Sign in to follow this  
Followers 0
ParaffinPower

Trajexia and interpolation

5 posts in this topic

Hi all I'm using TJ1 to perform circular interpolation. Actually, it's lots of linear interpolation generated by Cad2Motion. The shape is a fairly complex with curves, though. Both ballscrews have different pitches. I've set appropriate gear ratios in the drives so the "UNITS" for both axes are the same. Here's my problem. If I set SPEED, the the periphery speed changes, but only up to a limit - it won't go any faster (the servos are happy to go faster) Any ideas what's imposing the limit Code snippet, so you can see how I'm achieving 'circular' interpolation MOVE(0.3979,0.1136) MOVE(0.3894,0.1273) MOVE(0.3825,0.1418) MOVE(0.3767,0.1568) MOVE(0.3705,0.1716) MOVE(0.365,0.1866) MOVE(0.3603,0.2018) MOVE(0.3551,0.2163) MOVE(0.3481,0.2293) Thanks guys, Pp

Share this post


Link to post
Share on other sites
Why aren't you using the movecirc command? You say the speed is being cut off at a certain level. What is that speed in RPM of the servo motor? I see it are quite small movements. You are probably moving using a trapezoidal movement, but by raising the speed, at a certain moment it becomes a triangle. So raising the speed even higher will not make any difference. This is why movecirc commands have been conceived. You get a decel and standstill after every small movement.... I would not change the values of the gearboxes in the drives but use the units of the trajexia instead. You only have problems with connect functions (which actually works in encoder pulses) Beegee

Share this post


Link to post
Share on other sites
Hi Beegee I would love to use MOVECIRC, but the shape is very complex with no 'real' (?) curves. The shape is virtually 'hand-drawn' in Autocad. The trapezium/triangle profile idea is good - I guess each move never achieves maximum speed. I'll take the ACCEL and DECEL through the roof and try that!! The Initially, I had the TJ1 managing UNITS, but the X and Y speeds were very different. MOVE (10,0) and MOVE (0,10) would move the axes at different speeds (SPEED same for both axes). Weird Thanks for the input. I'll try and report. Pp

Share this post


Link to post
Share on other sites
What kind of servos are you using and what is the control method? I have allways used units (MC402/MCW151/TJ1) and only had problems when the units were negative. I noticed the MAC in Barcelona of possible bugs in the firmware, they should be looking into this now.... maybe there are big differences in gains or (I had it once) you could be using 1500RPM and 3000RPM motors mixed and they have different default encoder pulses per revolution beegee

Share this post


Link to post
Share on other sites
you could also make a PSWITCH just before the end of your movement and use it to use movemodify. Then you will not decelerate to standstill to accelerate again. another possibility is making a continuous move on a virtual axis and do the movements with movelinks. Then you can eliminate the decelerations and accelerations and create continuous movements on X and Y. (I did this already on single axis machines for the replacemment of clutch-brake systems and it works fine) you could get something like this( but you'll probably have to improve to get good results) base(3) forward [i][color="#009900"]'acceleration for the first move [/color][/i]movelink(Table(0),Virt_dist,accel_dist,0,3) axis(0) movelink(Table(1),Virt_dist,accel_dist,0,3) axis(1) [i][color="#009900"]'continuous move[/color] [/i]for i=2 to (Max_number_of_pos-2) step 2 movelink(Table(i),Virt_dist,0,0,3) axis(0) movelink(Table(i+1),Virt_dist,0,0,3) axis(1) next i [i][color="#009900"]'deceleration for the last move[/color] [/i]movelink(Table(Max_number_of_pos-1),Virt_dist,0,decel_dist,3) axis(0) movelink(Table(Max_number_of_pos),,Virt_dist,0,decel_dist,3) axis(1) where Virt_dist is the step of your movement, all postions are in the table and you have max_number_of_pos divided by two positions of X and Y

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now
Sign in to follow this  
Followers 0