topic Re: Cycle time estimate in HSM Forum

Cycle time estimate

gcolomboTS6NW — Thu, 02 Sep 2021 07:10:59 GMT

Hi guys,

I know this is a topic that has been discussed in the past many times, however I cannot seem to find anything more recent than 2017 about this.

I'm having issues with the cycle time reported in the simulation. I understand that it is potentially caused by the wrong or a default rapid speed used by HSM. I did create a new machine, which has the rapid speeds that my Doosan DNM has, however that did not produce any chance in cycle time at all.

It is pretty important for us as we run very long cycle times and we try to organise them so that a cycle does not finish in the middle of the night when there's nobody around to do a changeover.

Thanks in advance for your help on this.

Re: Cycle time estimate

FTKnur — Thu, 02 Sep 2021 07:44:40 GMT

Hi @gcolomboTS6NW,

@gcolomboTS6NW wrote:
...
I'm having issues with the cycle time reported in the simulation. I understand that it is potentially caused by the wrong or a default rapid speed used by HSM. I did create a new machine, which has the rapid speeds that my Doosan DNM has, however that did not produce any chance in cycle time at all.
...

You can change the rapid speed in the "process time".

"Eilgangvorschub" is rapid feed 🙂

But your problem is IMO a complete other one...

A line with 10000 mm and a rapid feed of 10000mm/min should take exactly 1 minute.

If you have a Zig-Zag line with completely length of 10000mm your machine need to brake and accelerate at each point. If your machine is a real heavy weight, the time can go up to 2 minutes or more.

We have Heidenhain (530 and newer) controls and the estimated time from the simulation fro mthe control is more exact then the HSM time.

Re: Cycle time estimate

gcolomboTS6NW — Thu, 02 Sep 2021 08:04:59 GMT

Hi @FTKnur ,

You hit the mark right away, I did not know the existence of this "process time". As you pointed out by changing the rapid feed the time changes accordingly.

I will test and see how close it gets to the real cycle time.

In regards to acceleration and deceleration you're correct the machine will slow down and accelerate at each change of direction, I did however notice that in case of toolpaths with very few rapid moves the cycle time is usually quite accurate, so I think I'm safe to assume that in my case the biggest contributor to the error should be the rapid speed. It was set at 5m/min and we normally run it between 9m/min and 18m/min.

Now the real question is why is it possible to specify rapid speeds in the machine definitions if then this information it's not used.

Thanks again for your help @FTKnur

Re: Cycle time estimate

FTKnur — Mon, 06 Sep 2021 06:23:51 GMT

I cant help you with the "Machine Setup". I don't use it. But perhaps you can try to use in the Linkin Tab not the rapid move. Don't know if this change the time in a setup sheet.

In the Post Processor is the rapid feed not defined ... so you have to try it with "Always use high feed". You should use this only to get the Time and not use this for the G-Code 😕