[Noisebridge-discuss] getting that goddamned mill to work

[Michael Wright]

EMC2 was just really really nice running on the 2.8GHZ box I borrowed  
from work, whereas it was it was a bit weird and slow on the compaq.   
I noticed the same running in VMWare on my macbook and netbook.  It  
runs with less that 1 ghz cpu, but it's bumpy and annoying.  If it  
runs great on a $80 celeron then that just leaves hacking together a  
functional HAL file.  If we get bored we can always hack it to be  
controlled with a gamepad.

Basically I'm betting that it will be $80 more fun to have a crisp UI  
for hacking on the mill.  From what I can tell, getting the computer  
able to move the mill is only the first step in being able to make  
parts, with the other steps being larger.

mike


On Oct 5, 2009, at 11:38 AM, Dr. Jesus wrote:

> On Mon, Oct 5, 2009 at 11:26 AM, Michael Wright <mike at smallip.com>  
> wrote:
>> Yep, it cycle one way when going forward, then the other when going
>> backward.  It also maintains it's pin state indefinitely when not
>> moving.  It turned out that all the original (down to 0.001") jog
>> sizes were a factor of four, so it would do a full encoder cycle,
>> ending with the original state.
>>
>> Halscope and halmeter are somewhat clunky, but really useful for
>> seeing what's going on.
>>
>> Though we don't actually need the timing to be perfect (because we're
>> doing quadrature rather than pulses) EMC2 will still throw an  
>> annoying
>> error any time the latency crosses the set threshold (max of about
>> double the default),   It looks to me like we've got a path to  
>> success
>> with EMC2 so if the box at Weird Stuff (1.8 ghz celeron) passes the
>> latency test I'll just give them $80 for it so we can move on to the
>> next problem (how to make parts via gcode).
>
> I take it that none of the systems in the space satisfied your
> parallel port requirements, then?
> _______________________________________________
> Noisebridge-discuss mailing list
> Noisebridge-discuss at lists.noisebridge.net
> https://www.noisebridge.net/mailman/listinfo/noisebridge-discuss