RetroBrew Computers Forum
Discussion forum for the RetroBrew Computers community.

Home » RBC Forums » General Discussion » ZX81+38 Build (Description of my build of the ZX81+38 board )
ZX81+38 Build [message #7908] Tue, 11 August 2020 22:49 Go to next message
ale500 is currently offline  ale500
Messages: 44
Registered: April 2018
Member
Using the info available here: https://revspace.nl/ZX81plus38_simple_to_build_ZX-81_clone I decided to get some boards made and order some parts (mostly from ali) and get (again) a ZX81 working. This is a computer I had as a kid, well a Brazilian clone called TK-85. We had quite some fun even if it was b&w and without sound.

The boards arrived some week ago, I got 5 made and populated one. Most chips I tested with a MiniPro and for the Z80 I made a breadboard SBC. Yesterday I finally got the crystals and the first very funny issue appeared (I should have got all parts from Reichelt, I know). The oscillator produces a square wave of only ~30 kHz !... never seen that. The crystals are from Mouser. Nonetheless It seems to work, it produces the video signal and one can see that the processor runs and memory gets checked but at snail pace (the Z80 is a CMOS version).
Re: ZX81+38 Build [message #7909 is a reply to message #7908] Wed, 12 August 2020 01:46 Go to previous messageGo to next message
ale500 is currently offline  ale500
Messages: 44
Registered: April 2018
Member
I did a couple of measurements with the 74HC14 (used in the oscillator), and it has a input-to-output delay of ~6 ns. Using a 15 MHz Crystal and 2 10 pF caps I get a nice 15 MHz square wave. With the 13 MHz crystal I get a nice 13 MHz sq. wave. I think I may have used the wrong caps, these 22 pF caps I had may be are something else.
Replacing the caps with two 10 pF ones makes the oscillator work at the desired 13 MHz.
Re: ZX81+38 Build [message #7910 is a reply to message #7909] Wed, 12 August 2020 09:17 Go to previous messageGo to next message
techdad is currently offline  techdad
Messages: 2
Registered: August 2020
Location: Virginia, USA
Junior Member
Were you able to find the BOM anywhere?
Re: ZX81+38 Build [message #7911 is a reply to message #7910] Wed, 12 August 2020 11:49 Go to previous messageGo to next message
ale500 is currently offline  ale500
Messages: 44
Registered: April 2018
Member
I used the schematic to populate my board. It has some issues, the processor freezes. It seems to have a very noisy +5 V rail, ringing all around and noisy reset line.
Re: ZX81+38 Build [message #7913 is a reply to message #7911] Thu, 13 August 2020 21:56 Go to previous messageGo to next message
ale500 is currently offline  ale500
Messages: 44
Registered: April 2018
Member
The frozen processor I traced to nWAIT being low all the time. no NMI pulses, the '74 flip flop (U18A) doesn't get preset, nWAIT never goes high. I wonder how the circuit starts: the flip-flop U13C/D starts with nNMI being high both inputs are high.
If I force nWAIT high, then the processor starts normally.

[Updated on: Fri, 14 August 2020 06:24]

Report message to a moderator

Re: ZX81+38 Build [message #7969 is a reply to message #7908] Mon, 14 September 2020 09:47 Go to previous messageGo to next message
ale500 is currently offline  ale500
Messages: 44
Registered: April 2018
Member
I got my board working. I swapped the outputs of the NMI flip-flop. Why it works when it shouldn't is still a mystery to me.
Firstly I AND'ed /NMI and /RESET before the PRE input of U18A (the '74 that generates /WAIT), as I thought that for some reason the initial condition was not being met. Didn't solve the problem but showed that the processor runs till the first HALT opcode is executed and the NMIs should start. What I see is that while the NMIs are disabled, they are still being generated and after the NMIs are enabled, and HALT is executed the processor stops because the NMI never presets the '74 flip-flop. I concluded that the RS flip-flop U13A/U13B was the culprit, not being set/reset correctly, i.e. A0/A1 are swapped somehow.
At the beginning of the ROM it does a OUT FD to disable the NMI, it should set pin 11 of U13A.
I'll capture a couple of oscillograms to see the flipflop in action. Anyways my TV cannot display the picture, these new tvs are not forgiving at all..

Re: ZX81+38 Build [message #7990 is a reply to message #7908] Thu, 01 October 2020 23:25 Go to previous message
ale500 is currently offline  ale500
Messages: 44
Registered: April 2018
Member
I finally found the problem, it seems that the circuit around the Flip-flop to generate the nWAIT signal does not work, it never gets 'Preset' because the NMIs are disabled. One get a picture if the nWAIT signal is pulled high but with some distortion on the upper part of the screen. A two transistor fix, like in the original ZX81 is needed. A One transistor circuit should also work, I think. The ZX97 has two gates to generate this nWAIT signal, I'll try that because at the moment I have no PNP transistors....
Previous Topic: 386EX marking question: EXTB or EXTC... (3.3V or 5V)?
Next Topic: Help with booting OS-8 on SBC6120 Community Edition board


Current Time: Fri Apr 19 17:04:08 PDT 2024

Total time taken to generate the page: 0.00806 seconds