D.I.Y. JAMMA adapter - cool idea and a question

[Akuma]

Hello everyone. I just got the idea to look at a certain tech-geek site called gamesx.com which I have been browsing over a year ago. Now, looking at everything at a new, home arcade cabinet usage point of view, I found something I didn't read last time which seems to me like an awsome idea which is almost screaming to get used in my cab.

Check out this link: http://www.gamesx.com/arcade/cheapjamma.htm

A self made Jamma adapter. It's vice versa from the PC2Jamma thing. Now there is only something I wonder about: Doesnt't a Jamma board put out the video in 5V RGB? Not to mention the 15 kHZ. So, can you safely plug this to a 15 kHz sync-able monitor? Or do you need to solder any resistors or something in, to get it down to 75 ohm? Might be a stupid question, but I am only an amateur electronic engineer ; )

I know, this might be better suited for the monitor forum, but I thought the adapter might be suitable for many fellow forum members that frequent the main forum more than the video forum.

[Stobe]

That is a very simplistic design.  You may want to add some pots to the RGB lines to give you some manual control on the levels.   You did take notice of his intent to use a RGB to VGA converter, which in some cases does have adjustments for sync and RGB.

-Stobe

[Akuma]

ah, I've seen now - so this would regulate the 5v 15kHz to low volt 31 khz?

[SavannahLion]

ah, I've seen now - so this would regulate the 5v 15kHz to low volt 31 khz?

No. It would drop the 5v to 1.5v (IIRC), but it would do nothing to change the frequency. To do that, you would need to find yourself a scan doubler or something similar. Given the cost of an RGB 15KHz to VGA 31KHz converter, it would negate the whole concept of "cheap".

[Akuma]

I see, I see... Since I live in europe a TV is also an alternative because of the RGB-scart connector's compatibility. I just have not figured out yet, how a MAK / Supergun modifies the RGB signal from 5v to 1.5 volt...

[SavannahLion]

I see, I see... Since I live in europe a TV is also an alternative because of the RGB-scart connector's compatibility.

I hate you. 

I just have not figured out yet, how a MAK / Supergun modifies the RGB signal from 5v to 1.5 volt...

If it works, it works, why worry about it?

[MonMotha]

Scaling a signal smaller is really, really easy.  If the monitor has known input termination (with most TVs and computer monitors, this is standardized at 75 ohms), you can actually just use a single resistor in each line to form what is colloquially known to EEs as a "voltage divider".  If you want to go from 5V (common to arcades) to 0.7V (PC standard, and also the active video level for NTSC and PAL composite video), you want either a 470 (slightly dim) or a 430 (slightly too bright) in series with each video line.  You can pass composite TTL sync (as output by arcade boards) straight to a PC type monitor, and though most SCART connections expect CVBS, you can sometimes get by with just attenuating it down to 1V with a similar divider.

If you want to make things adjustable, get yourself some 500 ohm pots and put them on the line (one side at the line and the wiper hooked up to the TV side).  Then you can adjust.

Be aware that this only works when you know the monitor's input termination.  Arcade monitors tend to vary a lot from 75 to about 10k ohms.  In this case, you need a buffer amp to isolate the attenuation stage from the output, or you can do things by trial and error or look up the input termination for your monitor.

Now, making signals bigger requires a little more effort.  You need a video amp.  For most people, it's probably far easier (and possibly cheaper) to just buy one, unless you have some high bandwidth op-amps laying around.

[SavannahLion]

If you want to go from 5V (common to arcades) to 0.7V (PC standard, and also the active video level for NTSC and PAL composite video),

Hhmm... why was I thinking 1.5V for PC standard?

[Akuma]

I just have not figured out yet, how a MAK / Supergun modifies the RGB signal from 5v to 1.5 volt...

If it works, it works, why worry about it?

Well - actually I wanted to refer to this article: http://www.engadget.com/2006/11/28/how-to-consolize-an-arcade-game/

The author says that it is easy to hook it up to a european TV but only tells in detail how to build a RGB2NTSC converter for american TVs. Furthermore, all the garage-manufactured MAKs and Superguns that are available here plug to standard TVs using RGB-Scart. I don't own one and really don't want to buy one since I love the idea of building it myself. Today I succeeded in creating an Naomi / JVS adapter for an ATX power supply, after I get the Naomi wired completely I will add some connectors to make the ATX psu compatible to jamma boards too.

