FBM drill problem

[pullo]

Here is a simple part that was brought to my attention which seems to show there is a problem with FBM drill , specifically when Clearance and Retract depth values are set to “Top of hole plus” and the depth set to incremental (self evident as what we are asking for is incremental in nature).

 

The linking depths do not come out right. Now the weird thing is I played around with these depths for about two hours and all of a sudden I got the right results . (the right result but one in a lifetime opportunity.jpg). I saved the parameters as default and tried them on the same part starting from scratch. I was not able to duplicate the same right result. I also grabbed the screen , with the parameters set to give me that one off time when the result was correct. No help from that

 

The only result that is near enough for real world solutions is if you set the linking parameters to manual and their value is an absolute -10 mm (best with manual minus 10 abs.jpg).

 

If I were to try to diagnose the situation , I’d say that the settings read “Top of hole plus” , but in reality they are not that . Somehow by heavily tweaking the parameters I was able to set the depths to “Top of hole plus” , but I can’t repeat that anymore.

I have sent this to QC , but as I said there seems to be the right  output (as I have a fleeting proof of that) if you tweak these parameters in the right order. So anybody want to give it a try.?  The Absolute, -10mm , Manual  input is a close enough result , but it is only a lucky shot as that will not work on the next part 

If this would work  as it should , it would be definitely a  powerful tool to use.  

Gracjan

 

 

FBM drill problem 2018.mcam

[pullo]

Hey where did all my superheros go ? Gcode , JParis  , etc ,  anybody ?

Gracjan

[crazy^millman]

No super hero sorry to disappoint. I don't use that part of the software and not sure any of them do either so sorry can't help. You spent 2 hours trying to solve an issue with FBM that I would have just programmed the old fashioned way(still my current way) and got done in less that 5 minutes and had everything just like I wanted the way I wanted it and been done sir.

Sorry took me 2 minutes. Here is the results.

5th Axis Answer to FBM drill problem 2018.mcam

[pullo]

It's a customer's problem , so first time around to get the job out of the machine , I did it also 5-axis style, but I see lot's of potential with FBM, only if it would work like it should...

BTW , I was supposed to add you to my superhero  list , just didn't want to get the "spelling" wrong  C^Millman  , there I did it .

Thanx anyway for the effort... I'll nail this one over Xmas break.

 

GracjanInsert other media

[crazy^millman]

10 hours ago, pullo said:

It's a customer's problem , so first time around to get the job out of the machine , I did it also 5-axis style, but I see lot's of potential with FBM, only if it would work like it should...

BTW , I was supposed to add you to my superhero  list , just didn't want to get the "spelling" wrong  C^Millman  , there I did it .

Thanx anyway for the effort... I'll nail this one over Xmas break.

 

GracjanInsert other media

FBM to is still not something I push. I have given parts to Feature based system that other CAM offer only to see them fail. End of the day interaction is the best teaching tool for manufacturing and pushing FBM for even things like this is dumb down a programmer in my humble opinion. It was such a good idea of the sales force at CNC Software they thought putting out even a 6 year old can program was a good idea. That is really a majority of the problem with manufacturing is everyone wants it to be so simple even a 6 year old can do, but guess what it is not. It takes a certain amount of thinking and ability to do certain things. I will take music and use it as my example. Our 16 year daughter has played piano for 10 years and she is very good for her age, but the 6 year old's she is teaching are not. Why is this? They must be taught muscle memory and understand what each note means. They must hear, see and feel the music on the piano the way the composer of the music wrote it. They must be able to think and train their brains to use both hands not just one hand. They must be able to sight read music and then take that and turn into something no computer can even come close to matching. FBM is trying to take learning out the job down to a level where that 10 years of learning how to play the piano are gone and play the piano for you. Guess what just like having a computer play the piano today it can be done, but it is not the same to the trained ear. To the trained ear it has no soul, it has not body, it has no feeling it is just noise. When we keep pushing our profession into the corner and saying everyone needs to have easier ways to do their programming job we are in effect doing the same thing. We need to show people how to think abstract and get their soul and mind into the process of manufacturing. We need to take ways the excuses of this is too hard and teach others how to think critically. How to hear the sound of the machine, tool and everything in the environment that is telling the story of the process.

