[SBC Cycle]

I'm going to try documenting my setup of a Haas VF2 with an HRT-160 in BobCAD. I may eventually get around to the machine base and the spindle but I'm starting with the basics of the table and the rotary. This picture shows the Haas machine geometry I am working with and how the Machine Coordinate System is laid out. Looking straight down on the table, all X and Y Work Offsets are in the negative region. The tool is at the "Home" position in the upper right hand corner:

Attachment 207672

The rotary is positioned in the center of the table. Relative to the Machine Coordinate System:

G53 X-15.0000 Y-8.0000

One of my first questions/decisions is where Machine Coordinate Z0 should be. Trying to keep everything as accurate as possible, my first try would be to set Z0 up high at the tool change position:

Attachment 207674

I'm thinking this may not be practical as I have yet to see any of the BobCAD examples defined that way. The 3 axis machines have the top surface of the table as Z0 and the 4 axis machines establish Z0 as the center of the rotary. But I do want to check limits on the machine, use real Work Offsets, and basically make it as close as possible to the real world. This machine has a Renishaw probe and the way tools are set and offsets established will be important (I think). So is this the proper way to start? I don't know, right now I have the table as Z0 instead of the rotary and maybe that will be my first error but that's where I'm going to start.

Attachment 207676

[jrmach]

Well,I certainly have no experience on a 4th,,,BUT,I have a lot of experience with all kinds of things that go round and round,ha.Just from everything I have ever done,my mind would always say the center of the Rotary is Z0.More advantageous than not,as for what I have done,usually.My .02 cents from a Manual Machinist view.

[SBC Cycle]

Ok, it seems like having the center of the rotary positioned somewhere other than X0 Y0 Z0 is not going to be a problem. You just have to tell BobCAD in the Machine Definition where the center of the A rotation is:

Attachment 207682

This matches the position where I have placed the rotary on the physical machine. My exact probe measurements were within 0.100 in XY and Z is exactly 5.000 inches off the table. Looking good so far:

[SBC Cycle]

Well,I certainly have no experience on a 4th,,,BUT,I have a lot of experience with all kinds of things that go round and round,ha.Just from everything I have ever done,my mind would always say the center of the Rotary is Z0.More advantageous than not,as for what I have done,usually.My .02 cents from a Manual Machinist view.

You are definitely thinking ahead! At first it seemed like a bad idea but then I found that I had to move the centerpoint definition for the A axis in the Machine Configuration. That basically makes the center of the rotary Z0 like it should be. That took me a bit to figure out.

So basically it shouldn't matter where you establish Z0 when you save out your STLs. I chose to stick with Z0 at the table like it is for the 3 axis machines. That makes it easier for me to share simulation STLs between Machine Configurations. I can simply delete the 4th axis and make the regular 3 axis Haas mill configuration. Maybe add a vise in.

If you do not save your STL with the center of the rotary at Z0, you can account for that in the A definition.

[SBC Cycle]

So basically it shouldn't matter where you establish Z0 when you save out your STLs. I chose to stick with Z0 at the table like it is for the 3 axis machines. That makes it easier for me to share simulation STLs between Machine Configurations. I can simply delete the 4th axis and make the regular 3 axis Haas mill configuration. Maybe add a vise in.

If you do not save your STL with the center of the rotary at Z0, you can account for that in the A definition.

I did forget to mention one thing. Now that I have messed with the "easy" setup. There is one more thing I'm going to have to do from now on when I need FULL machine simulation to look and act correctly. Without using "real" Work Offset values, the simulation looks a little screwy:

Attachment 207690

It simulates the stock just fine but when the Machine is visible it looks bad and part seems to jump around the screen.

I have to correct everything by setting a Work Offset in Machine Setup 1 to match my setup (accounting for fixture in X):

Attachment 207692

Now everything looks like it does above.

[btomlinson]

Did you model your own 4'th or is this available from Bobcad ?

[SBC Cycle]

Did you model your own 4'th or is this available from Bobcad ?

Actually my rotary originated with the files from this link: HAAS VF5/40 - SolidWorks, STEP / IGES, STL, Other - 3D CAD model - GrabCAD

I reworked the SW parts so that they were configurable, then created a new configuration for the HRT-160 based off of my measurements of the actual rotary. So now I can move it around in SW or suppress the 4th axis and load a vise on there. Save the assembly out as ACIS (.sat) and open in BobCAD. I save out the individual pieces as STLs for the simulation. That's my current workflow anyway.

Here is another one that I didn't download but looked good: CNC machine table - SolidWorks, Other - 3D CAD model - GrabCAD

[SBC Cycle]

Well you can see that I have a problem, the HRT-160 is only 5 inches from the table to the center of the rotary plate:

Attachment 207792

Obviously I'm going to have to raise it up but I want the simulation to catch this. I believe you MUST be in Kinematic mode to detect collisions but someone can correct me if I'm wrong about that.

