In Mill Expert, I am not having much luck getting a useful estimate of the time required to machine on the job sheet. Anyone got any useful tips on this?

Are you having good luck? Maybe its because of my post processor selection? You would think that getting the length of the various chains, and multiplying by the applicable feedrates would be child's play?

Hu,
I've had real good luck with this feature. I recently ran a job that had 6 tool changes, used the rest robot, and the time OneCNC calculated was 8 minutes and 26 second. Actual machine time was 8 minutes and 42 seconds. I attribute the difference to tool change time and that OneCNC figures rapid moves at 10000 (which is cofigurable). So I should change this. Just need to figure out what to change it to.:confused:

Hu,
Just a thought, and maybe you have already tried this, but have you tried running your program thru a standard post, say Haas or Fanuc? Maybe the custom post you have dosen't know how to handle all you are throwing at it. If it is better time wise thru a different post then you could use that for estimation purposes.

The thing is that the job sheet is independent of the post selection, that is you can pull up the same job sheet regardless of what machine you want to use.

So it seems to me that the machining time should be a function of the CAD chain length multiplied by the feedrates.

I've tried many different values in the NC setup Machine rapid speeds and this value seems irrelevant at this stage, too. The job sheet gives exactly the same time, no matter what I put in.

Give you an example: I did a planar finishing pass at at stepover of .005 and a surface accuracy of .001. producing 47000 lines of code. This was clocked to be 6 hours and something.

Then, I redid the finish pass again at a stepover of .004 and a surface accuracy of .0005, producing 69000 lines of code, but the machining time is only 3.5 hours.

It sounds like some kind of a math glitch in the internal formula.

HU,
I believe your right. The machine time appears to be tied to the cad drawing not the post.

I did a test and here are the results:

z-level rf at .025 depth of cut: time 6 hours 40 minutes.
same at .050 depth: time 5 hours 42 minutes

then radial finish at feed of 24.: time 5 hours 42 minutes
same at feed of 40.: time 5 hours 9 minutes
then same with feed at 80. : time 4 hours42 minutes

So it appears that my system (or circumstances) is at the very least making changes to the time sheet.

Hu,
I just did another test with Planar finish to make sure that was ok also.

Planar finish feed at 15 : time 2 hours 29 minutes
same as above feed at 30 : time 1 hour 14 minutes

Just as it should be. About 1/2 the time by doubling the feed.

I don't know what to tell ya. You cuold e-mail me the file and I'll run it thru my system and see it it differs from your figures.

I took the tap file and checked it with another program and it seems to give me the same results as Onecnc's job sheet. So it must be correct.

I must have misjudged the parameters I was using that I spoke about above, because I cannot reproduce that type of discrepancy again.

Also, I guess that maybe I am just expecting it to take longer because of the way I used to cut these molds. Beforehand, I didn't dare use as large of a cutter as I am now, because I was fearful of leaving too much material in unknown areas. Gotta hand it to those Onecnc guys, the Rest Robot is a cool feature and takes some getting accustomed to. :)

Hu,
That Rest Robot is a real time saver, yes indeed!
Glad you got the time thing figured out. I thought I was loosing my mind. Well maybe I am anyway.:D

Results of running the gcode program: the machining time is basically correct as output by the job sheet. So there is no problem after all.

Call me doubting Thomas, I guess. :D