Lying to HST toolpaths to get motion you want??

[crazy^millman]

I have a process improvement project that I am working on. I am using the HST toolpaths and would love to have one operation to cut the 18 features on one side of the part using my 3/4 endmill I have chosen and then come back with by 1/2 endmill to pick out the corners. What I am finding is I am having to break it up into about 40 operations to cut the 18 features the way I want. I want certain cuts on certain depths for closed pockets and on open pockets that are deeper I want more control over them and on the open areas coming up to one wall I want a different type of motion all together. One thing that is making extra work is the no respecting of the steep and shallow settings. I want one cut to -.845 deep and one cut only. Now I get a -.840 and a -.845 deep cut. I limit the range to -.8449 to .845 and I still get 2 cuts. If I make the range .05 to -.845I get one cut, but then it wants to helix all the way down to the floor even thought there is a .812 drilled into the each one of the closed pockets and the stock model has them referenced in them. I am using toolpath editor to limit this and thankful the cuts is respected in the editor, but any change to the file and I will have to redo all the edits.

 

Has anyone else run into situations like this were you just motion from the HST toolpaths you don't want? I hate all the extra work this is creating, but I know what I want for motion and how I want to cut the part. Am I just being to picky about what I am getting? Should I just do like I see so many other programmer doing and just making toolpaths and let what we get be what it is and just see what happens? I don't think I am being to picky and a customer is paying me for my best and if a certain way to go about it is the way I think it should be done then I should do that has always and will always be my thought.

 

What do others think? Let it do what it is going to do? Take the time and break everything out like I am doing to really dial it down? sorry IRAT part so I cannot share what I am working on.

[jlw™]

Practically every time I use it Ron.  It about drives me nuts.  I wish the depth of stepdown as well as the steep/shallow worked better.  In my opinion it should find flats within the given stepdown and hit the flats on the step up.  I find I will often have to make a unit for each flat level.

 

JM2C

[Matthew Hajicek - Singularity]

IMO it all depends on how much cycle time it saves and how much that's worth to the customer.  If the want to pay me to hand edit to get out the last second of cycle time that I can, then I will.  I guess I have it easy since all my parts are short run, so cycle time isn't a strong pressure.

[crazy^millman]

Well we are talking about something running well over 20 hours and reducing it by at least 40% if not more. I think the extra effort could mean the difference in about 30% to 40% better tool life and it should help reduce the time overall by 5%. It is 8 parts per week so if my math works out right the customer should see some serious reductions. One off part wouldn't give it a second thought, but on production jobs every minute adds up.

 

Had a conversation with a Swiss trained on a different product not to long ago. He was just teaching basic moves. I asked why all the air cuts and he was no worries. I took the example file and showed him on one toolpath 8 seconds of waste and that by spending and extra 5 minutes programming it differently we saved 2 seconds per index for the 4 indexes on that one tool. Okay and what does that mean? I asked him to do the math on a 10k run of parts and tell me how much saving was that. 22.22 hours of savings is what is worked out to is what he said. I said nope that is 44.44 hours difference. People always forgot the 2 for 1 rule. For every minute saved is a minute you can put on something else. Most times I do extra work for customers and never get paid for it, but I know they got my best and if I know that then I am good with doing my best. Not that everyone I have done work has been happy with it, but can't win them all like some one said today on the forum.

[Jayson Kramer]

Don't just follow. I will spend hours some time picking things to cut away air or making the tool do what I want. it would be nice at times to have more control in a single operation. But then more people would complanine saying to many settings and options.

[civiceg]

 

 

. One thing that is making extra work is the no respecting of the steep and shallow settings. I want one cut to -.845 deep and one cut only. Now I get a -.840 and a -.845 deep cut. I limit the range to -.8449 to .845 and I still get 2 cuts.

 

You can make them the same to get one cut.

 

 -.845 min and -.845 max.

[jlw™]

I will say it again, manning up and creating the geo or putting in the work to get the best toolpath is what separates the men from the boys. IMO

[civiceg]

I think it also requires a judgement call. How many parts are being ran, how much time am I going to shave off, how much longer is it going to add to the programming time, what's the likelihood of seeing the part again, is the tooling cost worth it? etc. 

 

Have I nitpicked tool path to make it perfect and optimized, absolutely. But, I have no problem having a less than 100% optimal program hit the machine, when it would be sitting there idle, not cutting for a week. But, I don't think contract programming has that option. 

[crazy^millman]

You can make them the same to get one cut.

 

 -.845 min and -.845 max.

 

Worked perfect. Thanks in my mind I was thinking okay no effective zone no cut, but shows what I know. Thanks for the suggestion.

 

I think it also requires a judgement call. How many parts are being ran, how much time am I going to shave off, how much longer is it going to add to the programming time, what's the likelihood of seeing the part again, is the tooling cost worth it? etc. 

