Saturn RGB issue

[leonk]

Syncing over composite video is the SCART standard.

This is a USA Sega Saturn. Pin 1 of the 10 pin mini DIN is CSYNC. CSYNC SCART cables is recommended by many cable sellers for Saturn 1. Sync on Luma recommended for Saturn model 2. Well, that's just BS / Hack! Sure, sync on composite video or luma will work, but it doesn't solve the inherit problem.

I believe the Sega Saturn model 2 has a design flaw that causes the EMI filter to fail prematurely.

[viletim]

Syncing over composite video is the SCART standard.

This is a USA Sega Saturn. Pin 1 of the 10 pin mini DIN is CSYNC. CSYNC SCART cables is recommended by many cable sellers for Saturn 1. Sync on Luma recommended for Saturn model 2. Well, that's just BS / Hack! Sure, sync on composite video or luma will work, but it doesn't solve the inherit problem.

SCART, Japanese RGB21, and other television RGB connections all have composite video inputs and the sync must be delivered through this composite video input. It's electrically identical to the round yellow phono socket you're familiar with. Any signal connected to this pin needs to match the composite video standard signal levels or it's not guaranteed to work.

A composite video input is always terminated with a 75 ohm resistor. If you're ever in doubt whether a monitor's sync input is composite video or TTL sync, you need only connect a multimeter between the input pin and ground and measure its resistance. TTL sync inputs usually have their inputs terminated with a 1k to 5k ohm resistor (2.2k is the most common and is the VGA standard).

The signal level for TTL is 2.5 Vpp to 5 Vpp. The signal level for the sync portion of a composite video signal is 0.3 Vpp (sync + luma + chroma = 1.23 Vpp max). If you connect a composite video signal to a composite sync input it doesn't work because the signal is too small and it's confused by the video information (if present). If you connect a TTL sync signal to a composite video input, the result depends on the characteristics of the TTL sync driver circuit.

The Sega Saturn has a push-pull TTL sync output stage (inside the graphics chip) which has a 3.3 Vpp swing and a low output impedance (roughly 40 ohms, I think, though it's non-linear).

If you connect this to a 2.2k TTL sync input load, you have a typical signal current of:
I = V/R = 3.3/(2200 + 40) = 1.47 mA

If you connect this to a 75 ohm composite video input load, you have a typical signal current of:
I = V/R = 3.3/(75 + 40) = 28.7 mA

This is overly simplified, because the sync signal is not DC, but a mostly positive signal with fast negative pulses. My point is no TTL output should be expected to source that much current.

I believe the Sega Saturn model 2 has a design flaw that causes the EMI filter to fail prematurely.

Based on what evidence? Blind faith? The same EMI filter that's on damn near every pin of every connector of every Sega console? You hear about them failing all the time don't you?

It seems much more likely to me that a Saturn's TTL sync output with 75 ohms (at the end of a cable) is stressing the EMI filter in that circuit which causes its premature failure.

[leonk]

Hi Tim, thanks for your input. Just trying to understand more what's going on here. Sorry for the brain dump.

It's very unlikely that the SYNC pin on these 3 saturn consoles ever got used. Not in this part of the world anyways. These consoles for the most part were used using the pack-in composite video cable. Yet for some reason, the EMI filter is shot on the CSYNC output / pin 1 of the 10 pin mini DIN port. 1 console, ok, 2.. maybe. But 3 in 1 month?

The SCART RGB cable I'm using (to connect to my PVM) is straight through. I confirmed it. It has no extra components in the cable.

Maybe the CSYNC pin was never meant to be used. But for some reason is there. According to the schematics I found for for a VA 0.5 (http://www.geocities.jp/atx197/Ssmn052S1_10.pdf) going backwards from the output port:

- CSYNC exits out of CN5 / AV out and can be verified on TP4
- if goes through a 100pF EMI filter (EM4)
- following CN5-1 --> it is used as input into CXA1645M via 2.2K resistor but feeds from VIDEO-SYNC (_H34)
- _H34 originates on pin 86 of 315-5690 / FH3006(YAMAHA) / Video Display Processor2

The above is for Model 1. But the 2 I fixed recently were all VA7 model 2. The only major difference is the EM filter marking, but the rest is pretty much the same between the model 1 and model 2 for the CSYNC circuit.

It seems I'm not the only one that discovered this issue only with the model 2 systems. Take a look at what RetroRGB posted: http://retrorgb.com/saturncompare.html

"I found that the handful of Model 2 units I had produced weird horizontal lines on the screen when using a cable that gets sync from the composite sync pin".

retro_console_accessories has this warning on her CSYNC Sega Saturn SCART cables:

"Some Model 2s have a compromised picture with csync, it isn't all of them or even most of them (per my testing) but it's enough that we have to include this warning. White Japanese Saturns especially suffer from this issue as it seems the majority of them use the problematic circuitry involved."

So why do you suppose that only model 2 are seeing this failure? Why do you suppose no model 1 consoles are seeing this issue? I wish I had a model 1 + 2 so I can fully trace it out.

[darcagn]

Syncing over composite video is the SCART standard.

This is a USA Sega Saturn. Pin 1 of the 10 pin mini DIN is CSYNC. CSYNC SCART cables is recommended by many cable sellers for Saturn 1. Sync on Luma recommended for Saturn model 2. Well, that's just BS / Hack! Sure, sync on composite video or luma will work, but it doesn't solve the inherit problem.

I believe the Sega Saturn model 2 has a design flaw that causes the EMI filter to fail prematurely.

I'm not even talking about the Saturn specifically. You asked why viletim's SCART page shows syncing over composite video. I'm simply answering that it's because that's how SCART is meant to be.

The fact that people make custom SCART cables with composite sync on that pin instead of composite video doesn't negate that the spec is the spec. These people are simply making out-of-spec cables.

Tons of people in the vintage gaming community do stuff they really aren't supposed to do electrically, but continue to do so because it works--and favorably. Myself included. In the specific case of using composite sync, it works favorably because it eliminates chroma interference from messing up your picture. The general thought around here seems to be "if the SCART cable using c-video doesn't work, try one using c-sync; if the SCART cable using c-sync doesn't work, try one using c-video." viletim seems to be the only one pointing out around here that this isn't the best way to go. In retrospect, sync over luma is probably the best way to go, since I believe it adheres to the composite video standard, but without chroma.... but my consoles pretty much all have c-sync cables.

[gojira54]

My setup is; JPN model 2, 3rd party RGB cable wired for euro into a 14" & 20" BVM via kramer matrix, all other consoles/15khz PC work fine on both so no problem with the setup.
Before looking at the scart cable I got audio buzzing but stable picture on both monitors - added 470ohm resistor to pin 20, result no more audio buzzing, picture stable on the 20" but rolls on the 14".
I had a LM1881 I never had to use kicking about - built the circuit and installed into the scart, result stable on both with a small amount of jitter, after a few minutes the picture fades in and out of black but no rolling. I measured the voltage and found it is ~3.3v on pin 8 - too low for the lm1881?
Would like to get a decent picture but not sure if issue is with the saturn or the cable, any help appreciated :s

[mvsfan]

Ive had a Sega saturn model 2 almost since it was new. I bought it off the first owner in the late 90s. I dont have problems with the video either in composite or RGB.

Recently, I picked up a Minty Model 1 Saturn that looks like it was never used. It also has great video.

mabeye the saturns having problems with video is just because of AGE and the fact that some of them have been used way too much over the years and also not taken care of.