GB . PTCS . AccelerationFeedforward (r1.1 vs. r1.4)

<<O>> Difference Topic AccelerationFeedforward (r1.4 - 19 Sep 2007 - TimWeadon)


Changed:
< <	Tim Weadon, PE
> >	Tim Weadon, PE

<<O>> Difference Topic AccelerationFeedforward (r1.3 - 24 Aug 2007 - JohnFord)


Changed:
< <	The system must be implemented on top of the friction compensation modification, with no disruption of existing systems. Personnel needed to do the job are J. Brandt, J. Ford, T. Hunter, and T. Weadon. All of these people must be available for the project to be feasible.
> >	The system must be implemented on top of the friction compensation modification, with no disruption of existing systems. Personnel needed to do the job are J. Brandt, J. Ford, T. Hunter, and T. Weadon. All of these people must be available for the project to be feasible. The command stream glitches are fixed by the trajectory improvements
Changed:
< <	K. O?Neil
> >	K. O'Neil
Changed:
< <	Joe Brandt 50% John Ford 10% Todd Hunter 20% Tim Weadon 20%
> >	Joe Brandt John Ford Todd Hunter Tim Weadon

<<O>> Difference Topic AccelerationFeedforward (r1.2 - 18 Jul 2007 - ToddHunter)


Changed:
< <	Todd Hunter, PS
> >	Todd Hunter, PS

<<O>> Difference Topic AccelerationFeedforward (r1.1 - 15 Jul 2007 - JohnFord)

Added:

>
>

%META:TOPICINFO{author="JohnFord" date="1184527424" format="1.0" version="1.1"}% %META:TOPICPARENT{name="ServoImprovements"}%

The GBT servo system suffers from velocity tracking errors at low velocities. One cause of this has been identified as the inability of the lag compensated velocity servo to follow a changing acceleration profile. A cure for this is to inject a feedforward command into the system to provide the signal to accelerate the antenna, and allow the rate loop to close on the rate. This is needed to correctly follow daisy scans and raster scans. In turn, these scans are needed to implement high frequency observing strategies.

Specific objectives and success criteria

The objective is to reduce the peak and RMS tracking error of a daisy scan. The success can be measured using a standard daisy scan, and plotting the position error as a function of time for each axis. Doing this for several values of feedforward gain will show the improvement gained by the modifications.

Primary stakeholders

PTCS Science team
High Frequency observers
Servo system engineers

Key assumptions

The system must be implemented on top of the friction compensation modification, with no disruption of existing systems. Personnel needed to do the job are J. Brandt, J. Ford, T. Hunter, and T. Weadon. All of these people must be available for the project to be feasible.

Project Approvals

Project Team

John Ford, PM
Joe Brandt, PE
Todd Hunter, PS
Tim Weadon, PE

GBT Program Management Authorizations

K. O?Neil
R. Prestage

Project Scope

The project will provide the coefficients needed for acceleration feedforward to be applied to the elevation and azimuth axes. The scope of this project does not include updated or modified servo loop tuning, or architectural changes to the existing servo system.

Project Tasks and Activities

Derive azimuth and elevation Acceleration feedforward coefficients from existing servo data.
Test elevation and azimuth coefficients using half-power tracks and daisy scans
Create documents describing how the modification works and how to adjust it.
Modify startup script to enable and adjust compensation constants.
Release system for production use

Schedule

See the attached servo_imp.pdf?

Personnel Request for this project

Joe Brandt 50%
John Ford 10%
Todd Hunter 20%
Tim Weadon 20%

-- JohnFord - 15 Jul 2007

Topic AccelerationFeedforward . { View | Diffs | r1.4 | > | r1.3 | > | r1.2 | More }

Revision r1.1 - 15 Jul 2007 - 19:23 GMT - JohnFord
Revision r1.4 - 19 Sep 2007 - 14:38 GMT - TimWeadon

Acceleration Feedforward

Project Justification