LM1881 in XRGB-3 downgrades signal with CPS-2 (and others)

[Artemio]

A while ago I decided to instal an LM1881 inside my XRGB-3 since I believed that the LM1881 could not alter the sync in a negative way. A few weeks later I built a superGUN and got an XRGB-2. When I compared what I was getting through the XRGB-2 vs teh 3 with CPS2 games I was shocked.

Here's a video of what I was getting through the XRGB-3 when using CPS-2. This is using the 46" Samsung series 950 LCD, please notice that only CPS-2 games were teh ones where it was most notorious (please watch these in "HQ"):

http://www.youtube.com/watch?v=zYXkhi5yEpo

Now here's what I think is happening. Of course it won't affect the signal when the sync is 60hz, but when fed a different sync such as CPS-2 the values of the capacitor an resistor used at the gamesx diagram might not appropiate for CPS-2 since it was designed with 60hz in mind. The LM1881 can handle other frequencies, but the circuit needs to be designed for that or have variable values. I am corerct in this assumptions?

As you can see in the following video these are gone after installing a switch that lets the original sync or the corrected sync by the LM1881 be selected.

Here's the final video switching between the corrected sync and the original CPS-2 sync:

http://www.youtube.com/watch?v=1OxmrN6BH0k

I had it installed since most of my RGB cables didn't have the +5V signal, and some of my consoles (The genesis/mega drive and my AES with Pulstar) needed it, otherwise teh sync was lost here and there when teh screen was bright or between sudden scene switches.

[viletim]

The LM1881 is a sync seperator. Its sole purpose in life is to take a video signal, find the sync information inside, and output only the sync and other timing as digital signals. What do you hope to acheive by sending a digital sync signal into its input? The best you can hope for is the same signal on the output. The worst is a damaged IC because feeding a digital sync signal can easily violate the Input Voltage rating in the Absolute Maximum Ratings of the datasheet.

While on the subject, the traditional LM1881 circuit should include a 75 ohm (or thereabouts, the value isn't critical) resistor between the input and ground, before the coupling capacitor. This sets the optimum video level and combined with the lower line impedance and reduced noise, give the LM1881 a much better chance of seperating the sync from a video signal which is changing from bright to dark rapidly.

[RGB32E]

Waviness non-sense!

Anyways, cool videos!  So, which XRGB (2/3) do you prefer for CPS2 games on your A950?

[Artemio]

viletim, I am feeding it the analog sync. What I am doing is just placing the LM1881 inside the XRGB for convinience purposes. It being inside or outside would bear the same effects here, since it is placed between the RGB-21 port and teh signal path, before the XRGB does anything to the signal. It is connected in exactly the manner described here: http://gamesx.com/wiki/doku.php?id=av:lm1881 and it works perfectly fine to fix the Sega Genesis sync problems I am having.

What I intended was to let other people know that in some cases it degrades the signal as shown in the videos.

RGB32E:  Right now I prefer the 3, because it let's me set lighter scan lines. Otherwise I would say the perform in a very equivalent way.

[viletim]

Snatcher,

Have a look at these two waveforms.

1) Composite video (terminated)
2) Composite sync TTL

(1) is the type of signal TVs like to see, it has video and sync information together in the same signal. Sync information is 300mV in amplitude.

(2) is the type of signal computer monitors and arcade monitors generally use for sync. Signal is 5v in amplitude.

The LM1881 circuit useful to generate a (2) signal from a (1). Connecting a (2) signal to the input of an LM1881 causes problems and should be avoided.

[RGB32E]

The LM1881 circuit useful to generate a (2) signal from a (1). Connecting a (2) signal to the input of an LM1881 causes problems and should be avoided.

Shenanigans!  While I would agree that a LM1881 should only be used when needed, there are some applications that it works just fine for (Genesis 2 CSYNC cleanup).  Check out the attached image.  I'm using that circuit for my genesis 2 RGB output.  If I input composite video (instead of sync), I get the same nasty vertical lines on the RGB output as composite video!  So, I've found that there are some appropriate applications that contradict your statement.   

[NFG]

Doubtless there are situations where the LM1881 is useful when fed a sync signal, but as the OP says, it's not working for him.

Your example of a Genesis benefitting is hardly discounting viletim's advice, the genesis is well known for having shitty RGBS output. 

[RGB32E]

Doubtless there are situations where the LM1881 is useful when fed a sync signal, but as the OP says, it's not working for him.

Your example of a Genesis benefitting is hardly discounting viletim's advice, the genesis is well known for having shitty RGBS output. 

No, not a discount...  I described an exception to the information in viletim's post.  My experience with the Genesis 2's RGB output is actually really nice, just that the CSYNC is crap with certain applications.  When using the Genesis 2 on my PVM, all that has to be done is add a 220uf cap, and a perfect picture can be had (plus RCs on RGB of course).  However, things changed when using my Genesis 2 on my FC-14, so the Redmond Cable buffer circuit corrected the bad CSYNC since a series cap wasn't enough.  I also found that feeding CSYNC into the LM1881 was a better choice than composite video, as the same type of vertical lines present on composite video were present on the RGB signal.

[Tiido Priimägi]

on another note - MD sync singal issues are caused by the fact that the sync signal is an open drain output with a relatively high value pull-up resistor. Decrease the value and all your problems suddenly disappear

[Drewman21]

Hey RGB32E,
what is that RGBS board from or for? AND where did you get it? Thanks!

[RGB32E]

Hey RGB32E,
what is that RGBS board from or for? AND where did you get it? Thanks!

The board is used for sync bufferring/stripping and filtering of RGB.  It has a LM1881 and CD4049 circuit (sync stripper + hex buffer), and 3 series 220uF for removing DC offset on RGB signals.   It also contains traces for stereo audio and shield ground, as this PCB comes in a tube that is intended to run inline with the cable.

I'm using one now for my Genesis 2 RGB lead.  The sync circuit cleans up the crap sync output from the Genesis, and the caps filter the RGB better than conventional caps, and 75 ohm resistors were connected to each respective RGB output to get proper levels.  (Note: feeding Genesis composite video resulted in the dreaded vertical line pattern that is visible when using composite video...?)

It was initially created for making RGB cables for the original playstation.  I got these from redmond cable (in Redmond, WA).  The company stopped making RGB cables about 5 years ago and has since thrown out all of their un terminated Playstation RGB cables...

Most of the cables I picked up are just un-terminated RGB cables (effectively scart cables with mini coax conductors for all signals plus outer braid sheild and drain).  However, some came with sync buffers as pictured above.

[Artemio]

The LM1881 circuit useful to generate a (2) signal from a (1). Connecting a (2) signal to the input of an LM1881 causes problems and should be avoided.

Thanks, learned that the hard way =)

It only helps my genesis RGB out so far. i posted this because I believe it may be a good idea to put this info on the LM1881 page for future noobs like me.

[Artemio]

Changed the wiki page from:

* You can leave the chip in circuits with both c-sync and c-video, it won't affect c-sync signals.

to

The LM1881 circuit is useful to generate a composite sync signal from a composite video signal. Connecting a c-sync signal to the input of an LM1881 causes problems and should be avoided.

http://gamesx.com/wiki/doku.php?id=av:lm1881

[NFG]

I wonder if we should mention that it can/may cause trouble?  I've used it with sync signals without issue before, tho obviously it's not always the case.

[Artemio]

You are right Lawrence, I changed that to *may*.