Sinclair ZX80 / ZX81 Forums

[sirmorris]

Hiya,

to make a reset-proof area you will need to have data stored above 32k. Or hack the ROM The RAMTOP poke will only protect against NEW.

It's usual to have rom and RAM shadows.

I've replaced the diodes on my board but there's no difference. The shift behaviour is still broken

I'll investigate some other options.

C

[superfo]

sirmorris wrote "I've replaced the diodes on my board but there's no difference. The shift behaviour is still broken"

That's what I thought, because only a few keys that don't work with shift key.
I don't have time to investigate it yet.

[siggi]

sirmorris wrote "I've replaced the diodes on my board but there's no difference. The shift behaviour is still broken"

That's what I thought, because only a few keys that don't work with shift key.
I don't have time to investigate it yet.

I also had some strange key-problems, when I built my ZX81 Laptop (see http://forum.tlienhard.com/phpBB3/viewtopic.php?f=7&t=305&start=12):
in command mode, when I pressed the "P" key, I got the command word "CLEAR" instead of "PRINT". The next press on "P" gave the correct letter "P". So it was NOT a keyboard problem.

That effect disappearded, after I connected an external RAM pack. So it was a RAM issue, not a keyboard issue!

The problem is the old design of the ZX80/81, where the internal RAM's output drivers are enabled together with the complete chip ( /CE conneted to /OE). That causes a crash at the data bus, when the Z80 wants to write data into the RAM, because the RAM is enabled immediately when the address is valid and then also drives the data bus. But the Z80 also drives the data bus! 1 clock cycle after the address is valid the Z80 enables the /WR signal to write the data into the ram. And that is the moment, where the RAM releases the data bus (disables its output drivers). But then the data from the Z80 may not have enough time to stabilize and nonsense might be written into RAM.
That design works with old and slow RAM chips, but not with modern and fast chips, which enable their output drivers much faster!

So I feeded the /RD signal to /OE of the RAM chip. And the problem was solved!
(Later I used the signal /RD AND /RFSH to enable the ram, necessary for TRUE HIRES)

Maybe that is also the problem here!

Siggi

[Lee Hart]

Paul wrote:
> I soldered U4 on the component side... the pins of C8 seem very close, so it
> doesn't look safe to install a heat sink. Maybe for issue2 boards, there could
> be a slightly larger gap (or an axial type for c8 with ground near the heat sink)?

C8 is intended to be laid down on the board, not standing up. This makes room for a heat sink on U4 that can extend over the top of C8. It also helps to mount U4 up above the board a little bit, with a screw, nut, and 1/4" (5mm) spacer between U4 and the board.

I moved C8 a little farther from U4 on the issue2 board; this should help as well.

> the signal coming out of the mic is very low.

Yes, it is. Sinclair used C14=47pf and R35=1meg. These values reduce the signal to something like 5 millivolts peak to peak.

To increase the signal, try changing R35 to 100k and C14 to 470pf.

> The printing near w5 doesn't seem to match the schematics.
> There is 5V at '27128' where should be A13, and A13 at '2732' where should be 5V.
> The schematic also states 2716 here, where it should read 2732 like on the board.

Thanks! I have corrected them on the schematic and board layout.

> I also wonder about the ROM and RAM mapping... everything above 32K is
> mirrored. and 16K of ROM and 16K of Rom would be usable in this configuration.

Correct. It's the same as the original ZX80.
0-16k = ROM
16-32k = RAM
32-48k = ROM image (same as 0-16k)
48-64k = RAM image (same as 16-32k)
When the RAM or ROM chip is smaller than 16k, the code in it repeats.

siggi wrote:
> I also had some strange key-problems, when I built my ZX81 Laptop... The problem is
> the old design of the ZX80/81, where the internal RAM's output drivers are enabled
> together with the complete chip ( /CE connected to /OE). That causes a crash at the
> data bus...

