MC6803 board, with RJ11

[saipan59]

Just finished a new version of a PCB for the MC6803/6801.
A perhaps unique feature is that the serial port goes to an RJ11 (modular telephone) jack. This is because I have a pile of 4-wire phone cables from the days of landline phones, modems, etc. So the RJ11 is a cheap, simple, and effective "interface".
The EEPROM code base includes a bootloader, so that I can burn new code without moving the EEPROM in and out of it's socket.
At the other end of the phone cable is a little dongle with another RJ11 jack, and an FTDI USB-to-serial converter from SparkFun.
I wrote a simple Windows app that takes an Intel HEX file, simplifies the format into the something the bootloader wants, and sends it out a COM port (the FTDI dongle).

Instead of RJ11, one could also use an RJ45 connector, and make use of unwanted Ethernet cables.

BTW, the board has a 68701 installed, because I have several, and they look cooler than regular chips... ;-)

Pete

[DiTBho]

looks great!
are you going to share the design on GitHub? 

[saipan59]

>Looks great! Are you going to share the design on GitHub?

Thanks! I'm happy to share. I've never put anything on Github, but I'll look into it. If it's not "obvious", feel free to give me advice. And PM me if you want the Eagle files directly.

Pete

[pqass]

A PDF of the schematic would be more useful here (vs. Eagle files). 

I'm also curious about your monitor program... what commands do you offer?

In my monitors, I've used a ":" (colon) command that interprets the line as a ihex file line; simplifying the host-side to just piping the ihex file directly to the serial port (thru the "ascii-xfr -s -l 150" line delay program) .

[rhodges]

Let me know if you want some 6803 chips. I have some number of boards with 6803 and EPROM I could donate.

[saipan59]

Let me know if you want some 6803 chips. I have some number of boards with 6803 and EPROM I could donate.

If the 6803's are not soldered in, I'm interested! I would glad pay for shipping, and/or swap you other vintage parts that interest you.

Pete

[saipan59]

A PDF of the schematic would be more useful here (vs. Eagle files). 

I'm also curious about your monitor program... what commands do you offer?
In my monitors, I've used a ":" (colon) command that interprets the line as a ihex file line; simplifying the host-side to just piping the ihex file directly to the serial port (thru the "ascii-xfr -s -l 150" line delay program) .

Two schematic pages attached.
Bootloader format: A Windows app reads an Intel-style HEX file. Each line is stripped down to only the start address followed by exactly 16 data bytes (shorter lines are padded to 16). No length or checksum. To further simplify things in the target Bootloader, the HEX-ASCII "A" thru "F" are changed to ":" thru "?", to make them contiguous with "0" thru "9". This makes the conversion of HEX-ASCII chars to binary nibbles very simple (just subtract 48). Since ":" is now in use, the Windows app prepends "#" as a start-of-line indicator, and appends CR as an EOL. After the loader receives a line (16 bytes), then it writes them (the Windows app does a short delay to allow for the line-write to complete). The bootloader on the target allows ANY address to be updated, EXCEPT the Reset vector at $FFFE.
On reset, the bootloader copies itself to RAM at $2000, then executes from there, so that it is free to write to the EEPROM at $E000. It then waits for a few seconds for any char to arrive on the serial port (indicating a bootload is starting). If nothing arrives, it JMPs to $E000.

Pete

[srb1954]

Two schematic pages attached.

I would consider putting a CPU voltage supervisor chip e.g. MC34064 or MAX809 in place of the RC reset circuit. The 6800 series of micros tends to be a bit fussy about getting a clean reset signal and can lock up if there is a glitch on the RESET input or a glitch on the supply when going through the reset phase. A supervisor chip would ensure that the RESET is not released until a well defined time after the main power rail has stabilised whereas the simple RC reset circuit is not so predictable in its response if the power rail doesn't rise smoothly and quickly.

