Cancel
Showing results for 
Search instead for 
Did you mean: 
Highlighted

CSE - Simulation Cycle Times

Creator
Creator

I am trying to get my simulation time to match my real time at my machine, i have matched the rapids, accelerations, deceleration but it seems to be way off when doing multi axis cuts. My post outputs TCP. Is CSE not able to calculate this accurately

22 REPLIES 22

Re: CSE - Simulation Cycle Times

Siemens Phenom Siemens Phenom
Siemens Phenom

Hi,

 

first of all please add requests about CSE Simulation to the general discussion forum or the private partner Forum

https://community.plm.automation.siemens.com/t5/Discussion-Forum-NX-Manufacturing/

https://community.plm.automation.siemens.com/t5/Private-Siemens-Partner-NX-Manufacturing-Forum/bd-p/.... The postprocessor forum is for requests about Post Processor only.

 

Back to your request... CSE IS able to calculate machining times accurately. First of all I would like to know what do you mean with "simulation time"? What numbers in NX CAM Simulation do you use for comparison? Which NX version do you use?

 

In general if the machining times in simulation are different the real ones this can have different causes in different areas. A first starting point are the axes properties in the kinematic model. Have you filled them with the real controller settings? Keep in mind there are some other properties (KV, jerk, ... ) which need to be filled to increase the accuracy. What to do mean with "... i have matched the rapids, accelerations, decelerations..."?  In which time area (primary, secondary) differs the machining time?

 

Sorry for all the questions but the answer is not quite simple ;-)

 

Thomas

Re: CSE - Simulation Cycle Times

Creator
Creator

X-AXIS SPECS.gifMaster X speed.gifThank you Thomas for the reply. So i am using NX 11.0.2.7. I have two problems really, so the machine time i get from my post processor which is $mom_machine_time is one on time, the time that shows on the control panel in CSE is another time and the time i get at the machine itself is another, they are all different, for example, the part i am testing on is as follows

Time from Post Processor = 1.8 min

Time from Control Panel = 4:49

Time at the Machine = 3:15

I have accounted for toolchanges in both the post and CSE, i have also matched the dynamics of the machine although i didn't know what the KV was. i' attached a snap shot of the machine parameters of my Osai controller and a snap shot of my machine dynamics in Machine Builder, note i'm only sending you the X but i have done this with all axis

 

Re: CSE - Simulation Cycle Times

Siemens Phenom Siemens Phenom
Siemens Phenom

Hi,

 

here is an explanation of the KV factor. I don´t know which parameter in your list covers this setting. But this factor has also an influence to the time calculation.

2018-10-31 07_54_55-Microsoft Edge.png

 

 

 

Another question what I have is. I´m not really an expert of the OSAI control and it´s commands. Are there any compressor modes or exact stop command calls in the NC code like the G64 in Sinumerik. I also don´t see any numbers for the lookahead which also can have an influence to the simulation time. Maybe they are not implementend/evaluated in your mcf/ccf? 

 

Do you have a chance to run the simulation in NX1202? In there we have a new Time Analysis feature. With that you´ll have an overview about alle times in cse. With that we can easily figure out in which area we´re losing the time.

 

About the deviation of the post processor time a can´t say anything to that. Maybe an Post expert can add some words here.

 

Thomas

 

Thomas

Re: CSE - Simulation Cycle Times

Gears Esteemed Contributor Gears Esteemed Contributor
Gears Esteemed Contributor

I know (at least the legacy) drill cycles are (or were) way off in time

When a G81 cycle takes the same time as a G83 cycle, all other params equivalent, you know something's wrong ;-)

Ken Akerboom Sr CAx Systems Engr, Moog, Inc.
Production: NX10.0.3.5 MP16/TC11.2
I'd rather be e-steamed than e-diseaseled


Re: CSE - Simulation Cycle Times

Creator
Creator

Ken,

as far as the drill cycles go, I only use the Drill and Peck cycles so i had the post changed to always output single moves just so i didn't have this problem

 

Thomas,

I sent your email to our OEM so find out about the Kv. i am not sure as to when we will be installing NX 12 here, our IT department said early 2019 but that was 6 months ago

Re: CSE - Simulation Cycle Times

Creator
Creator

Here is the response i got from our OEM. He made a mistake thou, null motion threshold is set to .01 mm on all the axis. So i played around with the settings and changing the Kv didn't seem to change the cycle time but only the way it displayed the motion. I doubled the Jerk limit on all the axis and got a faster time from 4:48 to 4:04, at the machine it is 3:15. I case this makes a difference, i am doing linear cuts (3D_Profile) holes (non-drilling cycles) and 5-axis cutting (Contour_Profile)

 

John, the part program is running in G27 mode, not G29 (stopping at every end of block).

 

Also G09 is for exact stop at end of block, only for one block.

 

G29 is modal, i.e. not changing the mode until G27 is programmed, same for G27.

 

There's a parameter to control the position tolerance at end of block before to go to the next.

 

It should be null motion threshold (0.001 mm normally).

 

You can find the Osai Open parameters manual here:

image.png

Re: CSE - Simulation Cycle Times

Siemens Phenom Siemens Phenom
Siemens Phenom

Hi,

 

thanks for the feedback. As you can see there are a few things which needs to be considered when comparing the real machining time with the simulated machining time. 

 

Now my question is how is the G27 and G09 implemented in the ccf/mcf file you use. When it´s doing nothing then you have to customize it. Another point could be, as already discussed, the tool change time. Maybe this could also be an area where you´re lose the time. 

 

Let us know if you have further questions.

 

Offtopic: I´m not sure if its possible but can you try to move this thread to the general manufacturing forum? Because this discussion and the information inside is very valuable for others too. IF you´re not able I will ask the admin if he could shift that when its ok for you.

 

Thomas

Re: CSE - Simulation Cycle Times

Creator
Creator

Thomas, can you please explain to me what you mean by ccf/mcf, is this the dynamic settings in Machine Builder or settings in Machine Configurator, if its Machine Configurator then i was told that those settings are not used anymore and that I should be using the ones from Machine Builder

 

Also, i don't know how to move this thread but i'm ok with you having someone move it. Thanks

Re: CSE - Simulation Cycle Times

Siemens Phenom Siemens Phenom
Siemens Phenom

Hi,

 

the axes properties (acceleration, deceleartion, ... ) are settings within the kinematic model. As you already have shown It looks like that you setup them as on the real machine. 

I talked about the used CSE functions with can influencing the motion profile. I don´t know which commands your controller uses to influence the motion profile. I only know the commands in Sinumerik Control G64, Soft, Brisk etc.

 

https://support.industry.siemens.com/cs/mdm/109752348?c=103778667275&t=1&s=Brisk&lc=en-US

 

The usage of these different motion profiles can have also an impact to the simulated machining time. I´m not really familar with OSAI control but I´m pretty sure that this controller does also have different modes for it. I´d suggest to figure out which modes uses the controller and which are active on the machine or machining type and then so setup the cse.

 

CSE commands are: SetDefaultMotionProfileType, SetMotionProfile 

 

Thomas

 

Learn online





Solution Information