Gracjan non of this is directed at you and I think we have meet in person at EMO some years ago, but manufacturing needs a revamp not just with automation. The people must have a desire in them of wanting to learn and put in the effort to learn. What is the saying there are many starving musicians. They are all willing to go without to do what they love. We need to make a generation of CNC programmers willing to think the same way about programming. To sit down and do the scales of programming. To be willing to humble themselves and learn and get the basics of what this profession takes. Understand how to take a piece of 1" (25.4mm) square stock tool steel and take a hacksaw and cut that block to make a perfect square by hand. To then make it a perfectly square down to .984 (25mm) block with a 2 finish by hand using hand files and Emory cloth down to 3000 grit paper. Go cut some threads on a conventional lathe and see and hear what happens when you forget to engage the half nut correctly and just scrapped 2 days of work. Learn by doing not by reading what happens when you do these things. FBM may have it's place, but I will repeat what I told a Manufacturing Director of one of the companies sending craft into space after he asked me this question when we brought to help them get their 5 Axis programming moving in a better direction. " Why are you doing this kind of work? Don't you know in 5 years you will be out of a job? They are working on software that will take any model given to it and make perfect CNC programs. " I laughed a mentioned the group doing this by name (I will not say in a public setting to avoid slander) They have a lot  PHD's working on that for the last 14 years with well over 20,000 hours of time invested into it. When I was a recent event they approached me and was hoping to interview me to see what they could learn from me about programming 5 Axis. Currently they can only do 1% of what they throw at it and those are simple shapes. I will keep taking my chances doing what I am doing. Some might read this and think what an arrogant person I am. I am not delusional in thinking one day we may not be needed. However those who don't give themselves any value also don't get value. If you cannot understand what you are worth how will anyone else give you worth. I always try my hardest to be humble and I know there are better programmers and machinist out there than myself. One thing I learned a long time ago is if you don't think of yourself as a business and invest in the business of yourself guess what not to many other will. Be willing to invest in yourself and take the time to learn something and show others you are willing to learn and are teachable then people will help you. Put your ear buds on and just hit the green button and not learn guess what you will not learn.

Sorry to go on, but FBM is cool, but using your brain and being able to know what and why something works and being able to figure it out for yourself is much cooler. Guess what it makes your smarter. The smarter you are the more valuable you are.  Don't be a fries with that programmer, but a composer of the process and make yourself create the music of the program that goes above and beyond and it will bring the reward you seek.

[pullo]

Wow , I did not expect that  )  .  I am going to have to read that again , but if I may suggest that we go back to the original problem.

I consider this a bug. So it's been sent to QC , that is it for me .

I actually looked at this more closely . I went from X5 to 2019 and found out it's always been like this buggy.  So I'd say if no one has found this and it's been there since X4 ,  it just sort of tells us of the "widespread " use . 

I put this out more as of  a brainteaser than looking for a solution  by employing  peer power .   

OK , there's 10 mold cavities waiting to be done and nothing  gets done if you don't do it , so I'll be off...

Gracjan

[Colin Gilchrist]

I think the issue may be where Mastercam "thinks" the top of each hole is. I know that probably sounds like "well, yeah, that's what I just showed you", but hear me out.

Try taking the part, 1st off, and making surfaces from the Solid, if your geometry is a Solid. Try running FBM on the surfaces, and see if your results change.

Now, with the surfaces, take your "hole cylinders", and un-trim them. The issue is probably related to where "computationally", FBM thinks the "Top" is. My guess is that some of those angled holes are so "inclined" to the surface that when you Un-trim the cylinders, the tops will be sticking out quite far from the face of the part.

Those are simply guesses on my part. Really, QC is going to need to look at your files for any chance to get this solved for you.

[pullo]

Like I mentioned before , this is a puzzle which has been already solved .  I  saw it work perfectly with the geometry at hand. 

I know that if I do not force the vectors to come from one side the depths look OK .  

QC  has the file and all the pics I posted here ....

Right now  I am recovering from one response here :)))))))

Gracjan

 

[pullo]

I sort of solved the problem . I noticed  one hole was behaving correctly , a hole  which  I had  added in Mastercam to the original "lump" solid that was created externally.

1.I first created arcs  with Hole axis .

2. I removed the holes with Find features / remove them and redid the holes with the arcs  sort of creating  a "direction" , which was picked up correctly on all counts the first time around by FBM drill.

The rest is history

But  as The original file was a legal solid , I call it  again a bug.

Gracjan

 

bang on.mcam