Estimating cycle times

[David Friedman]

Estimating cycle times has been a big problem for me. I usually run the program with the door open (so that the beam stays off) to get the time. For most raster engraving the times can be accurately obtained by the size of the drawing. Vector cutting is another story. This process can take a while and tie up the machine at the same time, but inaccurate estimates can really cost you a great deal!

However, a person can only wait 40 minutes for an estimate a certain number of times before reaching out for another solution. I have exceeded that limit and here is the solution at which I have arrived. A Visual Basic routine for Corel 12 that measures the length of all curves in the drawing and displays the estimated time. Now the same 40 minute vector file can be estimated in 5 seconds.

Be aware that:

1) all possible objects will be converted to curves
2) outline thicknesses are ignored (fat curves are counted)
3) grouped items are ignored
4) you assume all responsibility for your use of this routine
5) times are coded for 6% speed but can be adjusted (all IPMs are commented into the code)

The file is in GMS format and only requires copying to your GMS directory and Corel to be restarted to be available. To adjust the calculation for a different speed (I use 6% quite a bit for 1/4" acrylic) look for the 14th line in the file (it says "dSpeed = 12.12121") and change the number to one from the list below that line. Note that these numbers were developed on my laser and are probably different than yours.

Since I'm still not quite sure how to attach a file I have made it available at: http://www.creativeproco.com/documents/curvelength.gms. (http://www.creativeproco.com/documents/curvelength.gms)

Good luck with that.

Dave

[Mike Mackenzie]

David,

This is a great tool however it would really help if we no what machine you are using due to the different speeds and powers of laser systems.

[David Friedman]

Mike:

I have an Epilog Legend 36EXT. All of the values that I included in the comments are measured IPM (inches per minute) values. You must do your own evaluation of the IPM values on your machine in order to ensure accuracy of the estimate. I did a 10" x 10" square at each of the different speeds to develop those numbers. For example, if the cycle takes 1 minute then the speed is 40 IPM (the distance is 40 and the time is 1, or 40/1). If the same cycle takes 2 minutes and 30 seconds then that is 40/2.5, or 16 IPM.

The power setting will have no effect on the cycle time.

The display at the end can be easily changed to show more than one speed and cycle time. Let me know if you want some help on that.

Dave

[Michael Kowalczyk]

Hi David,
Thank you for sharing. Do you know how this would line up with a Trotec Speedy II that rasters at 140IPS?

Thank you,

[David Friedman]

Michael:

No, sorry. You'll have to run your own time tests to get the IPM values for your machine. I wouldn't count on any two machines being the same within 5 to 10 IPM at full speed.

Dave

[Mike Null]

Michael:

If you have job control on your Trotec it will give the the cycle times as soon as you complete a job. All you need to is identify them for easy reference.

[David Friedman]

Michael:

The idea here is to find the time before running the machine so that you can provide an accurate quote. Before I wrote this I would do exactly that... run the machine with the door open to get the cycle time. I don't want to quote 50 or 100 pieces and be off by several minutes each piece!

Dave

[Mike Null]

David:

I recognize your purpose, my point was that if Job Control is part of the Trotec software there is a job history to refer to. That history might be used as a guide for estimating based on similar jobs.

[Rodne Gold]

Generally vector and raster speeds are linear , so set your vector cut to lets say 50% speed or 80% or whatever , do the dry run (which should be real quick) and extrapolate down.
IE if you set 50% and your actual speed would be 5% , you would take 10x longer than the 50% speed.
There are a lot of ways to cut down cycle times as well with more than acceptable results depending on what you are doing. By using various strategies , we have often cut down cycle times by 1/2 or more and this is seriously significant when you doing volume work.
Merely measuring the length of curves and using the ips speed is not going to be very accurate due to the ramping function of the head as well as acceleration/decelleration etc. The rated IPS speed you use is most likely the maximum achiveable and wont be near an average speed , you would have to apply an averaging factor by using observed time/calculated time. Another way to avoid all the curve handling etc would be to select all hairline lines for your calculations which is very easy , instead of using all curves , I think all lasers will vector hairlines by default.

[Michael Kowalczyk]

Michael:

If you have job control on your Trotec it will give the the cycle times as soon as you complete a job. All you need to is identify them for easy reference. Hey Mike, I use the history quite often to check but as David has said his program may give an approx time with out having to run the laser. I have a stand alone program that will be pretty close for my CNC's but also have a statistics button in ArtcamPro that will give me a consistently close estimate. On big jobs, 100's to 1000's, of small parts or products like key chains I just run one sample and multiply it out. Experience helps you know set up time, change over of material, cleaning lense if it is a long run in acrylic, MDF, or stone. When doing photos most of us know about how long it will take but on the custom stuff that may be different sizes that's where this would me an enormous help. I also have a digital stop watch at the machine to get precise times and then carry it with me as I run parts that may take 30-40 minutes a run so I can do other things and be twice as productive. Only when it is running proven material that won't flame up. Thanks,