Re: waterjet cutting speed fluctuations
hello again, as shown in images from post1, there are cutted 4 geometrical groups :
- 1st group - 3 lines
- 2nd groups - lines and arches
- 3rd group - 1st one a bit scaled & mirrored
- 4h group - 2nd one mirrored
So i take into consideration only group 1 and 2
This post is about group 1 ( image 01 ), simple as it is, it contains 3 lines ...
I have extracted positions and feed from program attached at post 1. I have created some charts, and paste them inside a doc file attached. If you wish to take a look, please consider a view as shown in image 02, so to have a general perspective
Is easy to locate on charts position where one segment ends and another begins ...
Charts :
1) feed - feed variations among all 3 segments
2) df - changes between 2 consecutive feeds : Feed_current_position-F_last_position
3 & 4) dx & dy - changes on X ( Y ) axis between 2 consecutive positions
5) chart 1 again
6 , 7 , 8 ) absolute values for charts 2 , 3 & 4
9) dl - segment length between 2 consecutive positions; dl^2 = dx^2 + dy^2
[ observations ]
a) since material has been pierced, than feed can be accelerated and decelerated faster, as shown in image 03; maximum value may be equal to "program maxim" or close to it; i am not sure, but feed = "program maxim" may not work, because on chart this feed is mantained for a short period of time, and thus, delay helps to pierce material; speeding with feed = "program maxim" may not pierce, since there is no delay arround each position, so 80..90% may work
b) every chart has a sharp design, as shown in image 04 ... i don't know why ... maybe this behaviour is not present on other machines / water jet instalations ... this aspect can be improved from code
c) as seen in "df" charts, feed always changes with constant value, except for local oscilations; cause most likely similar to conclusion b) ... machine works smooth almost all the time, and suddenly something shakes; biggest feed oscilations, as shown in image 05, are because cutting direction is being changed; direction changing is done with low feed, but machine accelerates / decelerates to this values aggressive; changes should be smooth, and so to cause a delay, so waterjet will stay near this point for longer time, allowing better piercing
d) as seen in "dl" chart, cutting start with small segments ~0.1, and reaches also segments close to 0.9 length; this is because of the long acc / decc paliers
[ conclusions ]
1) speed changing palier ( acceleration / decelaration ) should be shortened, i guess from software interface
2) is necesary to :
2.1) ......... improve cornering
2.2) ......... reduce sharp design, described at observation b)
1 >>> less process duration
2.1 >>> same effect as 1) but with less impact
2.2 >>> smooth behaviour
- water jet machine is not mine, otherwise i would experiment with it ... i tried to rough some metal parts, but jet duration was too big compared to mill ; i don't know how much time i can reduce, but i am sure i can ... i have discused this conclusions with jet operator, and he discoverd an interface from where may be possible to change acc / decc paliers
- sometimes is good if things work
- much better when things work much faster; this type of improvements are time consuming ... kindly !
Ladyhawke - My Delirium, https://www.youtube.com/watch?v=X_bFO1SNRZg