Optimizing Servo Auto Tuning
Optimizing Servo Auto Tuning Tool
Based on the step response, the rise time given by the BA is 0.1 seconds and the settling time of 1.41 seconds with an overshoot of 50.8%. The system does converge to 1. 4.2.3 Case Study 3: Servo Motor Model This case study is to observe the algorithm behavior for tuning the controller for a servo motor model given a step input. Auto-tuning function then weeds out unwanted vibration and resonances—enabling smooth, fast and reliable position and speed control. Again, key to the MR-J4 is its holistic approach to servo axis tuning: One-Touch Tuning automatically and quickly sets control parameters at the push of a button. Auto-tuning then analyzes the machine. SVX Servo Suite provides a comprehensive and powerful software environment for setting up SV200 servo drives. All auto-tuning, parameter optimization, configuration and troubleshooting functions can be carried out within this software.
Optimizing Servo Auto Tuning Kit
Optimizing Servo Auto Tuning Parts
I've been looking for an inexpensive antenna tuner with a roller inductorfor a while -- 'inexpensive' is why I've been looking for a while. Not surewhy -- I have one of those LDG auto tuners that works just fine. So, whena tuner with a roller inductor finally came my way, I figured out what Iwas going to do with it -- make it an auto-tuner. In April/May/June 2002 issues of QST, W8ZR presented his EZ-tuner project. This is a T-network tuner controlled by BASIC Stamp (remember those?). According tothe project site, this is more of a memory tuner -- all matches need tobe set ahead of time and will be selected based upon recalled frequency.It also looks like the inductor is fixed with taps. My goal is to use a stepper motor for each of the capacitors and the rollerinductor controlled by an Arduino Mega. Match will be determined by sensingforward and reverse power and optimizing by some algorithm TBD. I would alsolike to integrate two radio inputs, three antenna output and bypass/throughfunctionality. Starting materials The SPC tuner design has an output capacitor in parallel and in series withthe inductor. I guess this gives some harmonic attenuation and wider tuningrange for the cost of twitchier tuning and higher loss. That all remains tobe seen, I guess. Proposed layout The front of the tuner will have an aluminum panel on which the NEMA23stepper motors are mounted. These will be coupled to the components viawhat I currently suspect is going to be fiberglass rod. The right side ofthe case will have switching, sensing and electronics. The components are mounted off the chassis with some plexiglass spacerswhich should be adequate for the 100W I anticipate. I came across an interesting find -- the RAMPS 1.4 3D printer controllerintegrates a display, stepper drivers, sensor switches and a bunch of otherstuff along with an Arduino Mega into an inexpensive and small package. I'll need all of the stame stuff for the tuner, that will probably be theelectronics package I use. Somewhere in there I need to do some limit switches to sense the end travelof the roller inductor and some arbitrary spot on the capacitors. How to auto tune in logic. For antenna switching I'll probably go with regular DPDT relays, modifiedslightly for use at HF frequencies -- which is really just DC with a bit ofripple on it if you ask the microwave guys. http://www.digikey.com/product-detail/en/panasonic-electric-works/HJ2-DC12V/255-1678-ND/570815 More to come. Sun Apr 30 22:43:48 CDT 2017It's a cold rainy spring day so I don't feel bad about spending time indoors onthis project. I think writing my own menuing system is a waste of my time --I would much rather be writing control and tuning code, making steppers andlimit switches work, rather than trying to go through cycle after cycle of making my menu code pretty and trying to figure out why returning from onemenu branch isn't like another. I did some looking around and found LCDMenuLib -- which I have working with my u8g2 LCD display with the rotary encoder and 'stop' button to go back. I'm hoping using this code will get me goingmore quickly on more interesting control problems. Additionally since my last update I found a cheap 4U enclosure -- which while a bitlarge -- seems to fit my components much more easily. The enclosure design allowsthe bottom of the chassis to be re-positioned about 1/3 of the way up -- which willallow for logic and power supply circuitry to be located outside of the enclosurewith all of the RF. I imagine the right-hand side to contain the relay crossbar forthe 2 x 3 (?) input/output matrix. (Maybe 3x5 -- don't know) I found some 1/2 inch teflon rod and a 1/4' to 5mm flex shaft adapter for the stepper motor on the roller inductor as well. The servo side of the roller inductor is at ground potential, but the ends of the capacitors have potentiallyhigh voltage. I put the teflon rod in my lathe, centered it in a 3-jaw chuckand eased a 1/4' drill bit in the end for the capacitor. A 1/2' collar with setscrew seems to do a fine job of holding the capacitor shaft securely. Added my most recent code to github: I think I'm at the point where I can start reasonably screwing down hardware, installingthe arduino hardware in the chassis and testing stepper code. This page last modifiedFri Oct 6 23:13:11 CDT 2017by timc! |