RE: [CAD_CAM_EDM_DRO] USB options ...
on 2014-06-18 11:31:40 UTC
My take on USB: It is a fine protocol foe memory sticks, mouse and keyboards. It has major shortcomings as a viable CNC interface.
1. It is a polled protocal. That means the Master (PC) has to ASK each device if it has any traffic then setup a "slot" and give that device the channel to talk. Once it all gets started it is pretty fast but the setup time creates bottlenecks in something that needs continuous communication
2. You can shield all you want to but the signal is NOT differential or isolated so the spec of 15 ft has to be reduced in really noisy environments. Try one of these things in a shop with plasma cutters, arc and TIG welders and see what happens.
As a long time developer of software/firmware for CNC (I started using MACH when it was Master 5 and ran on WIN98) we have looked at and tried a lot of the solutions. We even use USB ourselves but only for non-critical communications and we convert it to an industry accepted noise immune protocol called RS485. It can be used in high noise environments and over long distances (hundreds of feet) The RS485 is used for "backside" communications and doing things like setup and troubleshooting.
Ethernet was designed for and is best suited for high speed long distance (100 meters) bidirectional communications it has total ground isolation (where most of the noise comes from) and very established protocols. Any device can use the channel and initiate a transmission.
I contend that you won't find critical commercial systems running USB. If you are a DIY hobby guy and don't mind living with an occasional disconnect or "glitch" then USB might be acceptable.
As you may or may not know, when you move to ANY external pulse card with MACH you become "disconnected" from real time motion so it rules out any feedback type operations like THC to MACH
We recently introduced a new ultra high speed DTHC IV product that no longer uses the MACH THC logic and inputs. Shortly thereafter we announced our Ether-Cut series that uses Ethernet, runs MACH in WIN7 32/64 and has expanded I/O and improves motion versus a parallel port. We still sell parallel port solutions but virtually all our new sales are the Ether-Cut products. Just the fact the choice of PC platform is not as narrow and (just for fun) we ran a plasma table on a laptop with WIN 7 and surfed the net. Not something I advise but it proved a point.
As for wireless....I can't make most of my wireless stuff work 100% of the time so I don't trust it (as an Engineer) .
Just my views. Others will differ I am sure
TOMCaudle
http://www.Candcnc.com
1. It is a polled protocal. That means the Master (PC) has to ASK each device if it has any traffic then setup a "slot" and give that device the channel to talk. Once it all gets started it is pretty fast but the setup time creates bottlenecks in something that needs continuous communication
2. You can shield all you want to but the signal is NOT differential or isolated so the spec of 15 ft has to be reduced in really noisy environments. Try one of these things in a shop with plasma cutters, arc and TIG welders and see what happens.
As a long time developer of software/firmware for CNC (I started using MACH when it was Master 5 and ran on WIN98) we have looked at and tried a lot of the solutions. We even use USB ourselves but only for non-critical communications and we convert it to an industry accepted noise immune protocol called RS485. It can be used in high noise environments and over long distances (hundreds of feet) The RS485 is used for "backside" communications and doing things like setup and troubleshooting.
Ethernet was designed for and is best suited for high speed long distance (100 meters) bidirectional communications it has total ground isolation (where most of the noise comes from) and very established protocols. Any device can use the channel and initiate a transmission.
I contend that you won't find critical commercial systems running USB. If you are a DIY hobby guy and don't mind living with an occasional disconnect or "glitch" then USB might be acceptable.
As you may or may not know, when you move to ANY external pulse card with MACH you become "disconnected" from real time motion so it rules out any feedback type operations like THC to MACH
We recently introduced a new ultra high speed DTHC IV product that no longer uses the MACH THC logic and inputs. Shortly thereafter we announced our Ether-Cut series that uses Ethernet, runs MACH in WIN7 32/64 and has expanded I/O and improves motion versus a parallel port. We still sell parallel port solutions but virtually all our new sales are the Ether-Cut products. Just the fact the choice of PC platform is not as narrow and (just for fun) we ran a plasma table on a laptop with WIN 7 and surfed the net. Not something I advise but it proved a point.
As for wireless....I can't make most of my wireless stuff work 100% of the time so I don't trust it (as an Engineer) .
Just my views. Others will differ I am sure
TOMCaudle
http://www.Candcnc.com
Discussion Thread
Lester Caine
2014-06-18 03:00:58 UTC
USB options ...
Vince Negrete
2014-06-18 07:20:07 UTC
Re: [CAD_CAM_EDM_DRO] USB options ...
John Dammeyer
2014-06-18 09:09:59 UTC
RE: [CAD_CAM_EDM_DRO] USB options ...
