Motion Control and Motor Drives

cancel
Showing results for 
Search instead for 
Did you mean: 

Sinusoidal Velocity Profile

So, are you saying that I can use an analog signal connected to the ADC, as a master to be tracked, but I can't insert a digital signal, behind that ADC?
0 Kudos
Message 11 of 19
(2,521 Views)
That's correct. The reason for this is the fact that you can't communicate directly with the DSP that runs the PID loop but the DSP has direct access to the ADCs.

Jochen
Message 12 of 19
(2,520 Views)
Dear Jochen,
 
Thanks very much for the suggestion to use electronic gearing.  It is somewhat disappointing that we can't use a 150 Hz digital sinusoid, but going through the ADC should actually work out fine for my project, since I already have an analog output in the system.  The Slave Axis Gearing example works pretty well.  Just one problem, though.  While its position tracks a nice looking sinusoid, at 1 Hz, for example, I can hear the motor buzzing.  It doesn't do that when I'm running a circular arc or a velocity profile.  There is audible noise in the system somewhere.  Is it from the ADC quantization?  Have you heard of this problem before?  My step response is beautiful.  There is no overshoot; it just rises and settles to the target position in about 50 milliseconds, so it doesn't seem like a tuning problem.
 
Josh
 
0 Kudos
Message 13 of 19
(2,509 Views)
Josh,

yes, this is one of the drawbacks of analog feedback. Quantization and in general any type of noise on the analog line have an impact on the axis. In gearing mode this impact will be more significant the higher the gearing ratio is as this ratio amplifies also the noise. The better your signal uses the full range of the ADC the lower you can set the gear ratio (e. g. if your signal uses the full scale range of +/- 10 V and you need only +/- 5 V for your move, you could set a gear ratio of 2:1 as this will reduce the noise by the factor 2, too. In contrast if your analog signal is just +/- 1 V but you need +/- 10 V for your move you could set the ratio to 1:10 but this will also amplify the noise by the factor of 10).

If it's not an option to play around with the analog levels you could try to tune your motor for slower response which should reduce the buzzing but it will also reduce the quality of your trajectories...



Message 14 of 19
(2,503 Views)

I see!  Thank you very much!

Josh

0 Kudos
Message 15 of 19
(2,495 Views)
hi

How about using the sinewave velocity vi http://zone.ni.com/devzone/cda/epd/p/id/1438. Can this help me generate 10Hz frequency.
I too faced a similar problem.

Thanks
Regards
Sneh@l
0 Kudos
Message 16 of 19
(2,065 Views)

Hi 2photonGuy,

Have you read through the entire thread?  Did you try Jochen’s suggestions about sending an analog signal?  Did this give you problem?

Adam H
National Instruments
Applications Engineer
0 Kudos
Message 17 of 19
(2,043 Views)
I will be tryying it soon...

Thanks
Regards
Sneh@l
0 Kudos
Message 18 of 19
(2,037 Views)
Sounds good!!!  Let me know how it goes.
Adam H
National Instruments
Applications Engineer
0 Kudos
Message 19 of 19
(2,012 Views)