ZX81+35 Clone

[mahjongg]

yes, I could use several diodes to limit the voltage at the base, and use a schottky diode to turn off the emitter follower completely, (base below 0.6V) problem is that I have very little space in that corner, not enough to add three diodes.
I have to think about this, and experiment.

I also note that the actual value of the terminating resistance should not affect the signal amplitude too much, although it should normally be around 75 ohm, not far larger or smaller.

I'm sure now that now I know what the problem really is, I can find a suitable solution.
It seems I was simply blind to the fact that the signal was far too large 5V instead of about 1V, but I realize that now.
thanks all who helped, and gave me insight into the problem...

[mahjongg]

I did a spice simulation, to find the simplest circuit that would result in the required levels, and came up with this:

sync combiner in PSPICE.png

(23 KiB) Downloaded 365 times

This circuit produces the following levels:
White 1.2V
black 0.31V
sync 0.017V

[simon-zx81]

I was thinking of building my own homebrew ZX81 and wanted to look at existing attempts...so My questions is Does anyone have a list of ZX81 Clones...and links to them just out of curiosity ?

Thanks

[1024MAK]

Okay, this is how the ZX Spectrum 128 and the +2 (grey version) machine handles the composite video. It's labeled +3 as this is the circuit used to add composite video to a +3.


Of course it will not be "tuned" to the exact levels that you need, but the idea may be helpful.

Mark

[mahjongg]

In this case, everything happens inside the "blackbox" (ULA), so no information on how it works can be gleaned from it.

@ simon-zx81 I have an extensive list of ZX-81 clones, will report later with a list.

[mahjongg]

This is a list of material I used to design my ZX81 + 35 clone:

ZX81 rebuild projects I investigated to get inspiration from:
wilf rigter's "THE ZX97LITE SCHEMATIC DISASSEMBLED Rev.pdf"
wilf rigters "turbo.pdf"
wilf rigters "ZX81WAIt.pdf"
Kevin bakers, "turbo switch circuit.gif" redrawn by wilf rigter
Rodney knaaps ZX81 rebuilds # 1, #2 and #3
"ZX81 ULA Replacement By Andrew Rea.pdf"
"zx96mainboard schematic.gif"
Alwin Henseler's "ZX81_ULA-in-a-CPLD_v1.10.zip"
grant searle's "ZX80nmiVer4.1.jpg" slow mode extension for ZX80
PokeMon's "ZX80CORE-manual-EN.pdf"
Andy Rea's "mini ZX81" page 1 to 4
Other ZX80 /ZX81 info:
"ZX81 Service and Assembly Manual (scans) kopie.pdf"
"The Complete zx81 Rom Disassembly.pdf"
"Assembly Listing of the Operating System of the Sinclair ZX80..html"
"zx80_schema redrawn A3.pdf"
"Sinclair ZX80"81 etc Specifications.pdf"
"graphic view of zx81 rom.jpg"
"ZX81 schematic highres.jpg"
"Microace_Diagram_s1.jpg"
"annotated redrawn zx80-schema.gif" by NoCash
"brazilian slow mode adapter schematic .jpg" part 1 and 2
"ZX81CCP_schematic.jpg"
"jupiter ace 4000.pdf" by Bodo Wenzel
"cp-200 (ZX81 clone) schemas"
"tk82 (ZX81 clone) schemas"
"power 3000 (lambda 8300) schematic.gif"
"ts1500 schematic (scan).gif"

[mahjongg]

This is a list of ZX80 and ZX81 clones I know of (have pictures of, and sometimes schematics):
Beocomp ZX81 (a version of the ZX81 sold by B&O in the Netherlands)
CZ1500 plus
TS1500
ringo
CP-200
Prologica_CP200_system_s2
TK83
lambda8300
microace

[mahjongg]

okay, I made the modifications,
I have to say that even before the modification the picture quality was reasonably good. This time I had a camera so I took some pictures "before and after" of the TV and oscilloscope.
This is the TV screen before the modification, displaying the result of printing all chr$ from 0 to 255.

PTDC0252.JPG

(1.57 MiB) Downloaded 330 times

the scope picture shows there are some improvements needed, the screen was filled with chr$(5) characters resulting in alternating bars of black and white:

PTDC0257.JPG

(2.06 MiB) Downloaded 330 times

note the "glitch" on the front of the sync pulse, and note that the white level is almost 2 Volt.

The glitch is caused by the form of the sync pulse coming out of the NAND, here is a screen shot of it:

PTDC0270.JPG

(1.96 MiB) Downloaded 330 times

note the first microsecond of the pulse does not go lower than about 0.8V, while the rest of the pulse goes to about 0.2V above GND.

[mahjongg]

So this is how the signal looks after the modification:

PTDC0259.JPG

(2.01 MiB) Downloaded 327 times

note that the glitch is almost, (but not completely) gone, and that the white level is just slightly above 1 Volt, also the sync level is a bit larger, and now close to 0.3Volt. perhaps a better schottky diode will improve the glitch.
also note there is a pulse between the end of the sync pulse, and the start of the back porch. It seems that the sync and back porch signals do not overlap, and thus a pulse sometimes can be detected between them, (in this particular picture the pulse seems wider than it normally is) it does not seem to affect the brightness control. Perhaps stretching the sync pulse with a small R/C filter will remove the pulse.

I could not determine a signal quality change, but this is a screen picture after the modification:

PTDC0262.JPG

(1.58 MiB) Downloaded 327 times

[PokeMon]

Looks impressively good.

About the glitch when switching from sync to back porch:
Maybe it is possible to activate the back porch even during the sync pulse as it doesn't matter at this point (signal is low anyway) and would avoid that small glitch. Maybe this can be done easily with no or not much additional hardware support when accessing your counter. I think that TV would not have a problem with it in general as it is very short and in a non displayed region of the TV.