Pinball lamp driver boards, scr's, and the prop tv driver.....Strange Idea.
IncVoid
Posts: 40
Forgive me for my slang and lack of understand of electronics and bringing this here, but It just seemed to "click" together.
Mr and Mrs. Pacman pinball has a lamp driver board which I'm not quite sure about its operation.
But it has Silicon controlled rectifiers? SCR?
They appear to work as transistors that "lock on" until a voltage drops and they kick off.
The driver board seems to have ac bridge(4 diode network?) rectifier powering the lights.
Those lights get turned on by a bunch of decoder chips, that trigger the SCRs one by one allowing the lights a path to ground.
I was arguing with my dad about not seeing anything in the schematics about being able to turn off the SCRs. I think the they naturally turn off everytime voltage drops to 0. Which I think happens at 120 hz... So the mpu board would have to turn the lights on at least 120 times per second.
I did see a type of feed back coming through a SCR, or a "loop back". But I'm not quite sure if that is what it is used for.
It appeared that it turned on an SCR that went back to the lamp strobe line...I was thinking it would set that one last and wait for it to fall to 0 indicating it was time to update the lamp SCRs again.
What does this have to do with the prop?
I naturally thought hmm 60hz...(before I realize they are turning off everytime it hits 0 volts, 120hz)
I wonder if the prop tv driver could be modified, just hack off the signals for vertical retracing and all that, and have it spit out the decoder info as its "pixel data".
The particular board had a quad flipflop and 4 chip selects to an 8 line decoder chip..passing on the number to one of the selected chips, which would toggle a line high, turning on that SCR and thus allowing a light to turn on. So I was thinking of making "lighting" data as a scanline..where every other pixel is a clock pulse with the relevant light to turn on.
So a single scan line could hold the entire pinball lighting information.
Updating scanline and waiting for the loop back SCR to toggle and send out another "scan line"
Anybody else think of using cog "video" hardware to stream out other data?
I might have a concept error or two, but it just hit me randomly thinking, if those SCRs are turning off at a multiple of tv frequency 60 hz.
If we can update 640x480 pixels at 60 hz, I'm pretty sure we can update 50 SCRs in that time.
So fast that those lights are "on" Most of the time, producing flicker free lighting hopefully.
Just..always "day dreaming" about pinball design, and interfacing with a prop etc.
Just really seemed to mesh using the streaming unit in a cog to send out "lighting" frames..eh
Hope I wasn't too confusing, and that i'm not barking up the wrong tree in the parallax forums.
On a side note, I went to a pinball show in may, and won a silent auction that had 98-2001 nuts and volts magazines...I think everyone had a parallax stamp ad on the back. It is an awesome magazine and wish I was reading it 10+ years ago.
Mr and Mrs. Pacman pinball has a lamp driver board which I'm not quite sure about its operation.
But it has Silicon controlled rectifiers? SCR?
They appear to work as transistors that "lock on" until a voltage drops and they kick off.
The driver board seems to have ac bridge(4 diode network?) rectifier powering the lights.
Those lights get turned on by a bunch of decoder chips, that trigger the SCRs one by one allowing the lights a path to ground.
I was arguing with my dad about not seeing anything in the schematics about being able to turn off the SCRs. I think the they naturally turn off everytime voltage drops to 0. Which I think happens at 120 hz... So the mpu board would have to turn the lights on at least 120 times per second.
I did see a type of feed back coming through a SCR, or a "loop back". But I'm not quite sure if that is what it is used for.
It appeared that it turned on an SCR that went back to the lamp strobe line...I was thinking it would set that one last and wait for it to fall to 0 indicating it was time to update the lamp SCRs again.
What does this have to do with the prop?
I naturally thought hmm 60hz...(before I realize they are turning off everytime it hits 0 volts, 120hz)
I wonder if the prop tv driver could be modified, just hack off the signals for vertical retracing and all that, and have it spit out the decoder info as its "pixel data".
The particular board had a quad flipflop and 4 chip selects to an 8 line decoder chip..passing on the number to one of the selected chips, which would toggle a line high, turning on that SCR and thus allowing a light to turn on. So I was thinking of making "lighting" data as a scanline..where every other pixel is a clock pulse with the relevant light to turn on.
So a single scan line could hold the entire pinball lighting information.
Updating scanline and waiting for the loop back SCR to toggle and send out another "scan line"
Anybody else think of using cog "video" hardware to stream out other data?
I might have a concept error or two, but it just hit me randomly thinking, if those SCRs are turning off at a multiple of tv frequency 60 hz.
If we can update 640x480 pixels at 60 hz, I'm pretty sure we can update 50 SCRs in that time.
So fast that those lights are "on" Most of the time, producing flicker free lighting hopefully.
Just..always "day dreaming" about pinball design, and interfacing with a prop etc.
Just really seemed to mesh using the streaming unit in a cog to send out "lighting" frames..eh
Hope I wasn't too confusing, and that i'm not barking up the wrong tree in the parallax forums.
On a side note, I went to a pinball show in may, and won a silent auction that had 98-2001 nuts and volts magazines...I think everyone had a parallax stamp ad on the back. It is an awesome magazine and wish I was reading it 10+ years ago.
Comments
But keep thinking, I'm sure the Prop would be very useful inside a box of bells, lights, and bumpers
Which seems silly. I understand the point of using decoders to save pins.
Why have 28 pins to 28 scrs. They don't have any flipflops to "hold" the state of the SCRs. Which is why they need to update faster than the AC power frequency.
But as it is, I would just have to hook up 8 lines.. 3 address, 4 chip selects, and a clock and i'm turning lights on. I'll have to re-read the tv driver theory of operation.
It would only work in vga mode. the ntsc mode only outputs out a nibble if I read it correctly, and I'd have to disable modulation. Etc. I thought it was a good idea Guess it isn't that intuitive
The fact that there is a hardware decoder chip means that you will have to refresh it as you correctly summarized. We merely pointed out that SCRs themselves do not need to be "refreshed" as if they needed constant triggering when in fact they only require DC. The refreshing demanded is a trade-off in the original design to cut down the number of I/O and simply employ a tiny bit of CPU time. Refreshing 28 "addresses" 120 times a second is trivial even on the really old and slow 8-bit CPU chips, and even then there is a lot of time left over to do other things.
Use one or more 74xx595 chips and you can more or less control as many SCR's and Triacs as you want. Four '595 chips would allow you to control 32 scr/triacs and a pasm program could update them fast enough to use them as dimmers.