neatooooo
the long lines are your retracts? they seem very tall.
coolest thing I think I have done so far was a 4th axis chamfer along a curved edge
and I agree about the machining quality on certain markers
didnt really notice that stuff much until I became a machinist, then you never stop noticing flaws in parts
ABET accredited level II machinist - CNC Programmer - Mechanical Engineering Technologist
Rio Grande Inc.
as a person that literally measures flaws in machined parts in the single nanometer range ... this never goes away ... ever ....Originally Posted by ElPanda
didnt really notice that stuff much until I became a machinist, then you never stop noticing flaws in parts
social conservatism: the mortal fear that someone, somewhere, might be having fun.
They are relatively tall, but it's also deceiving in many ways.
Most of the retracts are 4"-6", the actual centre of rotation of the trunnion is 3" below the axis down the centre of the body tube. With the top body and everything else the centre of rotation is 3.75" below the highest part. So when it retracts and then spins the part under the tool, you are often far closer than the lines you see make it seem like you would be.
Now if you were using a machine with the rotational axis in the head, and RTCP, instead of a trunnion you can tell how the movements around the part will be and run it closer. Unfortunately with a trunnion you can end up with some pretty funky connections, that don't turn out to be anything like the lines you see on your screen (depending on your post processor) especially with rapid moving each axis as fast as possible and causing dog legs.
So you tend to run to a much safer retract height around the part, and even then it can be scary and the tool comes within an inch or less of the part. With 3 axis that's never very concerning, with 5 axis all moving at once it's pretty damn scary the first time. I may end posting a video at some point. If I go through and shorten them all to safe heights, it's a lot of work and not really that huge a time saving overall.
In reality I could go through and shorten each one down, and if I were running thousands of parts and was only there to feed the machine I would, but for me, the time taken to do that isn't worth it as compared to an inch or two of extra retract and approach in full rapid. I am better off letting the parts run and doing other work at the same time. I also only have one 5 axis machine and tend to play it safe
I will be running less than 100 of these in this batch, and maybe total forever.
Yep. Totally. And what's worse is I cut my teeth making concept cars and so I can't look at any car without picking up the flaws in the body panels and light lines etc.didnt really notice that stuff much until I became a machinist, then you never stop noticing flaws in parts
Last edited by Simon; 02-11-2014 at 05:20 PM.
For a sec, I read that and said to myself - "how in god's name did he cut his teeth? I bet that hurt like hell!" That's what lack of sleep will do to a brain...
Yep. Totally. And what's worse is I cut my teeth making concept cars and so I can't look at any car without picking up the flaws in the body panels and light lines etc.
I just have a quick CAM question for you because I have never worked in five axis, but how do you program the movement of the trunnion? I know in 3 axis you select your face, enter the parameters then your good, but for working with 5 axis do you have to enter any kid of angle parameter or does it do that automatically once the face is selected? Also for zeroing the machine is it the same as a 3 axis machine? and one last question but when setting up tool paths how do you keep track of the trunnion and make sure you don't crash? or do you just never setup any angled cuts on the underside of the XY plane? sorry for all of the questions!
So does anyone else try and rotate the part when they look at the picture?
5 axis can be programmed from any origin just as with 3 axis, but for organizational purposes it's usually done from the intersection of the centers of rotation. You can do that for everything so long as you know the dimensions of the trunnion being used ahead of time. (the specific measurements should come as an inspection sheet with the hardware).
The bad part, though, is your workpiece might be located in a different spot which is different from the origin point. If you need extreme part-to-part accuracy, or if you have an oddball fixture, you have to go back to the CAM system and start adjusting things. Not as simple as simply moving the offset as with 3ax, because the origin has to be linked to the axes' rotation.
Some machines will keep track of this on their own, but not all. And there's cases where either methods are preferable.
Hey simon are you programming those cardinal angles as separate offsets, or did you find a quaint way to do them all in one fell swoop?
We are pretty damned lucky to have Keith Belsey as our body "machinist". We are pretty tough on ourselves as to what we call acceptable or not, but he takes it to another level.
How long has he been working at PE? B/c I must say, I do like his work quite a bit. I really haven't seen one PE gun that I haven't really liked the milling on in recent years.
guys working on multi-axis cnc machines, check this baby out:
http://ibspe.com/category/machine-to...-solutions.htm
we were running this at a professional conference on some really nice basically new Haas 5 axis stuff ... and it was insane. even brand new, in profiling these machines were 30-90 microns error motion. but using this instrument, and having access to the source code of the machine, re-setting the axis alignment and error compensation maps (ie no mechanical adjustments at all), we were able to put these machines down to 5 micron error motion or better.
it works like a dynamic ball-bar test. the probe nest holds on to the target, and the axis move in synchronized moves while the probes record and then compute back out in spacial coordinates the error.
after mapping each axis, it computes out all the new alignment numbers, which basically tells the cnc computer to interpolate its moves differently to correct.
very cool machine. we were all trying to figure out an excuse to buy one, but i don't do much 5 axis stuff.
social conservatism: the mortal fear that someone, somewhere, might be having fun.
Posting Permissions
Reply With Quote