Need help fixing my wrongly wired cable with sync stripper

[an1]

Hi all,

Since I got my Saturn I've been using a SCART RGB cable from a seller recommended here. I have an NTSC Saturn, but the cable is using a sync stripper to output csync. Guess that makes it compatible with both PAL & NTSC, never really caused any issues. After some unrelated communication with the seller I suspected my cable might not be correctly wired up. I opened up the sync stripper and SCART head.

Here:

i.imgur.com/yjISp04.jpg

There's no 470 ohm resistor in the SCART head. After removing the hot glue from the stripper board I could tell there's none on the output there either:

i.imgur.com/X75gN7g.jpg

(a bit hard to see, sorry)

That's of course rather concerning as I don't want to fry any devices with TTL level sync. Apparently this has been corrected in more recent versions of the cable.

When removing the hot glue I accidentally disconnected the composite line but re-connected it to the coupling cap on the LM1881 input. I also added a 470ohm resistor in the SCART head on line 20. Unfortunately, I now no longer have working sync and the picture just flickers. I checked everything with my multimeter, composite video from the DIN goes into the little box, arrives at the input of the stripper, goes out through a 480ohm into the right pin on the SCART head.

Maybe I just don't understand the original circuit, for instance there's a 75ohm to ground wired to the composite before it goes into the stripper, but not having the 470 ohm on the output seems wrong and dangerous to me.

Any advice much appreciated!

[buckoa51]

Jeez that's some hot glue bukake going on there.

It's really daft to put a sync stripper in a cable for a console that already outputs clean sync. Looking at that I'd just replace it to be honest.

[an1]

Lord Voultar would not be amused. I don't quite understand the need for the glue since there is a strain relief for the cable and the little box is a sturdy and secure enclosure. Thankfully Voultar's suggestion of weakening the glue with isopropyl works very well. Didn't even need hot air.

I can only imagine the stripper was added to avoid the PAL/NTSC problem and make a universal csync cable. Unfortunately it seems this cable will actually put strain on anything connected due to the missing resistor. Thankfully it seems I caught the issue before any equipment was damaged.

I'll be getting a new cable, but I'm just trying to get a better understanding of all things console electronics, just curious why it works perfectly without the resistor but fails with it. That seems like the reverse of what should happen.

[buckoa51]

Yeah adding the stripper was probably done as you say to make one cable that fits all, but it would have been better to just use luma for sync in that case. As I understand it (and this is what Voultar said I believe) an LM881 generates a certain amount of line delay. And based on how many uS that line-delay is, the screen will be drawn later, causing a horizontal shift in the picture.

Given that clean sync is only needed if you're using Extron switches or an XRGB3, it seems a shame to degrade a cable for the sake of something a lot of people won't need.

The only thing I can think is, maybe the sync stripper circuit is outputting 75 ohm anyway?

[an1]

Yes, there's probably no issue with sync-on-luma. I try to avoid composite video since I had visible artifacts with that type of signal with longer cables or switchers. Never seen a problem with sync-on-luma in either sync stability or coupling in cables/switches etc.

The gSCART switch claims one of its advantages is that they use a custom FPGA based sync processing circuit that does not have the same delay & shift as the LM1881, so that's probably true.

From the datasheet (http://www.ti.com/lit/ds/symlink/lm1881.pdf) it seems like it should output TTL level sync. If I compare the circuit in my cable here with the design of the sync stripper board you're selling on VGP and the one RGC sells, I can see some noteworthy differences. The RGC one has the 470 ohm resistor on the board and the installation instructions for yours strongly encourage to add one. This cable has no resistor. Interestingly, the circuit used in the cable here has a 75 ohm to ground on the composite video line before it enters the cap & LM1881 input. The VGP/RGC ones do not. I'm not sure this is needed, and probably would not make a difference for the output of the stripper?

Also, something I've been wondering about - do you think the OSSC would be capable to warn about dangerously out of spec signals? Over the years most of the recommended cable makers seem to have made mistakes, i.e. the infamous 'boosted sync' cables or missing the resistor on the sync line like mine here. It can be hard for non-experts to check their cables as there are several valid ways to wire them, correct cables can be different for different console revisions / regions, circuits in the cables can be complex like with the stripper here, video signal levels can't be measures with inexpensive tools like a multimeter etc. There's probably countless people slowly damaging their TVs, scalers, switches with improper cables and they don't even know it. Picture looks good, nothing wrong. It would great if the OSSC could warn if it's digitizing a signal that is dangerously off-spec. Do you think that is possible at all and might make a good feature request?

[zyrobs]

The OSSC works perfect with the factory original Saturn SCART cables. In fact it works better with those than with the cable you posted on the top.

[an1]

Those are getting hard to get, so the usual cables from RCA/RGC are what I'd need to buy. Since I can't figure out why adding the 470ohm resistor makes the cable stop working I think I'll just convert it to a sync-on-luma cable. That should be quite easy, universally compatible and require no components.