FPGA replacment for the ZX8301 ULA and more

Nagging hardware related question? Post here!
lliont
Trump Card
Posts: 238
Joined: Sat Nov 22, 2014 9:18 am
Location: Athens, Greece
Contact:

Re: FPGA replacment for the ZX8301 ULA and more

Post by lliont »

thanks for the idea, we will see.
First I'll test the short pins tetroid suggested.


User avatar
Peter
Font of All Knowledge
Posts: 2023
Joined: Sat Jan 22, 2011 8:47 am

Re: FPGA replacment for the ZX8301 ULA and more

Post by Peter »

Chris_68 wrote: Sun Apr 07, 2024 7:07 pm I haven't all the posts but could it be possible to use your PCB and design an adapter board with the socket for CPU and 8301 routed to a DIN41612, this instead of having to reroute the PCB?
This can make sense for just two or three boards. But if a little series production is intended, I would not do it.
I don't have the best experiences with dual PCBs and lots of board-to-board connections involved.
Costs and manufacturing efforts tend to grow disproportionately.
If I was in Leon's shoes, I would finalize the decision, either internal or external, and then go for it, instead of a hybrid.


User avatar
Dave
SandySuperQDave
Posts: 2778
Joined: Sat Jan 22, 2011 6:52 am
Location: Austin, TX
Contact:

Re: FPGA replacment for the ZX8301 ULA and more

Post by Dave »

I looked at this project on Hackaday and it is quite interesting.

On the buffer, it would help to connect all unused inputs to ground or +5V. Between the buffer and the monitor connected pins of the 8301 socket, it would offer extra protection to add 5.1V zener diodes to ground.

Apart from that, everything looks great :D With some very careful design starting with SMD 'nail' type pins, the PCB could be reduced to the size of the 8301 IC itself.

To solve the problem of variable heights between the CPU and 8301 sockets, you might consider a hybrid flex/rigid design with two PCBs with a captive flex panel joining them. This would give them some flexibility to height of each half of the board.


lliont
Trump Card
Posts: 238
Joined: Sat Nov 22, 2014 9:18 am
Location: Athens, Greece
Contact:

Re: FPGA replacment for the ZX8301 ULA and more

Post by lliont »

It started as 2 separate boards but then I made a single board to eliminate the cables, that was ok in my ql with the changed sockets.
It was when I tested it to another QL I realized the height problem :(

I didn't tied the unused inputs to a fixed voltage because their outputs are always disabled (high resistance) by the G pin so they shouldn't oscillate at least their output stage.


User avatar
Dave
SandySuperQDave
Posts: 2778
Joined: Sat Jan 22, 2011 6:52 am
Location: Austin, TX
Contact:

Re: FPGA replacment for the ZX8301 ULA and more

Post by Dave »

Yeah, those red sockets are junk. I have had to replace dozens of them as they often fall apart if the CPU is stuck in them. Any time I installed an internal memory expansion in a QL at Sandy and it had a red socket, I'd fire up the weller first just in case. On my multitude of QLs, if I see a red socket I replace it with a plain low profile socket like the other ICs.

Do the equations implement the PAL/NTSC register? American "JSU" ROMs are the only ones I'm aware of that set this. Monitor mode is identical, but TV mode it sets different resolutions and uses a different character font. This breaks UK/EU Psion suite programs dramatically, and there is a custom US version of that suite.

Has anyone talked about the possibility of implementing a GD2-like driver to allow easy setting of different modes, bit depths, resolutions etc? It would be really nice to trade off colours to get much higher resolutions, or trade off resolution to get many colours.

Finally, would you consider adding an input on the logic "himem" which when asserted moves the address of screen memory to, say, 0xE00000? This way, forward-looking expansion designers could assert that signal to remap the video memory window up high. That would allow 2MB for a frame buffer.


lliont
Trump Card
Posts: 238
Joined: Sat Nov 22, 2014 9:18 am
Location: Athens, Greece
Contact:

Re: FPGA replacment for the ZX8301 ULA and more

Post by lliont »

As Peter and one's complement suggested an expansion card board is ready and it solves many problems.
It also needs two wires and a small board over zx8301 socket but it is so much easier to install and no need for 3d printed plate for the ports.
I also changed fpga to the newer cyclone 10 LP that I finally found in stock.

I am testing it now.

1713972338335.jpg


User avatar
Popopo
Trump Card
Posts: 175
Joined: Wed Apr 07, 2021 10:37 am

Re: FPGA replacment for the ZX8301 ULA and more

Post by Popopo »

Such a great job!
Congratulations & :) Thank you :)


lliont
Trump Card
Posts: 238
Joined: Sat Nov 22, 2014 9:18 am
Location: Athens, Greece
Contact:

Re: FPGA replacment for the ZX8301 ULA and more

Post by lliont »

Thank you
I forgot to mention it is now running at 16Mhz and I intend to add a fast serial port SER3.


Chris_68
Brittle Membrane
Posts: 102
Joined: Thu Aug 31, 2023 1:44 pm

Re: FPGA replacment for the ZX8301 ULA and more

Post by Chris_68 »

Leon, what an achievement, really great!!


User avatar
Peter
Font of All Knowledge
Posts: 2023
Joined: Sat Jan 22, 2011 8:47 am

Re: FPGA replacment for the ZX8301 ULA and more

Post by Peter »

lliont wrote: Wed Apr 24, 2024 4:35 pm It also needs two wires and a small board over zx8301 socket but it is so much easier to install and no need for 3d printed plate for the ports.
One or two wires are not too bad. For QIMSI I also needed one for the QL speaker.
I had the luck that one ROM port pin has no dedicated funktion, so the card remained pluggable.
Maybe some expansion port pins are forever so useless they could be cut an re-dedicated?
lliont wrote: Wed Apr 24, 2024 4:35 pm I am testing it now.
Good luck! I looks really traditional, sticking out of the QL case a little. Like in the good old days. ;)
You seem to design as many PBCs in a month as me in a year. :D


Post Reply