Improving elapsed time estimates...

[harryball]

Is there anyway to improve the estimated time to cut a file? I ran sheet of address plaques with an estimate of 1 hr 50 m. I expected a couple of hours... maybe 2 hr 30 m but it took 3h 20m.

What puzzles me is the software knows the cut speeds, the ramps speeds, the jog speeds... and it can't get any closer than that?

Is there some setting or adjustments I can use to improve it or should I ignore the estimates?

Robert

[David Iannone]

Rob, I hear you loud and clear. I have been tooling in V-carve lately for all my projects. The estimated cut time in v-carve is also not acurate. I did not know sb control software had a estimated time too. It would be VERY helpfull to estimate machine time more accurate, as I am running a PRT Standard and sometimes have log jams.

Dave

[fleinbach]

When estimating cut time there are many variabales to consider, one of them is needing to slow down to make a turn. Vcarve and Artcam have a feature called Scale factor. Run a file and see how long it actualy takes, then enter differant factors until the time comes close to actual. If you are making simular cuts use this factor.

It is still somewhat of a guessing game because the new file needs to have a simular number of straights and curves for an acurate estimate. However if your next file has lots more turns to make just raise the factor as the tool needs to slow down more often to make the turns. It will take practice to get a closer estimate of run times but you will be able to come a bit closer after awhile.

[harryball]

I abandoned the estimater in Vcarve for anything accurate. However, I will check out the scale factor and try to tune it.

When you run a file in the control software in preview mode, it still creates a log file. In that file it records EST time. I thought "Ahaaa! that's a better way to do it since the control software knows everything and cuts the exact same file for real." That does not appear to be the case.

The control software should have the ability to be very accurate. I understand loading, counting steps and actual speeds get involved... but I don't understand why it should be off almost 40%. I re-ran some files in preview mode that I have cut for real and the est time seems to have very little to do with real time. Surely it can be tuned or improved?

Robert

[David Iannone]

Frank,
Thanks for the tip on scale factor. I wondered what that was for. It makes sense though now that you explained it.

Dave

[patricktoomey]

I was also curious why a very accurate estimator couldn't be made. I run Mach3 now and I can feed it a file and have it simulate the run to get the true time. That time is bascially exact since it runs the file as fast as it can but counts exactly what is happening including all ramping. The only reason that I can see that accurate estimates aren't available from toolpathing software is that they don't know what the controller will do as far as accelerations. For example, I can dial in all sorts of different accelerations depending on what I'm cutting. That can affect overall cut times dramatically on complex 3D files with lots of Z movement. Unfortunately, the only answer I'm aware of is the time scale factor to allow you to fudge it. Having to alter it by hand for each different cut type is a pain but after awhile you get a feel for what factors work for different types of files.

[davidp]

Run the file in preview and it will fairly accurately give you a time. I have found it is usually within 2 or three minutes even over a file that takes over two hours'

[harryball]

An accurate estimation of cut time would be nice. David, I don't know what version you run but when I run a file in preview I get a time estimate of 14:46 When I run the file I get an actual time of 23:07 This was on the same system just in case my working PC was somehow different.

Is there someway to tune the preview mode for more accurate time estimates? The vcarve tuning only works when you are working in... well... Vcarve Pro.

Robert

[ffaelan]

I'm not sure about the accuracy of the preview time estimate myself. I find that it changes drastically depending on if I just run it in preview, if I run it with the simulation box check or if I run it with the show machine box checked. In the end I usually run three or more previews then take an average and add 10%. That's about as close as i can get it.

[erik_f]

Try putting your scale factor at 1.361. This was based off the two times you posted.