System-4 CPU Card

The system card is a fairly standard 1802 setup. A circuit remaps the 4K EPROM from the bottom of the address space (0x0000) to the top (0xF000) after the first access to an address in the upper block following a system reset. The circuit on the lower left side of the card replaces the original interface to an Apple parallel keyboard. A preprogrammed PIC PS2 keyboard decoder (PAK-VI) connected to a PIC 16F84 which acts as a serial to parallel converter connect to the original 1852 parallel IO port.

http://cottageworker.com/wp-content/uploads/2014/05/

System-4 Memory Card

This is the top view of the 24K memory card, only 16K is populated. The jumpers on the card are for setting card location within the address space, the need for more memory never materialized.

http://cottageworker.com/wp-content/uploads/2014/05/

The Bottom of the memory card. Back then building a computer was a little like knitting a sweater, the hours of handy work far out numbered the hours doing real engineering work.

http://cottageworker.com/wp-content/uploads/2014/05/

System-4

This is the last of several 1802 based systems I built between 1978 and 1983. This system like all “real” computers has a system bus and separate Front Panel,CPU, Memory, and Video cards. The original system used an Apple parallel keyboard that was damaged many house moves ago. Last year the system was modified to use a PS2 keyboard everything else is original and seems to work :-). The system uses Tiny Basic or a modified version of the original Popular Electronics monitor program.

http://cottageworker.com/wp-content/uploads/2014/05/

 

System 1

This is the first of many in a long line of 1802 based projects. This is basic design but it does the support an 1861 video chip, cassette I/O. The whole system is built with point to point wiring on Vero board. Unfortunately this system is no longer functional

http://cottageworker.com/wp-content/uploads/2014/05/

Every “real” computer needs a proper front panel

http://cottageworker.com/wp-content/uploads/2014/05/

Side and top view of system. The first card is the CPU card, the second is the interface to the front panel subsystem and the third is a 256 byte memory card. The motherboard also contains a few bits of the circuitry.

http://cottageworker.com/wp-content/uploads/2014/05/

4K byte memory card, the point to point wiring construction is much neater on this one.

http://cottageworker.com/wp-content/uploads/2014/05/

256 byte memory card with a switch to select bus or external power. Note power connector on right side of card. How many times can one person toggle in that monitor program!

http://cottageworker.com/wp-content/uploads/2014/05/

System 5 – Designed to Run ELF/OS

Last Updated Mar: October 31, 2005 11:13:27 AM

As you can see from the older 1802 systems documented on the home page I have been tinkering with the 1802 since the original Popular Electronics ELF articles first appeared. When Mike Riley published ELF/OS I was motivated to dust off my old System-4 and start the long process of updates required to make this home built system compatible with the hardware requirements of ELF/OS. Although I had designed and built System-4, all the work was done between 1979-1982 so some of the documentation was hard to make out if I could find it all. Also all the cards were built with wire wrap techniques, which can be a bit unpleasant to modify. All this lead to the decision to keep the basic bus definition and create System-5, a complete new system using printed circuit boards. This would allow me to update one card at a time mixing old wire wrap cards and the new PCB cards. Another requirement of this approach was that all new cards must be backward compatible with the older System-4. Yes it would have been easier to order a MicroELF from Mike or a STG ELF2000 but I just couldn’t help myself 🙂

The new schematics and PCB layouts as well as this web site finally fix the lack of documentation problem. All the cards are designed using EAGLE CAD software. The new cards liberally borrow ideas from Mike Riley’s MicroElf and PicoElf, Spare Time Gizmo’s ELF2000, the , and several very old BYTE articles. Many of the cards are still under revision, so if you have any thoughts about using these designs and layouts please contact me for the latest status information. I will try and keep the System-5 documentation up to date as the boards are revised. All comments and thoughts are welcome.

http://cottageworker.com/wp-content/uploads/2014/05/

CPU Card: Revision B

Last Updated: October 31, 2005 11:13:27 AM

NO PCB YET

Revision B not produced yet. The bottom connector provides signals to the 1861/keyboard card that are not available on the standard system bus. The upper connector supports Front Panel functionality, again it provides signals not available on the bus. The complete system does not require a front panel to operate. This card replaces System-4 CPU card.

