2 basic questiions on coolant and CAM

[SCzEngrgGroup]

Interesting. The code is basically correct, but the passes do appear longer than needed. It clear does understand where that upper-most corner is, or it wouldn't generate so many step-downs. The stock display in simulation mode is just wrong. That is clearly a bug in HSMXpress. You should report that.

Regards,
Ray L.

[Hirudin]

Why not try a solid model for your stock? It's a less error-prone method, and has other advantages too.

What is the sketch I circled in the screenshot? The incorrect stock shown in your other screenshot seems to be an extrusion of this sketch. Getting rid of it might fix this issue.

If you're using multiple jobs you don't want to define the stock each time. Use the option that's called something like "use stock from previous job".

[Hirudin]

Yeah, looks like HSMWorks has a bug that incorrectly displays the stock when you do a normal simulation. My assumption that the sketch might be causing the problem was apparently wrong. Sorry about that.

If you do a "stock simulation" the stock will (probably) appear correctly.

[mountaindew]

I use sprut9 and would have no problem setting this part up with ucs and operations on every side complete with machine stock, fixtures and tools for every feature.
Then again what do I know I only been doing this for a year.
md

[EvolveES]

Here is a quick and dirty way i have been doing things like this. We do a lot of one-offs, so it is really more about getting the machine running than having the most elegant or correct code. I am sure this is NOT the best or "correct" way to make a part like this, but it has always worked for me. I make a quick secondary part called "machine fixture" or something like that, and then make an assembly with it and then insert the angled part i want to cut. You can then either use an angled mate, or you could have made your "machine fixture" part as a 3D wedge shape with a 30 degree angle, then just use normal mates. Then when you convert your SW model to a STEP file to use in your CAM software it already has the part referenced how it will be in the mill, the correct side for machining is up and ready to be selected for an operation.

Another thing i often do is to model in small features that will provide referencing marks in the CAM software, usually it is a small shallow hole or boss (usually 0.05" diameter x 0.05 tall/deep), that way in your CAM software you have a quick center to place the origin on. these features will not be in the final part and just serve as reference marks in CAM.

ans yes this way will make your stock look funny, but it really doesn't matter what the CAM looks like as long as YOU know what your part will look like. There are a lot of offsets and transformations you can do in CAD and CAM but i have found this way usually gets me to the machine faster and with less mistakes (hitting the wrong key or entering the wrong value etc. - but i am a very visual person).

Hope this offers some insight and good luck with it

[popspipes]

I drew a similar part in Rhino, rotated it 30 degrees so that the angled plane was perpindicular to Z, imoprted into sprut and machined the flat plane and the pocket and hole, then flip the part and counterbore the hole using the bottom of the model angled to the Z axis 30 degrees and indicating off the hole for alignment.

It is much easier for me to do the plane changes in CAD rather than in Sprut, just my personal preference.

[widget_maker]

Why not try a solid model for your stock? It's a less error-prone method, and has other advantages too.

What is the sketch I circled in the screenshot? The incorrect stock shown in your other screenshot seems to be an extrusion of this sketch. Getting rid of it might fix this issue.

If you're using multiple jobs you don't want to define the stock each time. Use the option that's called something like "use stock from previous job".

That's the plane I created parallel to the 30 degree face so that I could create the new coordinate system. I am not experienced enough to know if I actually need it - perhaps I could have created the coordinate system off the face? I will play with it and see.

[widget_maker]

Yeah, looks like HSMWorks has a bug that incorrectly displays the stock when you do a normal simulation. My assumption that the sketch might be causing the problem was apparently wrong. Sorry about that.

If you do a "stock simulation" the stock will (probably) appear correctly.

When I do a simulation and enable "show stock" the incorrect stock appears and gets machined.

[widget_maker]

EvolvES and popspipes

thanks for the ideas. Yes, I have a lot to learn, I'm wondering if I'll ever get weaned off the manual mill at this rate!