 

Have I nitpicked tool path to make it perfect and optimized, absolutely. But, I have no problem having a less than 100% optimal program hit the machine, when it would be sitting there idle, not cutting for a week. But, I don't think contract programming has that option. 

 

You do, but would you want to go back to someone who gave you okay stuff or really did hard work and treated every project with care and concern? Fine line we walk of doing enough or too much. I would love to get paid for every hour we put into a project, but I will always do my best and like that about the group we all care do our best. After almost 4 years I can still put food on the table and keep a roof over our heads and value the relationships we have with the different companies we get to work with.

 

HST toolpaths are a game changer for today's manufacturing and they do amazing work, but I always seem to be one that pushes the limits. I will take the time and document every thing I want different. I will make videos and screen shots and give feedback to CNC Software like I have done for years. There are what many consider High End Software's that use a 3rd party to do HST toolpath creation. This is one area CNC Software made a very wise decision investing in doing this on their own. It does amazing things and I am amazed what we can do with these toolpaths.

 

Funny few years a go we were not paid for $20k of work and one of the programmer's comments was there is no way that carbide endmill will hold up for 5 hour in Ti using HST toolpaths. Never even tried the programs and long story I will no go into, but just seen an endmill cut for 9 hours in Ti and looks like it didn't even get used good yet. Might get another 9 hours out of that endmill. I am picky and I am very detailed oriented when it comes to my work. Some might even call it compulsive and I am okay with that. The only way to improve something it to let the people know who made it that it needs improving.

 

I am not trying to rip apart anyone's hard work, but I cannot be the only running into these issues. The more of us that have a productive conversation about it the better. I appreciate the hard work any Software Company put into their product, but as an end user pushing things to places most never do I expect to see things most never don't. Not bragging and those who have worked with us know the work 5th Axis CGI does. Still some of the these things need to work better and hopefully the thoughts shared and conversation is well received.

[crazy^millman]

You can make them the same to get one cut.

 

 -.845 min and -.845 max.

 

Spoke too soon, it will not respect the stock model and plunges the tool into the stock and not on the center of the hole. If I give it the .0001 room in the steep and shallow it does respect the stock model and not plunge into the stock model, but I get 2 passes. Thanks for the suggestion.

[Watcher]

Worked perfect. Thanks in my mind I was thinking okay no effective zone no cut, but shows what I know. Thanks for the suggestion.

 

 

You do, but would you want to go back to someone who gave you okay stuff or really did hard work and treated every project with care and concern? Fine line we walk of doing enough or too much. I would love to get paid for every hour we put into a project, but I will always do my best and like that about the group we all care do our best. After almost 4 years I can still put food on the table and keep a roof over our heads and value the relationships we have with the different companies we get to work with.

 

HST toolpaths are a game changer for today's manufacturing and they do amazing work, but I always seem to be one that pushes the limits. I will take the time and document every thing I want different. I will make videos and screen shots and give feedback to CNC Software like I have done for years. There are what many consider High End Software's that use a 3rd party to do HST toolpath creation. This is one area CNC Software made a very wise decision investing in doing this on their own. It does amazing things and I am amazed what we can do with these toolpaths.

 

Funny few years a go we were not paid for $20k of work and one of the programmer's comments was there is no way that carbide endmill will hold up for 5 hour in Ti using HST toolpaths. Never even tried the programs and long story I will no go into, but just seen an endmill cut for 9 hours in Ti and looks like it didn't even get used good yet. Might get another 9 hours out of that endmill. I am picky and I am very detailed oriented when it comes to my work. Some might even call it compulsive and I am okay with that. The only way to improve something it to let the people know who made it that it needs improving.

 

I am not trying to rip apart anyone's hard work, but I cannot be the only running into these issues. The more of us that have a productive conversation about it the better. I appreciate the hard work any Software Company put into their product, but as an end user pushing things to places most never do I expect to see things most never don't. Not bragging and those who have worked with us know the work 5th Axis CGI does. Still some of the these things need to work better and hopefully the thoughts shared and conversation is well received.

 

Funny how there are professionals and professionals. There are people with amazing opportunities, technologies and resources at their disposal, or reachable if they want to have them, but they never bother to walk the extra mile. Some even feel they´re are entitled to the privilege they have, because in many cases they never had to walk the walk and work for years without resources or honing their skills with what they had at the time.

 

It´s a privilege to know people who walk the extra mile just because it´s how their minds operate. People who take pride in what they do and don´t make excuses or a lazy job no matter what.

[BenK]

I have seen the same thing many times. It would be nice if we have a number of passes option and a manual depth option. Those two would make my life so much easier. 

[crazy^millman]

I made the suggestion with 2018 to use the new selection process to set percentages of cut in the areas picked. We now can decide to leave different amounts in different areas so it would be nice for the Opti-Rough to pick 100% depth of cut in this section and then maybe 120% depth of cut in another area and then 65% depth of cut in other area. Going even further it would be nice to pick direction like in and out with those choices. Really like the new selection process and if I could get some customers using 2018 I would move to 2018 right now and not use 2017 again. I had to go in X9 for an old project recently and had to remember where things were.

