Re: spend on DRO, pwr feed or conversion?
Posted by
caudlet
on 2004-11-11 09:50:51 UTC
--- In CAD_CAM_EDM_DRO@yahoogroups.com, "Chris Loiacono"
<chris01@t...> wrote:
consider. The next thing is having a DRO to be able to produce
accurate postioning although it won't help with cutting curves or
irregular objects.
As for converting to CNC here are the challenges:
While you could in theory use the output of the DRO sensors (the
shootingstar DRO uses rotary encoders in the sensor heads) you would
have to build a machine with virtually zero backlash. Why? because
of the nature of servo feedback and it's desire to keep trying to
find the perfect position versus the real position of the feedback
device. What all of that means is that your machine will move to a
position and the feedback has a distinct gap between the encoder
lines and it will adjust one direction then back the other (called
hunting) by a tiny amount. In a system where the sensor is on the
motor itself that just results in a slight hum an imperceptable shaft
movement. On a system where the table postition and the drive
position have slop, the servo loop will drive then reverse over the
span of the backlash (and even overshoot). Users have reported that
the machine will shake itself like a wet dog!
If you want to go to CNC (open or closed loop) you probably want to
consider replacing the leadscrews with precision ballscrews. Low or
zero backlash models are best. Once you do that you then have the
options of adding motors with encoders. If you just want to start
with a DRO then you can still take the encoder into a PC based
software to give you positional information. A that point though all
you need to do real CNC is the Software controller and digital motor
drives.
Now that I have really confused you I will tell you that I went the
descrete route first by putting PM DC motors on my existing
leadscrews with a simple DC speed control and bought one of the
Shooting Star 3 axis DRO's. I bought ballscrews but they are still in
the box. I used my mill to build a lot of the parts for my CNC plasma
and router tables. 90% of what I currently use the machine for is
accurate positioning of holes in a plate or for simple flat milling.
I don't regret the money on the stand alone DRO and when I have
(finally) finished my conversion to full CNC I will just move it to
my non-CNC lathe.
The irony of this is that it would be nice to have a CNC mill to make
the parts to finish the CNC conversion of my mill!!
<chris01@t...> wrote:
>feed or
> I have a 9 x 44 Nantong mill that I have been using with no power
> DRO. Lately it's been getting used every day and I'm tired ofturning
> handles.cost is
> My first instinct was to order an X-axis feed and 3 axis DRO. The
> only about $1,000 and a couple of hours to modify the end of thelead screw.
> My next thought was: "Hmmm... how much more would it cost toconvert this
> mill to CNC?that can
> I haven't a clue. for example, is there a DRO system I can install
> be used now as a DRO, and later as I complete the conversion?I can
> Can anyone offer advice and perhaps ballpark cost numbers for me so
> make a decision?Having some form of power feed on a mill is the first thing to
>
> Thanks In advance!
consider. The next thing is having a DRO to be able to produce
accurate postioning although it won't help with cutting curves or
irregular objects.
As for converting to CNC here are the challenges:
While you could in theory use the output of the DRO sensors (the
shootingstar DRO uses rotary encoders in the sensor heads) you would
have to build a machine with virtually zero backlash. Why? because
of the nature of servo feedback and it's desire to keep trying to
find the perfect position versus the real position of the feedback
device. What all of that means is that your machine will move to a
position and the feedback has a distinct gap between the encoder
lines and it will adjust one direction then back the other (called
hunting) by a tiny amount. In a system where the sensor is on the
motor itself that just results in a slight hum an imperceptable shaft
movement. On a system where the table postition and the drive
position have slop, the servo loop will drive then reverse over the
span of the backlash (and even overshoot). Users have reported that
the machine will shake itself like a wet dog!
If you want to go to CNC (open or closed loop) you probably want to
consider replacing the leadscrews with precision ballscrews. Low or
zero backlash models are best. Once you do that you then have the
options of adding motors with encoders. If you just want to start
with a DRO then you can still take the encoder into a PC based
software to give you positional information. A that point though all
you need to do real CNC is the Software controller and digital motor
drives.
Now that I have really confused you I will tell you that I went the
descrete route first by putting PM DC motors on my existing
leadscrews with a simple DC speed control and bought one of the
Shooting Star 3 axis DRO's. I bought ballscrews but they are still in
the box. I used my mill to build a lot of the parts for my CNC plasma
and router tables. 90% of what I currently use the machine for is
accurate positioning of holes in a plate or for simple flat milling.
I don't regret the money on the stand alone DRO and when I have
(finally) finished my conversion to full CNC I will just move it to
my non-CNC lathe.
The irony of this is that it would be nice to have a CNC mill to make
the parts to finish the CNC conversion of my mill!!
Discussion Thread
Chris Loiacono
2004-11-11 08:52:18 UTC
spend on DRO, pwr feed or conversion?
caudlet
2004-11-11 09:50:51 UTC
Re: spend on DRO, pwr feed or conversion?
Chris Loiacono
2004-11-11 11:56:26 UTC
RE: [CAD_CAM_EDM_DRO] Re: spend on DRO, pwr feed or conversion?
R Rogers
2004-11-11 12:12:58 UTC
Re: [CAD_CAM_EDM_DRO] spend on DRO, pwr feed or conversion?