[popspipes]

I have an old 1952 vintage Bridgeport, I have only owned it for around 20 of those years, and it was a no brainer for me, I use the Tormach almost exclusively, There are times I still use the old Bridgeport but they are rare these days.....

Getting the cad cam under your belt will make it much easier.

[widget_maker]

I removed the plane and referenced the 30 degree WCS to the 30 degree face. The plane was not necessary, however, the simulation did not improve. The stock I defined is being ignored and a stock box as shown before is still being generated. The simulation shows the tool coming down to the surface and taking a single face cut - and way past the 30 degree face so it really thinks there's a stock box positioned horizontally, not coplanar with the desired final shape.

Showing stock as defined (yellow)




Shows stock the CAM thinks exists (green)

[SCzEngrgGroup]

You do realize that is a harmless bug? The only effect is the toolpath is longer than it needs to be, but the part will be cut correctly.

Regards,
Ray L.

[widget_maker]

Ray,

I don't know anything, but it looks like the facing operation is taking about a 1" deep cut because it thinks there's no material there when there is?

[SCzEngrgGroup]

Ray,

I don't know anything, but it looks like the facing operation is taking about a 1" deep cut because it thinks there's no material there when there is?

No, if it's taking a 1" deep cut it's because you haven't set the step depth properly. Go back and look at the screenshots I posted. It was doing the correct number of 0.1" step-downs. It KNOWS where the top of the stock is, it's just got the shape wrong, so it does some air-cutting to remove material that really isn't there, which is harmless.

Regards,
Ray L.

[widget_maker]

But it doesn't know where the top of the stock is - it thinks it's the green stock which is planar with the top of the finished part. The actual stock is the yellow block in the top picture above. The first thing the simulation does is come straight down to where it thinks the top of the block is to take a facing cut and that is 1" or so below the top of the actual stock in yellow. The cutting in air doesn't bother me - it's the fact that it plows through the top inch or so of the actual (yellow) stock thinking it is taking a light facing cut off the green block which doesn't exist.

At least that's how I see it........

[SCzEngrgGroup]

Again, look at the screenshot here: http://www.cnczone.com/forums/tormac...ml#post1653164

It is taking multiple 0.1" passes, starting just below the corner of the stock, as it should. If you're not getting this, you have something in the operation setup wrong.

Regards,
Ray L.

[Hirudin]

When I do a simulation and enable "show stock" the incorrect stock appears and gets machined.

Do a "stock simulation".
If you don't know what I mean, watch the video I posted.

I implore you, watch the video. It shows going from your part, to creating the assembly, to creating the stock model, to setting up a coordinate system, to creating the toolpaths, to confirming the bug, to doing the stock simulation (where the bug isn't present), to optimizing the tool paths. There's even a couple other things in there that might be helpful.

You'll also see that the stock bug has no bearing on the length of the toolpaths. At least, not on my computer.

[widget_maker]

Do a "stock simulation".
If you don't know what I mean, watch the video I posted.

I implore you, watch the video. It shows going from your part, to creating the assembly, to creating the stock model, to setting up a coordinate system, to creating the toolpaths, to confirming the bug, to doing the stock simulation (where the bug isn't present), to optimizing the tool paths. There's even a couple other things in there that might be helpful.

You'll also see that the stock bug has no bearing on the length of the toolpaths. At least, not on my computer.

I watched it - thanks very much. What I need to try to understand is why your "phantom" stock box extends all the way to the leading edge of the "real" stock box, whereas my phantom block only extends to the finished 30 degree face of the real stock block. Yours knows that the "real stock block" material is there to be cut away, mine doesn't!

[Hirudin]

I see. Do you have multiple jobs? I tried messing with your model and ran into the same problem. I got it to work, I HATE having to work around bugs in software though! Argh.

Basically, what I did was just define the stock again in subsequent jobs. If you're using a solid model, just select it. If you're using a sketch, extrude the sketch again. Checking the "Continue machining from previous job" is suppose to cover this scenario, I think, but it doesn't seem to make a damn bit of difference. Also, only the "Stock simulation" mode showed the stock correctly for both jobs.