Interesting... thanks! But I don't see why this would cause a problem on the ZX80. There are 1k resistors between the RAM and Z80 data buses, so a bus crash shouldn't hurt anything. When reading RAM, /MREQ is low; when reading the keyboard, /IORQ is low. This should mean only one or the other can be enabled at the same time, even with the RAM's /CE connected to /OE.

But, maybe there is something odd in the way the Sinclair software reads the keyboard that makes it matter if /CE is tied to /OE? It would be simple to lift the RAM's /OE pin and tie it to the Z80's /RD as a test.

[siggi]

siggi wrote:
> I also had some strange key-problems, when I built my ZX81 Laptop... The problem is
> the old design of the ZX80/81, where the internal RAM's output drivers are enabled
> together with the complete chip ( /CE connected to /OE). That causes a crash at the
> data bus...

Interesting... thanks! But I don't see why this would cause a problem on the ZX80. There are 1k resistors between the RAM and Z80 data buses, so a bus crash shouldn't hurt anything. When reading RAM, /MREQ is low; when reading the keyboard, /IORQ is low. This should mean only one or the other can be enabled at the same time, even with the RAM's /CE connected to /OE.

The hardware is not destroyed, but the data on the data bus.
And pressing a key on the keyboard affects the address bus load, so the address lines of the RAM might get stable later when a key is pressed ...

But, maybe there is something odd in the way the Sinclair software reads the keyboard that makes it matter if /CE is tied to /OE? It would be simple to lift the RAM's /OE pin and tie it to the Z80's /RD as a test.

Yes, that would be a good test.

Siggi

[siggi]

Hi Lee
there might be also another timing problem concerning HIRES graphics. Wilf describes the problem here:

http://www.rwapservices.co.uk/ZX80_ZX81/forums/ram-at-8-12-eeprom-at-12-16k-t66s10.html#p752

"In the previously posted ZX RAMPAGERv1 schematic there is a problem with timing of memory access during the time that RFSH is low while MREQ is high. This short interval is important as the video pattern datat is fetched from ROM or SRAM and the data bus settles during that time. The ZX RAMPAGERv2 circuit shown fixes that problem ..."

I had that described problem with my Laptop during hires display: some hires columns were not shown correct (completely black or white). After I added the /RFSH signal to the address decoder, the display was o.k.

Siggi

[KnightFire]

This forum topic has been very quiet, is no news good news?

[Paul]

Neither good nor bad.
At the moment 4 out of 5 Boards are completely built.
One does not work (Jens), it gives no signal to the screen, there will be further investigation this weekend.
Two work with the known error of the keyboard.
SirMorris replaced the diodes with no effect.
My board still works fine.
I cannot save and load data because I don't have a working cassette recorder with mic input, only with line in.
The signal is to weak for line in.
I don't own cassettes with programs for zx80 or zx81, so I cannot test the loading as well.
Without the possibility of saving and loading, i don't like the idea of typing in big test programs.
Hopefully the board of Jens will be running by the end of sunday.
If it isn't, he's planning to come along on the 31st. Then he might bring cassettes and recorder to test, and we might be able to get his board running.
Lee is still working hard on his board and his manual.
Meanwhile, has anybody good suggestions for a simple way of loading/saving software to the zx80 to test using a pcs audio equiment? As I only have several macs and one Netbook, I dont own a parallel or serial port. Do usb to serial converters work for transfering data to the zeddy?
I just think about using eightyone as a source, create a file with the test program and burn it to my eeprom from 8 to 16 k.
Any tipps on how to do this in practice?

Greets Paul

[zx81jens]

hi to all!

well, my ZX80-LEGACY is fully assembled and... not working ... there is just a blank screen and i can´t find the mistake!

i have a problem with the 5V-line - it doesn´t come thru to some points... but a solution is near, because OLIVER will come to my house on sunday.

i will also drive to PAUL on the 31. of Juli !!

so there will be a lot of fun for the next few weeks

many greetings
jens

[sirmorris]

Jens - I'm sure you already know but it's worth asking if you put the required 3 jumpers on the appropriate pins of the 50 way expansion connector?

Without ar least one of these jumpers there will be no video signal. The information is in the manual but it's easily missed.

Charlie