I am taking ALOT of pictures and plan to write up a guide about this, since I realised that the web lacks some good (in other words detailed) home arcade guides - especially the Naomi scene! Therefore I am grateful for everyone who tells me, being an amateur, about ohms and voltages and RGB and so on. The more I spend my time on this matter, the more I learn : ) Hey kids, stay in school - learning rocks! ; )

[MonMotha]

The signal that comes out of a computer into a high-impedance load is 1.5Vpp (actually 1.4Vpp, but close enough).  There is a 75 ohm back termination on it which forms a divider with the 75 ohm termination in the monitor resulting in the monitor seeing 0.7Vpp when everything is properly terminated.  This is done because the coax cable used has characteristic impedance (Zo) 75 ohms, and terminating everything in this manner prevents reflections and other transmission line end-line artifacts, which can become quite prevalent on even moderate length lines at the high resolutions PCs sometimes run (like 1600x1200).  This termination scheme is not often employed in arcades.

If you hook your PC video card straight up to a scope and display a black/white checkerboard, you'll see 1.5Vpp signal levels, but if you hook it up to a monitor and probe the signal lines then, you'll only see 0.7Vpp signal levels.  If you drive a PC monitor (or other monitor with input termination) from a low-impedance source such as directly off the output of an op-amp, there's no back termination to divide with, and the monitor will see exactly what you are driving it with.

The choice of 0.7V for active white level has to do with some things you have to do when you AC couple the video and want to restore the DC level, but is otherwise not of much concern in this application.  Note that if you have embedded sync on a PC video signal (sync on green), the total signal will be 1Vpp, with +0.7V as the full intensity level, 0V as black level, and -0.3V as sync.  NTSC-M (US) plays some tricks and defines the black level to be slightly above blanking which is again above sync.  NTSC-J (Japan) does away with this.  You can adjust the difference out by tweaking brightness.  IIRC, no PAL standard incorporates this "feature".

On monitors where you have control of brightness and contrast, you can adjust some variance back out.  Contrast is actually the "white level" adjustment (i.e. you are adjusting what voltage is considered to be "fully white"), and brightness is actually the "black level" adjustment (i.e. you are adjusting what voltage is considered to be fully black).  PC monitors often seem adjustable to handle between about 0.5Vpp and 1.25Vpp signals.  Arcade monitors are often adjustable much, much further, and TVs sometimes feature automatic gain, especially on their composite input where there are some reference voltages to lock the auto-gain to.

The signal levels of composite NTSC and PAL video are somewhat undefined, also, but people seem to have pretty well standardized on 1Vpp (since it includes sync).  It makes the math nice, since full scale is then 1, and everything is otherwise defined as a fraction of full scale.  This has to be done since, when the analog signal is recovered off the air, the TV just scales it (using an auto gain mechanism) to something usable internally, as the actual signal level received on the antenna can vary a lot.

Also, as a warning if you plan to build that RGB to NTSC converter, while it should work great [I've built them on custom boards, and while the composite output is only so-so, the s-video looks nearly as good as what you'll be feeding it], be aware that those "crystal in a can" oscillators are often not tight enough for some TVs' chroma PLLs to lock on.  I had to build my own oscillator using a parallel resonant crystal and some inverters and tune it up with a variable cap.  Also, if you're in Europe, you can set the chip to PAL mode and get something that has PAL style color modulation, but it will still have whatever timings the arcade board uses, which is normally closer to NTSC.  Many European TVs will apparently sync up just fine, though.

[brandon]

I thought that a Scart TV was the same as an arcade monitor and thus could simple be connected to a JAMMA board without much fuss.. so the voltages aren't the same?  In that case maybe it'd be cheaper just to convert a TV into an arcade monitor with one of those 8liner chassis.  

[Akuma]

I read at gamesx.com that an easier alternative is to find an old Amiga monitor.

[brandon]

I read at gamesx.com that an easier alternative is to find an old Amiga monitor.

This is true.. I used a Commodore 64/128 monitor for some time.  Just be sure its an RGB monitor.  some of the c64 monitors were just NTSC video.

[Fozzy The Bear]

I thought that a Scart TV was the same as an arcade monitor and thus could simple be connected to a JAMMA board without much fuss.. so the voltages aren't the same?  In that case maybe it'd be cheaper just to convert a TV into an arcade monitor with one of those 8liner chassis.  

