Electronic multi-switches

Hello,

Small world init' :-) I'm doing my ST right now so have everything fresh in my mind from only a few minutes ago. :-)

With modern televisions and monitors you should be ok since if the sync signal is out of range then the monitor simply shuts down.

In the olden days when almost everything was done in an analogue fashion this clearly was a problem. But almost every signal today is processed digitally. As long as it's not over voltage then it's not a problem.

Possibly the only change to your circuit would be to make the 150R RGB resistors presets. Only reason I say that is that I'm finding here right now that 150R is a bit too much, 100R seems more appropriate.

I'm making up a low/med/high for a Samsung 19" LCD television which has a VGA input (LE19R71BX). Literally all I have is a DPDT switch which connects ST-pin-4 and GND, and in the other direction supplies 12v to the SCART-pin-8 function pin. It's working fine. I just flick the switch from 31.5Khz to

Only thing I have noticed is that with the VSYNC connected to both the SCART and the VGA input at the same time, it does just slightly alter the VGA picture.

Maybe the other thing I've found is that an LCD doesn't quite process the ST's 15.25Khz low/med as well as what an old Philips CM8833 *analogie electronics* CRT does. I get what I'd describe as digital artifacts if you know what I mean.

My suspicion is that these LCD televisions with VGA inputs are nothing more than VGA monitors with a scan doubler bolted on for the television. VGA signals seem pure and precise. Yet the 15.25Khz SCART images seem a little bit processed.

Hope you liked that 16/32 :-) You came into the wild west wilderness of Usenet and found someone doing something equally retro.

All the best,

Aly

Analog video is my world and you're preparing to commit a lot of sins here. 'Y' cord connections are bad. Mis-terminated (other than 75 ohm) transmission lines (cables) are bad. Series connected level controls are bad.

Wrong line terminations will cause SWR (ringing) problems along with changing signal levels. Changing R vs G vs B will shift the color balance - a _lot_. 'Y' connections introduce unintended tuned stubs into the signals which will form a trap filter. The cable lengths determines the tuning. Series level controls manage to simultaneously foul up the termination (SWR) and levels. Any wires longer than a few inches need to be 75 ohm transmission line and all transmission lines need to be properly terminated. 'Shielded' cable won't cut it.

So Mr Downer, what's the answer? Use distribution amplifers with a gain of 2 to correct for the incoming and outgoing terminations. If you truly require separate RGB levels (why is beyond me), use a separate amplifier for the variable gain stage. Maxim, ADI and many others make triple video amplifiers for just this purpose. I strongly recommend dual balanced power supplies and DC coupling to avoid lots of coupling capacitors. Video is not difficult if you avoid the pitfalls I mentioned. If you think your solution is 'good enough', it won't be.

GG

Sure is! I didn't expect that! Cool :-)

True. I want my circuit to be as "idiot proof" as possible, meaning that whatever combination of monitor/TV and switch setting I choose I can't blow anything up. The Atari colour and mono monitors can probably be regarded as "old" and care has to be taken with the sync signals, so this is why I'm designing it so that I'll be on the safe side.

You mean for the SCART connector? Yes, those values aren't definite, so you may very well be right. I might even have to experiment a little.

That sounds like a great solution for your setup. I used to have an NEC Multisync II which I used with my ST. Now I have an SM-144 for mono mode however and it's a LOT better. I really use mono mode the most, but want to make the computer setup as flexible as possible.

Probably the same sort of compromise my Multisync II gave me. The opposite was the fact there (although I've never had a dedicated Atari/ Atari compatible colour monitor to compare with) where colour was great but mono was fuzzy and unclear. Not good on the eyes. It was a great relief to exchange it for an Atari SM-125 mono monitor, then later the SM-144.

That may be true. I don't know what I'll be ending up with myself when it comes to display options, but most likely it'll be a VGA monitor of some sort, connected to the VGA (mono) output. The Mega STe will be recased (inside a 19" rack enclosure) and modified in many ways and will be more or less a dedicated MIDI computer for music purposes. Still, I might play the odd game now and then which is why I don't want to restrict it just for mono mode.

Yeah, sure did :-) Glad to see other Atari users out there.

Hallvard

I must admit that I'm pretty ignorant when it comes to electronics beyond the practical side of it, and have almost no idea how to design a circuit myself. The schematic I've made is an adaption of several other circuits I've found, so I may very well have made some grave mistakes along the way.

This is so that with a VGA colour monitor I can choose if I want the standard paper-white display in monochrome mode or adjust it to green, orange, red, yellow or whatever I may want. Still monochrome, but not just the plain black/white. I just got an idea that perhaps I should add a "bypass" switch so that if I want to revert back to black/white I don't need to adjust the trimmers each time.

