we managed to get a few of these compact 5 pc from a local school. As most of you I started searching the internet for info on the lathe. I found this artical

http://www.maxton.com/ebay/emco/EMCO...to%20Mach3.pdf

from the digital machinist mag that explained how to remove the octal latch from the interface board, switching the board to step/dir and setting up mach. I did a search for the 74ls374 chip discussed in the artical. It looked like I could use linuxcnc to send the clock signal to the board. So after some figuring and searching I found figured out the printer port pins..

x step pin 2
x dir pin 3
z step pin 4
z dir pin 5
index pin 12
estop pin 11
100 ppr sensor pin 10
74ls374 enable pin 14
74ls374 clock pin 1

So I hacked together a hal file that used the 'reset' function of the printer port driver to send a clock every base period to latch the 74ls374 chip. After not remembering to set switch 1 to on on the emco interface board (sets the interface to step/dir) and scratching my I got linuxcnc communicating. The performance was not stellar. The article above talks about setting your max velocity to 19ipm. This is what I was seeing. Just pretty poor performance. Sounded crappy. Now it could be that I didn't have the timing perfect but read on.. (according to the article above - the original software did around 30ipm)

When I forgot to set switch 1 on the emco interface board - I had decided (by twiddling bits on the printer port with linuxcnc) that the default setup must have been phase drive. The original software controlled the 4 phases of the stepper drives/motors. I didn't have much to lose so I modified the linuxcnc hal file to do 4 phase drive for each axis and set switch on on the emco board back to 'off'.

HAL Component Descriptions

So pins 2-5 ran one stepper and pins 6-9 ran the other.

the stepgen in linuxcnc will do a bunch of wave forms - I was interested in - 5 through 10. (4 phase patterns)

I started with pattern 9. This is Unipolar Half Step
-This acted the same as the step/dir I started with. 19ipm max and sounded crappy.

Next I tried pattern 10. This was a long shot as it is Bipolar Half Step
-I don't think it worked at all - or at least it was way worse..

Then I tried pattern 5. Unipolar Full Step (one winding on)
-I had to half the steps per inch obviously - but it acted the same as the above. around 19ipm and sounded bad.

finally I tried patten 6. Unipolar Full Step (two windings on)
-Wow. This sounded nice. We started pushing it and it started sounding rough at around 40ipm. (didn't seem to be losing steps but I think I might be just fine with 35ipm max)
We still have to up the acceleration as that could be pushed a bit also. this is around .0005 per step - very usable.

Very happy.

Now we need to get the index and 100ppl setup in linuxcnc and I will take some videos (threading and such). with this performance initially - I don't know if I will change the drives/steppers until they fail...

one last thing I would like to try is putting the interface back into step/dir mode and from what I have read - you can put it into full step mode.

So - Again - this is without modifying the emco electronics in any way. Setting up linuxcnc to output the 'clock' to latch the chip.

when I get the encoder setup within linuxcnc - I will post the hal/ini files here.

sam