Andy Wander
2014-06-18 09:13:40 UTC
RE: [CAD_CAM_EDM_DRO] USB options ...
Benny Croonen
2014-06-18 10:20:56 UTC
Re: [CAD_CAM_EDM_DRO] USB options ...
John Dammeyer
2014-06-18 10:38:58 UTC
RE: [CAD_CAM_EDM_DRO] USB options ...
Lester Caine
2014-06-18 11:31:14 UTC
Re: [CAD_CAM_EDM_DRO] USB options ...
2014-06-18 11:31:40 UTC
RE: [CAD_CAM_EDM_DRO] USB options ...
Lester Caine
2014-06-18 11:45:04 UTC
Re: [CAD_CAM_EDM_DRO] USB options ...
Lester Caine
2014-06-18 12:02:16 UTC
Re: [CAD_CAM_EDM_DRO] USB options ...
Benny Croonen
2014-06-18 15:41:49 UTC
Re: [CAD_CAM_EDM_DRO] USB options ...
Malcolm Parker-Lisberg
2014-06-18 15:46:07 UTC
Re: [CAD_CAM_EDM_DRO] USB options ...
2014-06-18 15:52:46 UTC
Re: [CAD_CAM_EDM_DRO] USB options ...
2014-06-18 16:32:49 UTC
Re: USB options ...
Peter Homann
2014-06-18 16:52:22 UTC
Re: [CAD_CAM_EDM_DRO] Re: USB options ...
Stan Stocker
2014-06-19 03:44:37 UTC
Re: [CAD_CAM_EDM_DRO] USB options ...
Lester Caine
2014-06-19 04:17:39 UTC
Re: [CAD_CAM_EDM_DRO] USB options ...
Jon Elson
2014-06-19 08:44:47 UTC
Re: [CAD_CAM_EDM_DRO] USB options ...
Hannu Venermo
2014-06-19 08:55:55 UTC
Re: [CAD_CAM_EDM_DRO] USB options ...
Jon Elson
2014-06-19 09:10:50 UTC
Re: [CAD_CAM_EDM_DRO] USB options ...
2014-06-19 13:44:51 UTC
Re: USB options ...
2014-06-19 14:14:24 UTC
RE: [CAD_CAM_EDM_DRO] USB options ...
John Dammeyer
2014-06-19 23:07:32 UTC
RE: [CAD_CAM_EDM_DRO] USB options ...
Steve Blackmore
2014-06-19 23:57:13 UTC
Re: [CAD_CAM_EDM_DRO] USB options ...
Lester Caine
2014-06-20 02:01:29 UTC
Re: [CAD_CAM_EDM_DRO] USB options ...
Hannu Venermo
2014-06-20 02:11:26 UTC
Re: [CAD_CAM_EDM_DRO] USB options ...
Andy Wander
2014-06-20 07:59:56 UTC
RE: [CAD_CAM_EDM_DRO] USB options ...
Hannu Venermo
2014-06-20 08:19:33 UTC
Re: [CAD_CAM_EDM_DRO] USB options ...
John Dammeyer
2014-06-20 08:48:35 UTC
RE: [CAD_CAM_EDM_DRO] USB options ...
Jon Elson
2014-06-20 09:01:16 UTC
Re: [CAD_CAM_EDM_DRO] USB options ...
John Dammeyer
2014-06-20 09:18:39 UTC
RE: [CAD_CAM_EDM_DRO] USB options ...
Malcolm Parker-Lisberg
2014-06-20 09:48:50 UTC
RE: [CAD_CAM_EDM_DRO] USB options ...
Peter Homann
2014-06-20 16:24:24 UTC
Re: [CAD_CAM_EDM_DRO] USB options ...
Hannu Venermo
2014-06-20 23:22:48 UTC
Re: [CAD_CAM_EDM_DRO] USB options ...
John Dammeyer
2014-06-20 23:51:30 UTC
RE: [CAD_CAM_EDM_DRO] USB options ...
Lester Caine
2014-06-21 00:19:03 UTC
Re: [CAD_CAM_EDM_DRO] USB options ...
Alan Stanley
2014-06-21 09:32:55 UTC
Re: [CAD_CAM_EDM_DRO] USB options ...
fortino tan
2014-06-23 08:49:24 UTC
Re: [CAD_CAM_EDM_DRO] USB options ...
Malcolm Parker-Lisberg
2014-06-23 08:49:58 UTC
Re: [CAD_CAM_EDM_DRO] USB options ...
Vince Negrete
2014-06-23 09:13:15 UTC
Re: [CAD_CAM_EDM_DRO] USB options ...
Andy Wander
2014-06-23 09:52:57 UTC
RE: [CAD_CAM_EDM_DRO] USB options ...