I have a machine that I'm trying to upgrade from a DOS controller to Mach; the thing is it's already got an early VIA 500Mhz C3 MiniITX board in it -- so what I'd like to do is simply replace the board with a more modern one, and run Mach3 under a stripped-down version of Win2K.

The boards that I was wondering about are these:

Intel D945GCLF
or
Intel D945GCLF2
from http://www.mini-box.com/Atom

I really would like to avoid having to change the board form factor; installing an ATX in there would be quite a chore (everything would have to be disassembled and re-drilled, and the control box is rather heavy and awkward)...

So - anybody have any experience with these? Do they work well with Mach?

I am using a 45CSX  Foxconn board,  XP and 2gig of ram.  Mach3 via LPT.

 http://www.foxconnchannel.com/Product/Motherboards/detail_overview.aspx?ID=en-us0000425

Only running a couple of months but so far it is going sweet.

Greg

Thanks for the reply...


Anybody else?

The boards that I was wondering about are these:

Intel D945GCLF
or
Intel D945GCLF2
from http://www.mini-box.com/Atom

I really would like to avoid having to change the board form factor; installing an ATX in there would be quite a chore (everything would have to be disassembled and re-drilled, and the control box is rather heavy and awkward)...

So - anybody have any experience with these? Do they work well with Mach?

Hello,
Reviving an older thread is not the best way to get an answer, but here goes.

As my ancient dos laptop running turbocnc was falling apart, I needed a new pc for a 3axis mill.
at the same time, I came across mach3.
Mach3 does everything and much more than turbocnc, so that was fine.

the machine proves a greater challenge. to cut the story short, I now have 3 machines, 2 of which don't work , one which does.
the working one is a full desktop but too large and cumbersome for my workshop
the problems with the other always center on the parallel port
1. a fujitsu lifebook with docking station - the X-axis (pin 2 & 3) don't work (or give a wrong voltage?)
it works fine for normal office work with parallel printers/plotters.

2. now to the point : the last one is a Atom based small factor pc with Intel D945GCLF2 motherboard, running XP

Mach3 runs great, the Gecko motors run even smoother than ever before.
all in a cheap, small tidy package, yippee!

the first time ....

after a few test-runs, the X-axis began to falter and then become fully unresponsive.

I replaced the atom pc with another new one: the same thing happened.

The problem is with the parallel port again, not the mill and controller.
the mill works fine with the ancient laptop as with the full desktop.

If anyone has a solution, I'd love to hear it.
glenn

Do you have a breakout board, if so which one, and which gecko are you using?

Hello Britt,

Thanks for your reply.

some clarification:
1. according to my coworker who built the controller/breackoutboard, the breakoutboard is based on 3 gecko 201 controllers.
The brand of the motors is almost certain Alstron IS34023 http://www.alstron.com.sg/Motor_MCG/PDF/IS34023.pdf (reseller had erased brand name)

2. the atompc seems to work fine the first few minutes. then the movement of the X-axis (pin 2&3) begin to falter and finally stops altogether.
after a reboot , it works again for a length of time which seems related to the time the pc was off (cooling down, capacitor discharging?)
Y(pin4 5) and Z (pin6 7) show no problem.
Strangely, and worryingly so, the trouble is with the same pins as on the laptop.

3. I then recabled the pinouts of the Xaxis to pin 8 and 9.  the problem is still there ?!

4. Next attempt will be to use a separate parallel pci card.
This will require some case alteration as the case is ultra low profile.

for now the bulky regular desktop is doing the work.
glenn

Hmm - I think that the drive voltage or current capacity that the parallel port is putting out is just marginal for that gecko... you might try swapping the gecko module for one of the others, and/or a buffered breakout board.

Hmm - I think that the drive voltage or current capacity that the parallel port is putting out is just marginal for that gecko... you might try swapping the gecko module for one of the others, and/or a buffered breakout board.

Many, if not most, parallel ports are incapable of properly driving the opto inputs of a Gecko.  A buffered breakout board is a necessity.

Regards,
Ray L.

Actually the new 203V and the new Servo drive are designed to work right off a 3.3 or 5v new LPT port.   No buffering required.

Dave

the geckos are not directly connected to the parallel port. there is an intermediate interface from Axxus.
Although, looking at the doc, I'm not sure there is much of a separation or buffer between the pc parport and the geckos.

I'm going to do a last test with the pc - I found on the forum that disabling ACPI might help, so I did that, testing tomorrow.

regards
glenn