I'm going to open up the Machine Definition under Current Settings and add a Collision Check between my stock and the table. I would also like to figure out how to add fixtures and check those as well but the fixture I built is already small than the 5" swing of the rotary. Here's the CC I added:

Attachment 207794

And it looks like that did the job. The two colliding pieces light up in Red whenever there is an oops. Moving right along...

[SBC Cycle]

The swing of my part in the fixture is going to be just less than 6" so I need a 1" minimum spacer to clear the table. I made a 1.5" riser just so I don't have a heart attack watching it swing so close, lol.

Attachment 207802Attachment 207804

I created a new Machine Definition to use with the riser plate because I make this part often and other larger parts with light machine work. I shifted all of my rotary STLs up 1.500" and just sketched up a solid block in BobCAD to represent my riser block. I added that geometry to my machine definition and moved my rotary centerpoint up 1.500 to make it a total of 6.500" off the table:

Attachment 207806

Now I believe I have proper machine definition. I already ran these parts before I had the Pro Simulation and they were good, it's nice to actually be able to check my tool clearance heights. I also had to adjust my Work Offset Z to 6.500 for the simulation to come out perfect in Kinematic mode. The Work Offset only seems necessary for full limit checking and machine simulation. I want to use both so I don't mind setting them.

[ChipMaker729]

I love this! You are absolutely right about having to adjust the work offset value for the machine setup in order to position the part properly. You can define a different XYZ shift for each machine setup you create in case you need to flip the part in a fixture. That's the primary reason you found that the machine 0 is set up in the center of the table or the center of the rotary, because it's easier to think with when setting the work offset value for machine simulation. The value is always just the shift amount from machine zero (simulation machine which the machine zero may be different than the actual machine).

The one thing I didn't see mentioned was adding collision pairs to the machine. In order for the simulation to flag collisions you must create a collision pair in the machine definition. Right click on the top level and add a collision pair. Then you just simply put the geometry you want to check for collisions in two different groups. For example with a 4 axis machine you probably want to have the tool, the holder, and the head of the machine in one group and put the components of the rotary axis in the other. Also perhaps another collision check of the workpiece and the table.

[SBC Cycle]

Thanks ChipMaker! I am just now getting around to adding collision pairs. If you see post #8, I added a CC between the stock and the table. If I want to collision check the tool and holder with the rotary, is that best done in a separate CC group?

[SBC Cycle]

I tried to upload the STLs to the forum last week but they were too large to upload. I figured if they were too big to upload, it was probably too detailed to use in simulation without eating up resources. My laptop handles it fine but my desktop choked. I finally got around to checking the settings in SW and instead of using BCC to create the STLs I exported them directly from SW as STL (I couldn't get BobCAD to reduce the number of triangles, it's probably me as I don't work with them often). I actually like the way it looks in simulation with less detailed STLs. It's a lot more "crisp" which is opposite of what I would expect.

Anyway, here are all the files I believe are necessary to simulate a full 4th setup for a 2012 Haas VF-2 with HRT-160. I saved these out for a "standard" setup - the centerpoint of the rotary is at XYZ zero the same way Bob is set by default. I still like using my actual work offsets but that's too much work even for my guys. I favor precision, personally. Of course you can modify them anyway you wish.

I just Zipped up my MachSim folder that had all the files, including the XML Machine Definition. I would NOT recommend just dropping the whole thing into the MachSim folder. It might work (I am very curious if it does) but I can't test it until my new computer comes in.

A question for ChipMaker or Al (or anyone who knows): Can we trade Machine Definitions like that? It would be nice if I could define the whole setup, then just save it to a jump drive and drop it in the MachSim folder of another computer running BCC. I am very open to sharing my files and it would be nice to have a thread here where people can trade Machine Definitions if they choose.

[SBC Cycle]

BTW, standard Simulation users can play along. Before I had full Sim I could run simulation then open a Machine Definition to play with the Axis controls. That's about all you can do with it but that's how I was able to easily create a "real" Sim in a short time. It helps familiarize yourself with how the Sim workflow.

[SBC Cycle]

I have finally accepted that the Machine Definition (MD) is one of the greatest tools ever added to BobCAD-CAM. I recently demoed the V28 software to play with the 5 axis portion. We purchased a Haas UMC to dip our toes into the 5 axis field. We didn't buy it to do simultaneous but it will and it's the most fun I've ever had watching a machine run

No, I spent the five days of my trial trialing the 5 axis tilts and indexing under every different scenario I could think of to dial in my Post Processor. I am very appreciative that BCC allows you to post your code for a full 5 days to test on your machine. In five days I was able to complete my first 3+2 part on the first 5 axis machine I had ever laid hands on, and the posted code only required one simple hand edit (replace G54 with G254 Dynamic Work Offset in the appropriate place). The Simulation proved to be key to understanding how my machine worked by providing a visual aid with real machine coordinates (they match almost perfectly!) that I could test on the machine safely and with confidence that I wasn't going to have a crash. A huge tip of the hat to BobCAD-CAM software for helping me get going so fast on my new machine
:cheers:

In an effort to save other BCC users their valuable trial time on more exciting things (the 5 axis is very cool) I'm going to revive this thread and hopefully help other users recognize the benefits of full Machine Simulation. If you have it and aren't using it you might be missing out on one of the most valuable tools of the software.

