I would stay away from 7 segment displays and any logic IC for counting and 7 seg decoding. I would like to keep everything discrete. Either 10 leds for each player using simple transistors to shift increment the score, or, with the exact same transistor shift increment, provided the output transistors are open-collector and can withstand the HV, send the 10 outputs to nixie tubes for a 0 to 9 count display.
Basically, 1 all transistor circuit, 2 optional displays, 1 LV for leds, of with an HV supply, nice numeric tube display.
I would stay away from 7 segment displays and any logic IC for counting and 7 seg decoding. I would like to keep everything discrete. Either 10 leds for each player using simple transistors to shift increment the score, or, with the exact same transistor shift increment, provided the output transistors are open-collector and can withstand the HV, send the 10 outputs to nixie tubes for a 0 to 9 count display.
Basically, 1 all transistor circuit, 2 optional displays, 1 LV for leds, of with an HV supply, nice numeric tube display.
What about printing the score on the screen? In an earlier thread on a DIY analog computer, someone (possibly GK?) calculated the patterns needed to print numbers to a scope in a purely analog fashion, so that *could* be used, though I'm not sure how complex the implementation would be.
Failing that you could do a pair of discrete resistor ladder DACs for X and Y; the data for drawing the numbers would come from a ROM (you could use an array of diodes in place of the ROM to stay discrete), then just raster scan them out during a multiplexing time slot. That's how Tektronix did readouts on their CRT based scopes.
I would stay away from 7 segment displays and any logic IC for counting and 7 seg decoding. I would like to keep everything discrete. Either 10 leds for each player using simple transistors to shift increment the score, or, with the exact same transistor shift increment, provided the output transistors are open-collector and can withstand the HV, send the 10 outputs to nixie tubes for a 0 to 9 count display.
Basically, 1 all transistor circuit, 2 optional displays, 1 LV for leds, of with an HV supply, nice numeric tube display.
What about printing the score on the screen? In an earlier thread on a DIY analog computer, someone (possibly GK?) calculated the patterns needed to print numbers to a scope in a purely analog fashion, so that *could* be used, though I'm not sure how complex the implementation would be.
Failing that you could do a pair of discrete resistor ladder DACs for X and Y; the data for drawing the numbers would come from a ROM (you could use an array of diodes in place of the ROM to stay discrete), then just raster scan them out during a multiplexing time slot. That's how Tektronix did readouts on their CRT based scopes.
But you've just ~doubled the complexity of the project. If you want to accommodate that as an optional add-on the current multiplexer logic will have to be scrapped and majorly redesigned. I say just stick to the current specification.
Also, yes, that was me in the analog computing thread where development of my "Fourier synthesis character generator" was documented:
http://www.glensstuff.com/fouriersynthchargen/fouriersynthchargen.htm
Might as well mention it; The other night in a fit of immodesty for the first time I tipped a personal project for the Hackaday blog!
Tim, great work, if you're fussy about layouts they do take some real time.
May I ask what size annular ring size you're using and the pad hole size ?
I've had some pads lift when reworking PCB's and as a result my preference is for larger pads to minimize the chance of this.
When I can I stick with 100 mil and maybe alter the shape from round if more clearances are needed.
As promised:
http://timb.us/PDF/Scope_Pong_Pad_Sizes.pdf
Tim, great work, if you're fussy about layouts they do take some real time.
May I ask what size annular ring size you're using and the pad hole size ?
I've had some pads lift when reworking PCB's and as a result my preference is for larger pads to minimize the chance of this.
When I can I stick with 100 mil and maybe alter the shape from round if more clearances are needed.
As promised:
http://timb.us/PDF/Scope_Pong_Pad_Sizes.pdf
Tim, great work, if you're fussy about layouts they do take some real time.
May I ask what size annular ring size you're using and the pad hole size ?
I've had some pads lift when reworking PCB's and as a result my preference is for larger pads to minimize the chance of this.
When I can I stick with 100 mil and maybe alter the shape from round if more clearances are needed.
As promised:
http://timb.us/PDF/Scope_Pong_Pad_Sizes.pdfEnlarge all the pads hole to at least 0.8, especially C2.
Enlarge all the pad outer diameters to 1.6, and make C3's pad size 1.8.
Remember, home build PCBs usually are drilled with 1 drill size for as much of the PCB as possible as we may be working with something like a mini Dremel drill press.
For the 1N400x diodes, make the holes 1mm, pads 2mm.
Tim, great work, if you're fussy about layouts they do take some real time.
May I ask what size annular ring size you're using and the pad hole size ?
I've had some pads lift when reworking PCB's and as a result my preference is for larger pads to minimize the chance of this.
When I can I stick with 100 mil and maybe alter the shape from round if more clearances are needed.
As promised:
http://timb.us/PDF/Scope_Pong_Pad_Sizes.pdfEnlarge all the pads hole to at least 0.8, especially C2.
Enlarge all the pad outer diameters to 1.6, and make C3's pad size 1.8.
Remember, home build PCBs usually are drilled with 1 drill size for as much of the PCB as possible as we may be working with something like a mini Dremel drill press.
For the 1N400x diodes, make the holes 1mm, pads 2mm.
You may be fine with all the pads and holes Tim, it's just one of my pet hates, annular rings too small.
I'm somewhat thinking of those that will be buying PCB's or using your files to etch their own PCB. Those that have done some PCB work will have no problem with your work I don't think, moreso those with limited experience taking on this as their first big project. If it's populated without mistake and zero components need to be reworked, then I don't see any problems.
After the effort I know you're putting into this I wouldn't want to see criticism of your efforts by those that have taken on something more complex than they should've.
It's a heap of work Tim, you and GK decide what the outcome should be.
It also adds some constraints on the design - e.g. not using top connections to parts that can't be top-side soldered, like multiturn trimmers & connectors.
Okay boys and girls (if there are any girls), I have finally finished all of the equivalent diagrams and amended circuit diagrams and have collated them into a single 25 page (21 Mb) PDF document. Here it is:
http://www.glensstuff.com/pong/Oscilloscope_Pong.pdf
Considering the size and number of holes you'd be pretty nuts to home-brew a PCB like this. For something like this, the last thing you need is the additional hassle of tracking down shorts & broken tracks.
It also adds some constraints on the design - e.g. not using top connections to parts that can't be top-side soldered, like multiturn trimmers & connectors.
A PTH board will be a lot more robust, and it's so cheap now, especially as I suspect enough people would be interested to do a small group order, that it's probably not worth going too overboard making the design optimised for homebrew PCBs.
That being said there's never any harm in making track widths & gaps as wide as practical, as this also increases yield on production boards and improves reworkability.
I like GK's all discrete analog construction, so I'm in the camp that thinks 7 segment displays or LEDs would be out of place here.
Okay, so I didn't get nearly as much done as I planned tonight, feeling a bit under the weather so I decided to make an early night of it.
I finished the Function Generator section, but forgot to upload the PDF/images for them, doh!
I also did the Multiplexer FET Gate Drive section; I implemented it as a sub-section of the Multiplexer Logic; the Horizontal and Vertical section will be placed directly above the Multiplexer section. That way the X, Y and Z BNC connectors will all be facing the same side on one board.
I also added some test points and cleaned up some traces in the Multiplexer Logic section.