|Ian Hodgson 296 Malcolm Circle Dorval, Quebec, Canada H9S 1T7||
All features are optional. The
manual includes itemized parts,
lists for each option, enabling
you to build only the sections
I chose everything except the power supply and the dual cas- sette port, so the first job was to look for the necessary parts. Al-
|though the manual does men- tion parts substitutions, they donít cover the UART, a Western Digital TR1602B. I have used both an AV-5-1013A and an AY-3- 1014A with fine results. One word of warning: If you haven't bought any 74LS ICís lately, be prepared for a shock when you see the current prices. I kept the bill to less than $280, including memory, so the entire expan- sion was built for about $370 Canadian Ė a lot less than $1042 for the RS interface. nbsp;The RS232 port is well provid- ed with handshaking, and Re- quest to Send (RTS), Clear to Send (CTS), Data Terminal Ready (DTR), Data Set Ready (DSR), Carrier Detect (CD), and Ring indicator (RI) are all imple- mented, as well as provision for generatina a true BREAK signal. In addition, the serial data ap- pears at one bit (IN 0EBH, bit 1) and the port may be wired for 20mA operation if desired. Al- though the TRS-80 may be set up either as a data terminal (DTE) or data communication device (DCE), the handshaking lines are fully implemented only for DTE. There is no provision for|
While the Level II, 16K
TRS-80 is a lot of machine
for not much money, it becomes
evident that for serious work
more memory and disk drives
The first equipment I investi- gated was the RS expansion in- terface. At $837 for the interface with 32K RAM, $160 for the RS232 board, and $45 for the communications software it looked like a grand total of $1042 just to connect the drives and add a bit of memory. There had to be a cheaper way. I then saw the ad from LNW Research for their TRS-80 ex- pansion board. I ordered one for $59.95: It is a 9 1/2 x 11 1/2 inch, double-sided, soldermasked, glass epoxy board.
The board has provisions for:
Photo 1. The expansion board installed behind the TRS-80. At left, from front to back, are the dish drive connector, parallel printer port, and keyboard connector. At rear, left, is the expansion connector, and at right, the RS232 connector. The two 40-pin chips are the UART and the disk controller. The vacant area in front is for the power supply.
230 - 80 Microcomputing, May 1981
the UART to generate interrupts, nor is the interrupt line easily available to those who have time-critical applications. Inter- rupts are definitely usable on the TRS-80 (from your own machine language programs), since the vectors are stored in RAM. You must be careful, though, because many of the ROM routines depend on the in- terrupt vectors, and you may have to write your own keyboard and screen drivers depending on what interrupt mode you use. Although a power supply is built in on the board (and it is quite a sophisticated one, too, using an additional Radio Shack TRS-80 transformer) l used one I already had, to save money and because my RS store refused to order the transformer for me. The completed board works with an external supply. It re- quires + 5 volts at about 1A, + 12V at a few mA, and - 12V also at a few mA. The manual is well written, and details are given for con- struction, configuring the sys-
tern and theory of operation.
Data sheets for the UART and
the disk controller (1771) are in-
cluded with software for using
the 80 as a terminal. Software is
also provided to drive a serial
printer through the RS232 port,
or through the parallel port, with
certain board modifications. I
only found one error in the chart
of UART control addresses; the
bits for Request to Send and for
Data Terminal Ready are inter-
Construction went fairly well, but there were a few bugs that took several hours to find. Amazingly, these were all traced to a batch of new 100 ohm resis- tors, of which three were open.
One feature I didnít like was the vertically-mounted terminat- ing resistors at the expansion bus.
Finally, everything worked, in- cluding the TRSDOS 2.3 mem- ory test (TEST1) and stress test (TEST2).
Words of Warning
A few words of warning are in
|order. The bus expansion pro- vided should be used with due caution. It is a simple, unbuf- fered extension of the TRS bus, and any expansion will require more buffering with high imped- ance input buffers (not 7400 ICís). Remove the terminating resistors from the LNW board and terminate at the end of your expansion; as well. There are two edge connectors on the board. Either should work with the keyboard, and the other may be used as the bus expansion. However, when I used the con- nector next to the terminating resistors for the keyboard, I had unusual problems like random rebooting, and Incorrect opera- tion of a few programs. Using the other connector cured the problem. I also suggest that you use the shortest cable possible. Incidentally, the cable does not come with the board, but you can make one with 40-conductor ribbon cable. The identical con- nectors on each end are 3M #3464-000. They are Installed us- ing a small vice.||
Using disks, machine lan-
guage programs can be patched
to correct errors, and reassem-
bled with ease. Data files are
childís play, and of course, long
programs load in seconds. My
system has a KIM-1 converting
from parallel to serial on the
parallel printer port allowing me
to use the KIM to format my out-
put (paging, line length, mar-
gins, etc.). What is more, it has
4K of memory available to act as
a, buffer. When I LLIST a pro-
gram less than 4K long, the
READY returns in about one
I am using a Terminet 300 printer (upper and lowercase) at 300 baud. I use the RS232 port for packet radio transmission : and reception at 2400 baud.
To summarize: the LNW Re- search board is well laid out and inexpensive. The documenta- tion is satisfactory. With disk drives, 48K of memory and Elec- tric Pencil with lowercase modi- fication, it makes a good ma- chine.
232 - 80 Microcomputing, May 1981