 

Still not a fan of the ribbon ahd have the RMB and QAT highly customized, but good with 2017 and 2018.

[Matthew Hajicek - Singularity]

If only they'd fix the access key's I'd be all over it.

[Zoffen]

I agree to all points here it is frustrating getting them to do what you want sometimes.

 

I would like to see a graphical depth selection kinda like they use in NX. That is one thing i really liked about it when i got a demo.

 

Plus dragging things around on the screen makes for good demos!

 

This would give you easy and precise control of all the depths and stepups.

[crazy^millman]

I figured out something I was doing wrong with the toolpath. Only took me 2 weeks to figure it out, but that is life. I had step up checked on the toolpaths. When I disable check up and use the steep and shallow with the same depth I get exactly the results I want. No toolpath editing needed what so ever. I will call this one a PINIC. :vava: :ouch: :realmad:

[gcode]

I figured out something I was doing wrong with the toolpath. Only took me 2 weeks to figure it out, but that is life. I had step up checked on the toolpaths. When I disable check up and use the steep and shallow with the same depth I get exactly the results I want. No toolpath editing needed what so ever. I will call this one a PINIC. :vava: :ouch: :realmad:

another lesson well learned at the School of Hard Knocks... the kind you don't get reading books or sitting in a class

the kind  you won't ever forget

[crazy^millman]

Talking about those lessons. I learned in Vericut you must make sure the Tool Compensation number matches the tool number in the Vericut Tool Manager when using a custom defined profile brought in through the Interface. If it doesn't you can waste 2 weeks fighting false errors. Will not be forgetting than one anytime soon either.

 

I have gone back and stripped out about 25 operations that I was doing and now I have 4 operation giving me what I wanted. Still not doing open areas exactly like I wanted in one operation with depths by keeping as one, but 4 is much better than what 25 operations was doing. :thumbsup:

[Watcher]

Talking about those lessons. I learned in Vericut you must make sure the Tool Compensation number matches the tool number in the Vericut Tool Manager when using a custom defined profile brought in through the Interface. If it doesn't you can waste 2 weeks fighting false errors. Will not be forgetting than one anytime soon either.

 

I have gone back and stripped out about 25 operations that I was doing and now I have 4 operation giving me what I wanted. Still not doing open areas exactly like I wanted in one operation with depths by keeping as one, but 4 is much better than what 25 operations was doing. :thumbsup:

In case you want to change Vericut behavior in regards how driven points are parsed, there's a macro you use in the start of processing that gives you low level control about how Vericut must behave when reading a driven point, or even in the lack of one defined in the tool.

 

There are switches that can be passed to the macro.

[crazy^millman]

In case you want to change Vericut behavior in regards how driven points are parsed, there's a macro you use in the start of processing that gives you low level control about how Vericut must behave when reading a driven point, or even in the lack of one defined in the tool.

 

There are switches that can be passed to the macro.

 

Daniel, thanks. What I have run into has been documented and handled directly by CGtech personal to help resolve the issue I was running into. I gave it the good old college try and after some back and forth we finally got to the bottom of the issue which is what I stated above. It was not a dig on Vericut, it was just a issue I ran into that hopefully my sharing will educate someone else if they run into something where they are getting negative comp errors on a G41 toolpath that make no sense. In all reality it was a simple fix and just something I was not aware to look for. I know now and if someone else can be spared the lost time I lost on the issue then I am glad to point anyone in the right direction to prevent them from having to go through it. Like me sharing this topic. I have a much better grasp of the HST toolpaths after going through everything I just did. I am no expert, but I know what I know and seems like I know a little more than some. Sharing knowledge and helping others means we are willing to look stupid and be proven wrong. Without effort nothing is accomplished. 

 

Yes I work with a different product and like it, but that was not my reason for sharing the issue I ran into. Our customer requires we use Vericut and we give the customer what they want.

[Watcher]

Don´t worry about working with other products... They´re all great stuff... 

 

I imagined you shared to help other guys in a similar situation... We used to have this problem with driven points in our WFL VMCs, then I changed some switches in this macro and we never missed an error in our code or in VERICUT ToolManager anymore...

 

It´s good to share these things... most software these days are highly customizable but seldom we spend time reading the help file...

[crazy^millman]

Don´t worry about working with other products... They´re all great stuff... 

 

I imagined you shared to help other guys in a similar situation... We used to have this problem with driven points in our WFL VMCs, then I changed some switches in this macro and we never missed an error in our code or in VERICUT ToolManager anymore...

 

It´s good to share these things... most software these days are highly customizable but seldom we spend time reading the help file...

 

This was one of those once in a 10 year things. Normally never a problem and previous to this had not been an issue, but it became one. One I will hopefully never forget how to fix if it should ever happen again. Let's pray it never happens again. :thumbup: