I use SprutCAM for CAM and I generally create primitive shapes for no-go zones like fixturing. Like Tom, I also try to set the safe height above any fixturing. I usually set my safe height at some nice round number so I can verify my tool height easily on the first run. One thing that I have run into in the past was hitting fixturing with the collet chuck so I try to be generous with my clearances.
I only use models of fixture if I know the tool paths may be close. I use HSMWorks for CAM it has tool path simulation so if you setup the tool and holder it will show where there may be interference. I made my own models of my table, vise(s), fixture plates and specialty clamps or fixtures. Tormach has 3D models of 6" & 8" rotary tables & tail-stocks (4th axis) that are identical to the Phase II. Sometimes you can find 3D models on GrabCad.
Sometimes I draw in the fixturing / clamping. Normally I plan the tool path and set up to account for the clamping. If a job has more than a couple of operations, then I machine the whole job in my head before I ever make a chip. I usually try to design parts that have provisions for hold down bolts, in fact where possible, I make a lot of parts that have holes that line up with my T-slots. I have a drawing of my mill table and overlay the part to figure out the clamping. With a CNC, many times you can rotate the part to get existing holes to align with the T-slots.
Where possible, I normally drill all of the holes in the first operation to have something to through bolt with. I also use MDF as a spoil board. You would be surprised at how well deck or sheetrock screws hold. You can also drill & tap MDF for hold down. You can also make a part shaped pocket in the MDF about 1/8 or more deep to fixture the part once you have done the outside profile. That way it requires minimal hold down.
Rarely a vise. It's seldom in the way or even much of a player.
Always the clamps. I've heard of guys that have run into clamps - though, of course, no one i've ever met, right? Always the clamps, especially when there's a re-clamping operation. Gotta keep track of the clamps. They're easy to forget... uh... so i'm told.
All the hold-down tabs and hold-down bolts get modeled or coded. In fact anything that's not going to machine as easily as the part gets represented.
I normally just mentally put them in when I'm setting up the tool path in CAM, but if I need them on the drawing, I just draw a quick one. As @wrat says, normally the vice is not a factor except for height off of the table. I also normally have soft jaws in my vice, and will machine into the jaws if needed.
Three lines is a clamp. A circle is a bolt.
Needn't be rocket science.
I know the Big Boys all have complete McMaster-CarrLane-DeStaCo model catalogs, especially for Vericut and high zoot software titles like that, but I just rough something in, and only when needed.
Besides, there is a drawback. When one uses a solid model or some very nicely detailed clamp/hardware data, the temptation exists to actually believe it's true. Clamps are not precision items of precision placement. They tend to wander at the whim of the operator. They get replaced without telling anyone. They grow and shrink without notice, as it were.
More than one have been crashed and an argument ensued because it wasn't "where it belonged".
If it's not going to be exactly right to begin with, i tend to not sweat the detailing of it and just setup a "stayout zone" where it sorta-oughta be. JMHO.
Thanks guy. I'm slowly getting more in to CNC. I do have Inventor HSM so I've been playing it. Totally different than Creo/Pro-E. I just to prefer until I get more experience to have the model laid out how the layout would be on the machine. Yes, it's more work than I need to do but it gives me added CAD time in learning the software further.
I'm lazy and I usually just use sketches for my keep out zones. That said 3d content central, McmasterCarr, Carr Lane are all good places to start. Also any good fixture company will have CAD models that you can download.
We have a Haas TM1 used for a number of recurring parts one of which is the string termination part for bass electric guitars. The first program was finger CAMed at the machine and took less then 2 hours of spindle time each. The machine had a control problem and all programs were lost.
They then bought one seat of CAM for Solidworks, the first time the program ran the tool crashed right into the fixture, after this was corrected the program took nearly 4 hours to run the same part with no visible improvement in surface finish or accuracy.
It is not important to model the workholding, it is important to know where it is however.