What are the units for velocity?

Motion Scope Tool
1

Suppose I capture something like “Axis 0 Command Velocity.”
What units should I expect the captured values to be in?

  • counts/sec
  • counts/millisecond (i.e. counts/sample)
  • something else?
2

Also, what units should I expect for acceleration? counts/sample/second?

3

Looks like velocity is going to be counts/second.

velocity_counts_per_second
velocity_counts_per_second1828×943 16.2 KB

And acceleration should be counts/second/second

You can find the units of your trace if you “edit” a selected trace (see below).

accel_trace
accel_trace1209×582 16 KB

vel_trace
vel_trace1209×587 16.1 KB

1 Like
4

You caught me asking a Scope question when I guess I was asking a firmware question.

I’m capturing acceleration using recorders, and the raw numbers I get look different than what the scope captures (and saves in its .pan data file).
If I copy the firmware address of CommandAcceleration to a new trace and don’t set the units, here’s what I see:

image
image1402×866 69.5 KB

The data from the “unitless” trace is literally a million times smaller than “units/sec/sec”.
This seems like a queer scale factor, unless the raw data is actually counts/sample/sample.

Is acceleration (command acceleration, anyway) represented in the firmware as “per sample per sample?”

5

I tried using the recorder with a new phantom axis, and it looks like I got my velocity in terms of user units.
(I set user units to 1000, command a velocity move of 10, set my recorder to record RSIAxisAddressTypeCOMMAND_VELOCITY, and once the axis reaches speed the retrieved double value is 10)

From what I recall from some past internal discussions is that the Command Acceleration is generally not used or computed unless you are actually sending sending some form of command accelerations. Is this the case?
In with my same test setup as before, but recording RSIAxisAddressTypeCOMMAND_ACCELERATION instead, my command acceleration stays at a constant 0.

Perhaps you can compute your accelerations from your velocity commands, or record your actual acceleration. I need to confirm the command/actual velocity behavior, however.

6

I’m only paying attention to the acceleration when there’s some motion commanded, so that’s when I expect to see something. I am using the address that corresponds to RSIAxisAddressTypeCOMMAND_ACCELERATION. In the most recent plot above, I was using whatever address the scope uses when you select Command Acceleration. When I home (the operationg I’m testing), I can see meaningful values for these addresses. The units are just not what I’d expect.

7

RSIAxisAddressTypeCOMMAND_ACCELERATION should be in units of counts/sample/sample. Does that help reconcile your values?

1 Like