Description

The CPU card has several separate subsystems; clock generation, mode control, power on reset/run, address demultiplexing/decoding , 4K EPROM, and reset address remapping. The CPU card also has auxiliary connectors required for a full function front panel and an 1861 based video/keyboard system. Jumper blocks select -EF line mapping for the bus EF line, front panel EF line, 1861 Video EF and keyboard input EF line. Another set of jumpers selects the location of the EPROM within the address space. The last jumper block (VMA) determines if/how the address is remapped during reset from 0x0000 to the location occupied by the on board EPROM within the address space. See the operational description for details.

Issues/todo log

Date Description Status

Schematics, board layout, and operational description

Revision Production Date Schematic Layout Operational Description
A Jan. 11/05 cpu2-a.sch cpu2-a.brd Rev. A ops notes
B Current Development Board
*UNBUILT/UNTESTED*
cpu2-b.sch cpu2-b.brd Rev. B ops notes

CPU Card: Revision A

Last Updated: October 31, 2005 11:13:27 AM

http://cottageworker.com/wp-content/uploads/2014/05/

Revision A of the CPU card had a few mistakes and required some creative pin lifting, double socketing and rewiring to get it working. The bottom connector provides signals to the 1861/keyboard card that are not available on the standard system bus. The upper connector supports Front Panel functionality, again it provides signals not available on the bus. The complete system does not require a front panel to operate. This card replaces System-4 CPU card.

Description

The CPU card has several separate subsystems; clock generation, mode control, power on reset/run, address demultiplexing/decoding , 4K EPROM, and reset address remapping. The CPU card also has auxiliary connectors required for a full function front panel and an 1861 based video/keyboard system. Jumper blocks select -EF line mapping for the bus EF line, front panel EF line, 1861 Video EF and keyboard input EF line. Another set of jumpers selects the location of the EPROM within the address space. The last jumper block (VMA) determines if/how the address is remapped during reset from 0x0000 to the location occupied by the on board EPROM within the address space. See the operational description for details.

Issues/todo log

Date Description Status
Jan. 2005 SV6 pin 8 should be -reset not -run.

74LS123 should use Q (pin 13), not -Q (pin 4) to IC6
Gate A (PIn 1). 74LS123 CLR (Pin 3) should be tied
high, A (Pin 1) to 1232 RST (Pin 5)

1232 -ST (Pin 7) should be tied to 1802 Clock out (Pin
39) not to TPA.

74LS74 tie PRE (Pin 10) and D (Pin 12) high. CLK (pin
11) should be connected to A15 not GND.

Rewired

Schematic/Layout
Done

Rev. A

Rev. B

Jan. 2005 VMA, Boot/Remap silk screen entry touches pad Done Rev. B
Mar. 8/2005 Add jumper to disable automatic run after power on reset Done Rev. B

Schematics, board layout, and operational description

Revision Production Date Schematic Layout Operational Description
A Jan. 11/05 cpu2-a.sch cpu2-a.brd Rev. A ops notes
B Current Development Board
*UNBUILT/UNTESTED*
cpu2-b.sch cpu2-b.brd Rev. B ops notes

Breadboard Experiment

This is a recreation of the original test I ran sometime in 1977 with my brand new 1802. I wanted to try something that would require a minimum amount of work and see some sort of result.

1) Wire 0x7B by directly connecting D0 through D7 to the appropriate +5V and GND signals
2) Connect an LED and resistor to the Q signal
3) Connected up the power pins
4) using wires connected to the control lines reset the CPU and then set it to run
5) Apply pulse from signal generator (original test used a de-bounced switch) to CLOCK line
After 20 or so clock pulses the LED connected to the Q line lights up

http://cottageworker.com/wp-content/uploads/2014/05/http://cottageworker.com/wp-content/uploads/2014/05/

1802 with Led connected to Q signal and
LED connected to the CLOCK input signal.
Alligator clips are from pulse generator connected to CLOCK.
The Q LED is on the other LED is pulsing.