They are more or less the same if you're using the RGB input on the SCART.... You just have to use a resistor on each of the RGB lines to get the right level or it's too dark. I've got a SCART lead with the resistors inside the plug itself, which works fine.

Best Regards,
Julian (Fozzy The Bear)

[MonMotha]

Now forgive me here, as I'm not a SCART person (live in the USA, where SCART connectors taking RGB are unheard of), but I thought SCART RGB had an implicit DC level, meaning it needed external AC coupling, so there have to be capacitors, not resistors, in the line.  Putting resistors in the line would make it dimmer (due to the input termination).  I know that's what I've found inside all the PS2 SCART cables I've hacked up.

[Fozzy The Bear]

Now forgive me here, as I'm not a SCART person (live in the USA, where SCART connectors taking RGB are unheard of), but I thought SCART RGB had an implicit DC level, meaning it needed external AC coupling, so there have to be capacitors, not resistors, in the line.  Putting resistors in the line would make it dimmer (due to the input termination).  I know that's what I've found inside all the PS2 SCART cables I've hacked up.

You may have found caps in there.... but to pull down arcade 5V to SCART 1V input signal you need resistors. The lead you are looking at is for a PS2, which already outputs its signal at the correct level for a TV Set to handle and therefore has no resistors in it. 

I'm not quite sure why you think it needs external AC coupling though. Unless you mean that it needs external AC De-Coupling. Which is actually unnecessary at the input to an RGB SCART because the signal is de-coupled internally by the TV set before being processed. It is true RGB, not composite although SCART can also accept a composite sync signal, and S-Video. Depending on the facilities available on the TV itself. Unlike in the USA, 99% of them do accept true RGB though.

Best Regards,
Julian (Fozzy The Bear)

[MonMotha]

Ok, I misread you.  I thought you were implying that the signal would be to low (i.e. dim) without these parts in series on the line, not that it would be too high (i.e. bright).  This can happen due to odd DC bias showing up, and some monitors don't AC couple their inputs (didn't know if SCART called for this or not), resulting in whatever random DC bias exists from the source being passed right through.  Ah the joys of lacking a negative rail...

I know that the PS2 SCART cables I have taken apart for other uses do have 220uF caps in series with the video lines, but this may be due to a PS2 oddity (though I seem to recall reading that SCART has this all the time).  Again, we don't have that connection in the USA, so I don't get much experience with it.

Yes, if you want to take common arcade signals (3.3V to 5V) down to 1V levels expected by most consumer gear, you will need resistors in the line (see an earlier comment).  This is cheap (3 resistors, or about 10 cents), and pretty easy to do.

--
Totally offtopic:

Now, when I say AC coupling, I mean effectively building a high-pass filter using a cap and the input termination.  220uF is the common value I see used because it'll pass the 50-60Hz vertical sync when used with a 75ohm input.  Perhaps there's another american/europeanism going on here and you guys call that AC de-coupling (de-coupling in the US seems always applied to caps which are designed to overcome inductance/resistance on power lines/planes so that digital parts can draw their huge current spikes).  I've never seen inserting a capacitor in a video line to pass AC but not DC called "de-coupling", but I could see how such terminology might evolve (though I would personally call that "DC de-coupling" as it "decouples", i.e. blocks, the DC and "couples", i.e. passes, the AC).

[Fozzy The Bear]

Now, when I say AC coupling, I mean effectively building a high-pass filter using a cap and the input termination.  220uF is the common value I see used because it'll pass the 50-60Hz vertical sync when used with a 75ohm input.  Perhaps there's another american/europeanism going on here and you guys call that AC de-coupling (de-coupling in the US seems always applied to caps which are designed to overcome inductance/resistance on power lines/planes so that digital parts can draw their huge current spikes).  I've never seen inserting a capacitor in a video line to pass AC but not DC called "de-coupling", but I could see how such terminology might evolve (though I would personally call that "DC de-coupling" as it "decouples", i.e. blocks, the DC and "couples", i.e. passes, the AC).

OK now I understand what you meant.... I see your point here. It probably is an American/Europeanism going on. Why we can't have a common terminology is beyond me. Especially when most of the gear we use is Japanese on both sides of the pond. LOL 

Best Regards,
Julian (Fozzy The Bear)