ZX80 Core - new ZX80 motherboards

Any discussions related to the creation of new hardware or software for the ZX80 or ZX81
Lee Hart
Posts: 48
Joined: Fri Mar 19, 2010 11:00 pm

Re: ZX80 Core - new ZX80 motherboards

Post by Lee Hart »

PokeMon wrote:
Lee Hart wrote: HC and HCT are essentially interchangeable. There is no internal difference...
They are interchangeable from logic function but have not (definetely not !) internal design and not same data characteristic. Do you ever read datasheets?
Certainly! :-) The data sheets are different, but the parts are the same. The data sheet simply describes how the part was tested. With rare exceptions, the internal circuits are identical between HC and HCT logic. The only change is that HCT parts have a lower input switching threshold voltage. Do you have any references that show otherwise?
PokeMon wrote:...the HCT series are made for 5V operation voltage only as the HC series allows a wide range of operation from 2.0 Volts to 6.0 Volts. Try to compare HC and HCT at a different supply voltage and you will be surprised about their behaviour.
I have done exactly that. I find that HC and HCT are interchangeable in any circuit where the input threshold voltage difference is not an issue. For example, I have been selling this little computer for several years http://www.sunrise-ev.com/MembershipCard.htm. I've mixed HC and HCT chips, from various vendors, and operated it on supply voltages from 3v to 5v. I have never had any issues in mixing the two families, as long as there are no bipolar TTL parts to produce weak high logic levels.
PokeMon wrote:
Lee Hart wrote:The root cause is that Sinclair's cheap design does not meet TTL logic level specs at the inputs of U10... There is a lot of noise on the address bus...
this could be solved and easily managed with a custom program in FAST mode and a special written keyboard input routine.


That is a good approach, and I tried it. The problem is that the keyboard reads correctly most of the time. I have a good scope, but it is not a storage scope. I have no way to trigger on a bad read, or capture what happened during it. Even though a 10:1 scope probe has low loading, it still affects operation. Take a look at the ZX80 data and address buses. They are amazingly noisy!

I am sure someone with a logic analyzer could find this problem quickly. But I do not have access to one, and have to rely on simpler methods.
User avatar
PokeMon
Posts: 2264
Joined: Sat Sep 17, 2011 6:48 pm

Re: ZX80 Core - new ZX80 motherboards

Post by PokeMon »

Lee Hart wrote: Certainly! :-) The data sheets are different, but the parts are the same. The data sheet simply describes how the part was tested. With rare exceptions, the internal circuits are identical between HC and HCT logic. The only change is that HCT parts have a lower input switching threshold voltage. Do you have any references that show otherwise?
Your answer is quite inprecise. What do you mean with the data sheets are different but the parts are the same ? This is very simplified. If you mean the function is the same, you are right. But the "part" is the same ? I have at home two tables, one made of wood and one made of glass. These tables seem to have the same function but look different, have different weight, one you can see through, the other not. :roll:
The interface circuits (input) are definetely different and this is a major part of that circuit.
You can take a more detailed look at this application note from Fairchild:
http://www.fairchildsemi.com/an/AN/AN-368.pdf

And a datasheet describes how a part was tested but that's not the purpose of a datasheet. A datasheet describes the characteristics of a circuit under several conditions. The characteristics of a circuit under defined test conditions is the important information for an electronic engineer and not to know "how a part was tested". So the point of view is important and you have to transfer the characteristics into your design. Testing circuits is not a self purpose.

Lee Hart wrote: I find that HC and HCT are interchangeable in any circuit where the input threshold voltage difference is not an issue.
Well the correct interfacing of circuits is ALWAYS an issue especially when you have problems like with your Shift key.
If you have problems like this then your design is wrong. It's that simple. Computer do what you tell them to do. The same with electronic.

Lee Hart wrote: That is a good approach, and I tried it. The problem is that the keyboard reads correctly most of the time. I have a good scope, but it is not a storage scope. I have no way to trigger on a bad read, or capture what happened during it. Even though a 10:1 scope probe has low loading, it still affects operation. Take a look at the ZX80 data and address buses. They are amazingly noisy!
Any working computer system is noisy on the signals. TTL quite more than CMOS but is not sensitive in the same way. CMOS is more sensitive for induced voltages, parasitic effects and can be influenced from other signals. This is also an issue of routing the signals, pull up choosement. Impedance is also a big difference between TTL and HC/HCT series. :mrgreen:
TTL has a lower (asymmetic) noise gap but less sensitivity for influence effects.
User avatar
questarian
Posts: 57
Joined: Sun Jun 17, 2012 9:40 pm

