User:YOPERO

From London Hackspace Wiki

@yopero


Hacking Interfacing the Boxford 190VMC

15/09/2014

Long ago this CNC was controlled by a BBC Master from Acorn.

I just found a disk with the software used to control the Boxford 125 tcl lathe that happens to have a very similar electronic board as the 190VMC. To emulate the BBC master on a Windows7 sytem I have used "BeebEm 414"--->http://www.mkw.me.uk/beebem/ and the disk can be downloaded from here: https://groups.google.com/group/london-hack-space/attach/8f627b29677a61fb/Boxford200K.ssd?part=0.1&authuser=0&view=1

It could be a starting point for decoding the initial handshake to control the drivers. The documentation suggests that it uses standard gcode (ISO)---> http://tturkyilmaz.home.anadolu.edu.tr/dersler/telatturkyilmaz/programlar/cnc/millcamakz/millcam.pdf

Yesterday I was able to connect to the cnc via serial(b:9600 d:8 s:1 p:none) and get its model but could find any code that actually moves the motors.

18/09/2014

DataSheets for the existig electronics

  • DS80C310 High-Speed Microcontroller, driving each motor:

https://drive.google.com/file/d/0BwcYiBsr3w5_NWk3LTRhMXI5bUE/edit?usp=sharing

  • Lense Speed controllers 530 series Spindle:

https://drive.google.com/file/d/0BwcYiBsr3w5_bzZWSzBPQU5LbjQ/edit?usp=sharing

  • GS-D200S 2/2.5A BIPOLAR STEPPER MOTOR DRIVE MODULES

https://drive.google.com/file/d/0BwcYiBsr3w5_Yjh6UWN2NklVMmc/edit?usp=sharing

  • microstepping

https://drive.google.com/file/d/0BwcYiBsr3w5_VURPR1Z1cFJlOTA/edit?usp=sharing

http://www.nmbtc.com/step-motors/engineering/full-half-and-microstepping/

  • Serial cabling db25 and db9

http://www.lammertbies.nl/comm/cable/RS-232.html https://drive.google.com/file/d/0BwcYiBsr3w5_MlIyRWlRSFA5ejA/edit?usp=sharing http://www.lammertbies.nl/comm/cable/RS-232.html

[[parallel_port_pinout.jpg ]]

The cnc sends 5 blocks of 7 chars, the structure as follows:

P000000// Z axis
p000000// feed rate
 s 0000// spindle rpm
00000E5// X axis
@000000// Y axis

Null modem cable with loop back handshaking does not make any difference when sending regular GCode

20/09/2014

I was able to make the cnc move by sending 7 nibbles(each one with #A) and then gcode although it is not really stable.

eg:

line00: AA AA AA A \CR (hex)                    1010 1010 1010 1010 1010 1010 1010
line01: G00 X05 \CR (ASCII)

line00 needs further investigation as each bit should be enabling something on the uC

must be worth it to read from the eeprom

EEPROM https://drive.google.com/file/d/0BwcYiBsr3w5_TW03QkkwMWp6bGM/edit?usp=sharing

21/09/2014


FIRMWARE----------

Software controller sends "=" (asci) 0x3D (hex)before sending the actual gcode and then it waits for the CNC to:

  • Reset
  • Cancel Manual mode
  • Press both + and - axis buttons for each axis.

CNC sends this string on start-up:

�(c)2000 Rod Hine MT190HI,v019,MH,1234,25-09-2000 09:47:13
(c)1996 Rod Hine auto4t 1/3/96
ms190.asm    ml190ph.asm

SOFTWARE-------

I got the software working on Windows7.

Boxford software version 4.4.2 with DOSBOX x86 emulator(http://www.dosbox.com/). I have tested the port connexion only available on COM1 and COM2, for this I have created a virtual port with vspdxp(http://www.eltima.com/products/vspdxp/). I believe it will work with the real machine but I have not tested it yet.


24/09/2014

Eugene has got his hands on a Boxford190VMC_programming_manual---> https://www.dropbox.com/s/uphbkib56eaziz4/Boxford190VMC_programming_manual.pdf?dl=0 and it moves!!