Obviously, I do not have V28 and there are some additions in the newer software such as defining which pieces show up in the Simulation workspace. It won't matter as those are just additional tools in the toolbox, everything in this thread should apply to the future versions of the software.

Attached is a screenshot of our new Haas and the start of a Machsim simulation. The next few posts (don't know how soon) will be revisiting the previous machine I built in this thread and how it looks today. Thanks again BobCAD staff for putting out a stellar product.

[SBC Cycle]

Ok, so quick recap. At the time I created this thread I really had no idea of the extent of the relationships between these three things:

a) The part (and its location in the workspace)
b) The Machine Definition (MD)
c) The Post Processor

Obviously we all know the purpose of the Post, and the MD for simulation. What I did not realize is how important the MD would be when it came to posting code for rotary style toolpaths or even simple indexing. The rotation points (relative to your part's origin) become extremely important, especially when the zero of your part is not in the center of the rotation. When everything in the MD is centered up about the Origin you don't get an insight into how the MD can alter the posted code. Start shifting your center of rotation away from the MD Origin and things start looking a little goofy. Now shift your part away from the center of the rotation as well (which I'm learning is quite common with my Table-Table configuration) and it looks completely hopeless and confusing.

In the first post I asked:

One of my first questions/decisions is where Machine Coordinate Z0 should be. Trying to keep everything as accurate as possible, my first try would be to set Z0 up high at the tool change position.... So is this the proper way to start?

There is no real answer to this question, the closest answer is "it depends".

Having the Z Origin of the MD at the center of the rotary (and the part centered in the workspace around the origin) now makes sense as it simplifies setting up the MD by a mile and even if your post is not quite setup correctly you will still get good code. I did realize this early on, even when I asked this question, but what I really wanted to know is if the MD could be setup exactly like my machine. The answer is yes, absolutely, and why wouldn't you?

So why wouldn't you? I can think of plenty of reasons. It overly complicates the programming of even simple parts. It's difficult to setup with no documentation on doing it this way. It introduces unnecessary "room for error". Really there's a lot of reasons not to and it's not the purpose of this thread to change the way you program parts. However...

You will find that it's really not that hard at all to setup a "real" simulation and once it's done properly you never have to think about it again. You may not even have to alter your workflow at all depending on how you currently use the software. When I said it overly complicates the programming of even simple parts that's a nod to the laziest among us (I count myself in there too ). Usually there is only the additional step of setting your Work Offset in the Stock Wizard to your real XYZ pickup points. I approximate the location most of the time as I'm programming the part and if I feel the need to proof it before I hit Cycle Start it takes just a second to update my WO in BCC and run a collision check. I know I'm taking a $2k-12k bet that I'm not going to hit anything when I load code and just stay nearby (not on top of the Slide Hold) as I'm running the first part. That is what made it worth my effort and "saves" me money in the long run. Machsim is as accurate as you want it to be and I will try my best to replay my steps on how I got here:

Attachment 307846

[MikeC8]

Its also a good idea to set work limits for your 4/5 axis MD which will prevent bobcad from being able to post or even simulate a part with G-code which violates the work space.

[SBC Cycle]

Absolutely! My B axis only goes -35 to +110. If it's going to try to flip the wrong way, I would definitely want to know while I'm still at the computer, not at the machine. I like that it forces me to not shortcut.

I had an in house demo a few weeks back from brand X CAM software. I gave them a sample part to program and sure as I'm typing this the first time it flipped the part went B-180. "See how easy this is!" lol. Keep in mind, the Post Processor is a couple thousand dollars on top of the software, built to my machine. Needless to say I was not impressed.

[shaunwil]

Interesting read for sure.

I'm currently investigating purchasing a 5 axis Haas. We have Bobcad v28 in 3 axis pro at the minute, with the option to upgrade to 5 axis pro if my justification for the machine is passed by management.

I've got the v28 demo at home trying to figure round the machine parameters. I can't seem to get it to display the machine in the sim tool.

Shaun

[SBC Cycle]

Make sure you select "Machine Focus":

Attachment 316266

Or Hide/Show components in the Machine tab of Simulation:

Attachment 316268

At any time (with a valid part file open) you should be able to open the Simulation, then open a Machine Definition .xml to play with the moving components:

Attachment 316270

If none of these will show your machine, post back!

[shaunwil]

Thankyou, will have a play when I get home

Shaun