One encoder on Y axis installed. Y was the most accessible axis and I
already opened up the encoder box on it, so it was an easy choice. I
will from now on test with Y axis, as opposed to X axis. I will try to
make it a one axis CNC machine.

Might want to disconnect the timing belt to keep myself safer.

http://igor.chudov.com/projects/Brid...allation-On-Y/

Anyway, seems like tomorrow will be a great time to hook up that PC
with EMC2 and start working on configs.

i

"Ignoramus7096" wrote in message
...
One encoder on Y axis installed. Y was the most accessible axis and I
already opened up the encoder box on it, so it was an easy choice. I
will from now on test with Y axis, as opposed to X axis. I will try to
make it a one axis CNC machine.

Might want to disconnect the timing belt to keep myself safer.

http://igor.chudov.com/projects/Brid...allation-On-Y/

Anyway, seems like tomorrow will be a great time to hook up that PC
with EMC2 and start working on configs.

i

Looks good. Not sure how you centered the encoder. Ones I've used on
different applications are centred using sleeve with tapered ends. Like
he

http://usdigital.com/assets/general/...structions.pdf

On 2010-07-11, kelly wrote:

"Ignoramus7096" wrote in message
...
One encoder on Y axis installed. Y was the most accessible axis and I
already opened up the encoder box on it, so it was an easy choice. I
will from now on test with Y axis, as opposed to X axis. I will try to
make it a one axis CNC machine.

Might want to disconnect the timing belt to keep myself safer.

http://igor.chudov.com/projects/Brid...allation-On-Y/

Anyway, seems like tomorrow will be a great time to hook up that PC
with EMC2 and start working on configs.

i

Looks good. Not sure how you centered the encoder. Ones I've used on
different applications are centred using sleeve with tapered ends. Like
he

http://usdigital.com/assets/general/...structions.pdf



Kelly, the answer is that I did not use a centering tool, but the hole
diameter in the mounting plate is very close to the shaft diameter. On
a second thought , I will call them for such a tool.

i

Ignoramus7096 wrote:
One encoder on Y axis installed. Y was the most accessible axis and I
already opened up the encoder box on it, so it was an easy choice. I
will from now on test with Y axis, as opposed to X axis. I will try to
make it a one axis CNC machine.

Might want to disconnect the timing belt to keep myself safer.

Well, you can, but don't try to tune the servo response with no load,
you will
end up having to re-do it later. Just start out with the P gain low.
First, with
servo amps off, move the motor by hand and make sure EMC's DRO reading
changes in the right direction. If not, invert the sign of INPUT_SCALE in
the .ini file for that axis. Once you have the direction of motion
right, you
can then enable the servo amps. They may run away if the output direction
is wrong, so be prepared to E-stop. If it does run away, then either
reverse the
motor wires or invert the sign of OUTPUT_SCALE. Now, it should hold
position.
You can start making jog moves to see how well it performs. At this point
you want to read
http://wiki.linuxcnc.org/cgi-bin/emc...rvo_Amplifiers

This was written for the PWM servo system, but almost all of it is
applicable to
the PPMC and velocity servo amps, too. One complication is if the velocity
servo amps are not well-tuned, you have to figure out which bobbles are due
to the servo amp and which are due to EMC's PID tuning. You can contact me
for hints on how to decipher that.


Jon

In article ,
Jon Elson wrote:

Ignoramus7096 wrote:
One encoder on Y axis installed. Y was the most accessible axis and I
already opened up the encoder box on it, so it was an easy choice. I
will from now on test with Y axis, as opposed to X axis. I will try to
make it a one axis CNC machine.

Might want to disconnect the timing belt to keep myself safer.

Well, you can, but don't try to tune the servo response with no load,
you will
end up having to re-do it later. Just start out with the P gain low.
First, with
servo amps off, move the motor by hand and make sure EMC's DRO reading
changes in the right direction. If not, invert the sign of INPUT_SCALE in
the .ini file for that axis. Once you have the direction of motion
right, you
can then enable the servo amps. They may run away if the output direction
is wrong, so be prepared to E-stop. If it does run away, then either
reverse the
motor wires or invert the sign of OUTPUT_SCALE. Now, it should hold
position.
You can start making jog moves to see how well it performs. At this point
you want to read
http://wiki.linuxcnc.org/cgi-bin/emc...rvo_Amplifiers

This was written for the PWM servo system, but almost all of it is
applicable to
the PPMC and velocity servo amps, too. One complication is if the velocity
servo amps are not well-tuned, you have to figure out which bobbles are due
to the servo amp and which are due to EMC's PID tuning. You can contact me
for hints on how to decipher that.

I would add one thing - initially, reduce the power-supply voltage available to
drive the motors, or put a big power resistor in series with the motor, enough
to slow things down and reduce the forces when it gets away.

Joe Gwinn

On 2010-07-11, Jon Elson wrote:
Ignoramus7096 wrote:
One encoder on Y axis installed. Y was the most accessible axis and I
already opened up the encoder box on it, so it was an easy choice. I
will from now on test with Y axis, as opposed to X axis. I will try to
make it a one axis CNC machine.

Might want to disconnect the timing belt to keep myself safer.

Well, you can, but don't try to tune the servo response with no
load, you will end up having to re-do it later.

I agree. I have a pretty big task, which is to connect encoder to your
boards, connect analog output to the drive, make EMC read encoders,
send signals, etc.

That in itself can easily take a week or two.

The first thing I will do is, as you say, make sure that EMC can read
the motor's position.

Just start out with the P gain low. First, with servo amps off,
move the motor by hand and make sure EMC's DRO reading changes in
the right direction. If not, invert the sign of INPUT_SCALE in the
.ini file for that axis. Once you have the direction of motion
right, you can then enable the servo amps. They may run away if the
output direction is wrong, so be prepared to E-stop. If it does run
away, then either reverse the motor wires or invert the sign of
OUTPUT_SCALE. Now, it should hold position. You can start making
jog moves to see how well it performs. At this point you want to
read
http://wiki.linuxcnc.org/cgi-bin/emc...rvo_Amplifiers

This was written for the PWM servo system, but almost all of it is
applicable to
the PPMC and velocity servo amps, too. One complication is if the velocity
servo amps are not well-tuned, you have to figure out which bobbles are due
to the servo amp and which are due to EMC's PID tuning. You can contact me
for hints on how to decipher that.

I will definitely do so once I get there.

i