the positing of the electron guns. As long as the video device understands “reset” it works. That’s why using a 300 or 470 ohm resistor doesn’t really matter.

I’ve installed a few hundred of these NESRGB consoles and I can assure you that using a snes scart cable with resistor in it and setting it to TTL is safe

Some People choose to use luma sync because they don’t fully understand csync; Luma is basically csync 75 ohm with extra brightness data the video device ignores.

[maxtherabbit]

PVM have high-Z sync termination - you are not getting an accurate representation of what that waveform will be with a 75-ohm termination in a TV - it will be much less

[leonk]

The waveform on the oscilloscope is taken right at the CS# pad of the nesrgb. Not at the PVM.

[Ryoandr]

PVM have high-Z sync termination - you are not getting an accurate representation of what that waveform will be with a 75-ohm termination in a TV - it will be much less

PVMs have either 75R termination or auto 75R termination if their loopback ports are empty.
Its BVM s that require 75R terminators on loopback ports

[makar1]

Some People choose to use luma sync because they don’t fully understand csync; Luma is basically csync 75 ohm with extra brightness data the video device ignores.

Or they own devices that output S-Video natively but not C-Sync. E.g. SNES 1-CHIP 03, and the N64.

[jsteel]

You have to remember that csync is not the same as the R G B video lines

Yes I understand that, I'm purely wanting to make sure my sync level is perfect for my OSSC.

I'm not saying you are wrong but I am reading conflicting advice about this matter which is why I've opted for jumper 8 closed and a straight wire in the cable as I don't believe there's any confusion there.

Here's an interesting post (and discussion that follows).
viewtopic.php?f=6&t=47617&start=4500

Not according to what Tim said. Both the older and newer revisions have resistors attenuating the csync signal. So no, the NESRGB board does not output TTL csync in any normal circumstance. In his diagram he posted, revisions 12 and 13 have a 680 Ohm resistor on the csync line, while the current revision 14 has a 470 Ohm resistor plus a jumper to an additional 91 Ohm resistor. So even if you leave the jumper open, you've still got a 470 Ohm resistor on the csync line.

What all this means is you're in no danger of TTL csync levels coming out of the NESRGB board in any scenario.

[maxtherabbit]

The waveform on the oscilloscope is taken right at the CS# pad of the nesrgb. Not at the PVM.

that doesn't matter since the NESRGB has the series resistor in line before the pad

i.e. when you terminate the other end of it with 75-ohm it will drop at the pad as well

[leonk]

I'm not saying you are wrong but I am reading conflicting advice about this matter which is why I've opted for jumper 8 closed and a straight wire in the cable as I don't believe there's any confusion there.

and what you're doing is perfectly fine. Just don't use that SCART cable by accident on your RGB modded SNES or N64. It will damage your OSSC.

Many people prefer to have 1 SCART cable and multiple Nintendo consoles (again, NES, SNES, N64) and for these cases, the output from the console needs to be electrically compatible. Nintendo set the tone by having TTL output from SNES and early N64 models. So setting pin 3 of the Nintendo MultiAV to "TTL" level (keep J8 open) is the safest solution for single cable users on NESRGB.

[leonk]

that doesn't matter since the NESRGB has the series resistor in line before the pad

i.e. when you terminate the other end of it with 75-ohm it will drop at the pad as well

Not with PVM. PVM is already 75-ohm self terminating. Please re-read my post above. Going into PVM, CSync is 2Vpp. Not plugged in, it's >3Vpp. The drop you're talking about already occurred. Test done with system connected to PVM, playing game on screen.

[nmalinoski]

Nintendo set the tone by having TTL output from SNES and early N64 models. So setting pin 3 of the Nintendo MultiAV to "TTL" level (keep J8 open) is the safest solution for single cable users on NESRGB.

As far as the AV multi-out, TTL on pin 3 was the trend for NTSC consoles, but it's 12V on PAL consoles. :/

Side question: Where did Nintendo ever use pin 3 for sync? Did the JP-21 cables use it, or did they use composite sync like the SCART cables in PAL regions? Were there special, clean-composite-sync cables used with any of the development kits or store display units?

[jsteel]

and what you're doing is perfectly fine. Just don't use that SCART cable by accident on your RGB modded SNES or N64. It will damage your OSSC.

Yes this seems to be the only consistent information I can find. I'll stick to doing this. Cables are well-labelled

Many people prefer to have 1 SCART cable and multiple Nintendo consoles (again, NES, SNES, N64) and for these cases, the output from the console needs to be electrically compatible. Nintendo set the tone by having TTL output from SNES and early N64 models. So setting pin 3 of the Nintendo MultiAV to "TTL" level (keep J8 open) is the safest solution for single cable users on NESRGB.

One cable to rule them all could be handy but from the sounds of it, the NESRGB outputs a lower TTL level than the SNES so I don't understand how the same cable is best suited to both. It must work as you say, but isn't the NESRGB signal weaker than intended with a cable at spec for SNES TTL? There must be some tolerance but personally I like the idea of correct levels with different cables. Bare in mind I'm just a lowly, paranoid consumer so I'm not trying to give advice here; I'm just trying to help myself sleep at night.

[leonk]

the NESRGB outputs a lower TTL level than the SNES so I don't understand how the same cable is best suited to both. It must work as you say, but isn't the NESRGB signal weaker than intended with a cable at spec for SNES TTL?

That's because we are talking about CSYNC! From the OSSC perspective 300mVpp or 150mVpp CSYNC is still very usable signal. CSNYC is just co-ordinates. Nothing more.

If we were talking about the actual RGB data, then that's something totally different. 700mVpp vs 350mVpp would be a much darker image.