If you decide to retain the existing RC reset circuit you should at least put a diode in parallel with R1 to ensure that C1 is rapidly discharged when the power goes down. It also ensures that there will be a proper length of RESET pulse following a momentary glitch on the power supply and also protects the micro RESET input pin against possible damage . Lastly, a small value resistor, say 100 , should be put in series with the reset switch to protect its contacts from being damaged from excessive surge currents when it shorts out C1. 

[saipan59]

Two schematic pages attached.

I would consider putting a CPU voltage supervisor chip e.g. MC34064 or MAX809 in place of the RC reset circuit. The 6800 series of micros tends to be a bit fussy about getting a clean reset signal and can lock up if there is a glitch on the RESET input or a glitch on the supply when going through the reset phase. A supervisor chip would ensure that the RESET is not released until a well defined time after the main power rail has stabilised whereas the simple RC reset circuit is not so predictable in its response if the power rail doesn't rise smoothly and quickly.

If you decide to retain the existing RC reset circuit you should at least put a diode in parallel with R1 to ensure that C1 is rapidly discharged when the power goes down. It also ensures that there will be a proper length of RESET pulse following a momentary glitch on the power supply and also protects the micro RESET input pin against possible damage . Lastly, a small value resistor, say 100 , should be put in series with the reset switch to protect its contacts from being damaged from excessive surge currents when it shorts out C1.

Good ideas - thanks! I'll definitely add the diode and series R. I'll pass on the supervisor chip for now - I'm OK with depending on the Reset button. But I should mention it often doesn't self-start if the power has been turned off 'recently' (within several seconds or more). However, a Hitachi HD6303 (CMOS) doesn't have the issue - it seems to start reliably in all cases.
I should also mention that the "power supply" right now is a 5V wall-wart, with a not-so-great connection to the board. So a proper PS might make a difference too.

Pete

[saipan59]

Updated schematic attached. I added D1 and R13 as suggested by srb1954 (thanks!).
Also added the AS signal to J5.7 . I don't need it at the moment, but certain peripherals that could be wired onto the 2nd board might make use of it, such as an MC146818 RTC chip.

With D1 added, it now comes out of reset on power-up in all cases, with any type of chip (6801/6803, or CMOS HD6301).

Pete

[srb1954]

Good ideas - thanks! I'll definitely add the diode and series R. I'll pass on the supervisor chip for now - I'm OK with depending on the Reset button. But I should mention it often doesn't self-start if the power has been turned off 'recently' (within several seconds or more). However, a Hitachi HD6303 (CMOS) doesn't have the issue - it seems to start reliably in all cases.

The HD6303, being a CMOS device, will have an internal ESD protection diode from the RESET pin to the Vcc pin. This diode is acting as a discharge path for the external capacitor when the power supply drops which why you are getting a reliable restart every time with this chip.

However, this internal diode should not be relied on to provide this discharge path as the discharge current may exceed the diode's current rating (probably <20mA) and cause damage in the longer term. The external diode is a safer option to discharge the capacitor.

[rhodges]

Let me know if you want some 6803 chips. I have some number of boards with 6803 and EPROM I could donate.

If the 6803's are not soldered in, I'm interested! I would glad pay for shipping, and/or swap you other vintage parts that interest you.

Hi, Pete. I don't know if you (or anyone else) might be interested in the boards and cases, so here are some pictures. You can have as much or as little as you like.
I forgot how many I have to offer, but at least ten and maybe more than fifty.
The EPROM is an ST 2764 AF, 12.5 volt. Some of the 6803's have a proprietary Motorola number.
The board was made to plug in between a keyboard and PC to feed barcodes in.

[saipan59]

PM sent - thanks very much!

Pete

[Kean]

Hi, Pete. I don't know if you (or anyone else) might be interested in the boards and cases, so here are some pictures. You can have as much or as little as you like.
I forgot how many I have to offer, but at least ten and maybe more than fifty.
The EPROM is an ST 2764 AF, 12.5 volt. Some of the 6803's have a proprietary Motorola number.
The board was made to plug in between a keyboard and PC to feed barcodes in.

Oh wow, I remember those light pen style barcode readers!

I am sure we used those particular ones at my first job, a subsidiary of National Semiconductor (Datachecker/DTS).