I want the quality to be as good as possible. After all it can be seen as an investment in your eyesight if you use the computer for long periods of time. If I can achieve a sharp, vivid image I want to aim for that.

Video may not be difficult for those who know what they're doing, but for me I would have no clue as how to improve on the circuit with the things you just said. If you use Eagle I can upload the schematic file so you can edit the circuit with those improvements.

As for my initial question about an electronic switch which bypasses the physical 6PDT switch -do you have any thoughts on that?

Hallvard

you get a much sharper picture from a monochrome vga display. get some coloured glasses.

Hi Halvard :-)

Okies, progress this end.. I'm boxing mine up this weekend and using a QPDT switch to control the 8-FUNCTION, 4-MONO DETECT, VSYNC and HSYNC lines. That'll be about it really.

Aside from anything that should or shouldn't be done, I'm not going into mainstream production with it or attempting to win a coveted electronics design award so that's how it'll stay :-)

The ST definitely outputs weird slightly off standard signals though. I've had the Playstation running via both video and rgb and it doesn't mess up the reds, same with a Sinclair Spectrum 128k+ too. The ST's red just seems to be a bit over saturated and the composite sync is just slightly out.

The lead (almost identical to yours) has now been tested on three setups; Samsung LCD 19" TV, Panasonic 32" LCD TV, and a separate Commodore 1084/VGA setup.

Ahhh, the NEC multisync II. :-) Mine ended up with UV attacking the plastic so it looks horrible now. That and it's a big like looking at a fishbowl. :-)

As for an electronic switch. There are the 4016 variations although they're not really suitable for video signals. There are a few other video multiplexers and other things, but by the time you've messed about getting them, paying for them etc. etc. time has passed.

I wonder if one of these small little electronic KWM switches might be any good as a smart little out of the box solution...

Aly :-)

the Spectrum's red isn't full intensity. iirc Y=1/4 for the bright red, same as the colour bar test pattern.

Bye. Jasen

If it works for you you should be fine :-)

Anything that can be done to improve on this? Perhaps it's just an artifact we have to live with :-)

So these "electronic switches" are called multiplexers? I guess not all "switches" are suitable for video and you need specific ones that handle those frequencies. I've worked hard with my schematic lately and come up with some new ideas and a possible solution to the problem with the Atari monitors where the circuit wouldn't know if an Atari mono or colour monitor had been inserted as well as shutting off any accidental mono/colour switching which could fry the monitors. As part of my schematic is theoretical I could need some help finding which parts are to be used and how to connect them. When that's done I would like to improve on the actual signal and hope Stratus could help with that.

OK, first, here's my latest schematic:

This is how it should work: An electronic switch (U1) of some sort (a "multiplexer" IC?) works as a multiple 3 way switch. Those 3 positions are: A: when an Atari mono monitor (SM-124 etc.) is present B: when no Atari monitor is present at all C: when an Atari colour monitor (SC-1224 etc.) is present

In positions A and C the manual colour/mono switch (S1) is bypassed so changing the modes will have no effect whatsoever (and not allow anyone to accidently fry the monitor(s) in use). In other words, whenever an Atari monitor is used it will work just like with an unmodified Atari ST. In position B however the manual switch (S1) will be in control.

Control of the electronic switch (U1) for positions A and C is dependant on some logic (AND/OR gate ICs?), and to determine this a micro-switch is to be placed inside the 13 pin DIN (female) connector. This switch will sense whenever an Atari monitor's (male) connector is inserted. So now we have a system which "knows" when an Atari monitor (mono or colour) is present or not. Whenever the switch is off (e.g. no Atari monitor is present at all), position B is selected meaning that the manual switch (S1) is in control.

Next we need to determine if that Atari monitor is a colour or mono monitor. A mono monitor as we know has pin 4 (mono detect) connected to GND (pin 13) internally, which the Atari colour monitor doesn't. I assume this will generate a logic high/low (correct me if I'm wrong) and can further be used to control the U1 logic switch circuit and set it in position A or C. Positions A and C will as mentioned above shut off the manual switch, and also give sync signals only to those connectors (TV or VGA) which correspond to the mode which the computer is currently in.

Finally, there's the LEDs. If the pin 4/GND issue above works I would think that yet another logic circuit can be used, only this time for controlling two LEDs. One LED indication "colour mode" and another LED for indicating "mono mode".

Any comments on this? Anyone have any practical and concrete suggestions as to how I should get this done?

Hallvard

Are there any better suited newsgroups, web-forums or mailing lists for video electronics related issues? I'm really digging into unknown territory and need help in figuring out my circuit.