Re: ZX80 Core - new ZX80 motherboards

Post by questarian »

...A little late once again, but if it helps I'd be interested in a couple boards as well.
User avatar
PokeMon
Posts: 2264
Joined: Sat Sep 17, 2011 6:48 pm

Re: ZX80 Core - new ZX80 motherboards

Post by PokeMon »

Okay - thanks.
The prototype is in production now and will be tested first before the planned series will be run. ;)
User avatar
PokeMon
Posts: 2264
Joined: Sat Sep 17, 2011 6:48 pm

Re: ZX80 Core - new ZX80 motherboards

Post by PokeMon »

Here is the latest version of ZX80 Core.

There are some more modifications made in the latest (now final) version:

1)
I added a connector to place a NMI generator board inside. This will be made as additional SMD board to place it as sandwich on a "pin head line" (2x 8 pins). This board will have the same function as the NMI generator from Grant Searle but a little more simplified with less components. And this board will have the video circuit ZX81CCB on it. Lets say it is a combination of both, should be very small, not bigger than the ZX81SCP board. The video part can be used either with original 5V signal or standard 1V composite video signal. I think the modulator may take 1Vss signal as well as the Sinclair 5V TTL version.
This is just an option, the signals SYNC IN/OUT and VID IN/OUT can be jumpered if not used or can be soldered on soldering side if somebody doesn't want to place a connector on the board for beeing more close to the ZX80 original.

I placed a video chinch connector at exact position which can be uesd instead of the modulator. So either modulator or chinch connector can be used. If the NMI/VID Board is used, the signal can be transformed to standard composite video signal with 1Vss.

2)
I added an additional connector for adapting an external keyboard to the ZX80. This was discussed at the german board. There are solutions for the ZX81 to connect a keyboard via slot edge connector to ZX81. This solution does not work with the ZX80, because the keyboard lines KBD0-KBD4 are not open collector. I thought about maybe disabling the driver by a possible unused or very rarely used signal (BUSAK for example) but throwed this because this does influence US/UK version and TapeIn signal. And I throwed the idea to change logic at this point from tri-state output to open collector as there were not too many interests in this external keyboard. So I placed a 16 pin connector (pin head 1x 16) near the keyboard with KBD0-KBD4,A8-A15,/RESET and power supply (VCC,GND). So it would be possible to adapt a keyboard if really needed. So a compromise solution.


So here is the latest (final) version.

Schematics:
ZX80CORE - Schematic.pdf
(260.17 KiB) Downloaded 337 times
Top layer:
ZX80CORE - PCB_topsilk.pdf
(138.48 KiB) Downloaded 315 times
Bottom layer:
ZX80CORE - PCB_bottom.pdf
(101.15 KiB) Downloaded 310 times
User avatar
PokeMon
Posts: 2264
Joined: Sat Sep 17, 2011 6:48 pm

Re: ZX80 Core - new ZX80 motherboards

Post by PokeMon »

Addition I have the component view and a 3D view of the board. 8-)

Component view:
ZX80CORE - Components.pdf
(24.56 KiB) Downloaded 301 times
3D view:
3D.gif
3D.gif (33.82 KiB) Viewed 4301 times
So lets hope the board is running in first step without modifications. ;)
sirmorris
Posts: 2811
Joined: Thu May 08, 2008 5:45 pm

Re: ZX80 Core - new ZX80 motherboards

Post by sirmorris »

Wooohoooo!!

Great work, Karl!
XorA
Posts: 98
Joined: Thu May 10, 2012 9:14 am
Location: Glasgow, Scotland, UK
Contact:

Re: ZX80 Core - new ZX80 motherboards

Post by XorA »

Looking good, now someone needs to laser mill some diamond cases for the boards!
User avatar
PokeMon
Posts: 2264
Joined: Sat Sep 17, 2011 6:48 pm

Re: ZX80 Core - new ZX80 motherboards

Post by PokeMon »

Maybe black marble would be good. 8-)
Alternative could be ivory or made of a unicorn. :lol:
User avatar
iturbez
Posts: 154
Joined: Fri Dec 25, 2009 1:06 pm
Location: Spain

Re: ZX80 Core - new ZX80 motherboards

Post by iturbez »

Pokemon wrote:There are solutions for the ZX81 to connect a keyboard via slot edge connector to ZX81.
Pokemon: What are theese solutions like?

thanks in advance.
Post Reply