anafranil online
IPB

Welcome Guest ( Log In | Register )

 
Reply to this topicStart new topic
> TCPM and state parameters for 5d
ka67_72
post Mar 27 2015, 01:35 AM
Post #1


Member
**

Group: Members
Posts: 22
Joined: 31-May 13
Member No.: 3,531



I started testing 5d in foam tonight. It's working but something's way off. Could you point me in the right direction for getting the control parameters, TCPM and zero point settings correct? My machine is an old Matsuura with Yasnac I80m controls. It has a Tsudakoma rotary tilt. There is no compensation built into the machine. G54 is dead center on the surface of my rotary table. When A rotates G54 stays in it's original location. The machine just does what it's told. I need Sprutcam to do the work.

Thanks,
Kevin
Go to the top of the page
 
+Quote Post
Sprut_UK
post Mar 28 2015, 10:07 AM
Post #2


Advanced Member
***

Group: Administrators
Posts: 1,091
Joined: 12-December 07
From: United Kingdom
Member No.: 4



Hi Kevin,

I am not familiar with that machine or control.
I'm not sure what you mean by "there is no compensation built into the machine", do you mean that it does not support TCPM? TCPM only corrects for parts that are not positioned at exactly the same position that they are programmed for.
It also depends on how your postprocessor is setup to work with your machine? Is the postprocessor specifically for your machine / control combination?
What machine configuration are you using in SprutCAM?
Does the CNC control compensate for the 5D tool length?

You can find out more information from the SprutCAM manual. The V9 manual is available online here: http://www.sprutcam.com/files/documentatio...l?html__352.htm
I would suggest looking at the section called '5 axis machining' to see what options are available.

Kind regards

Dave




--------------------
"Never interrupt your opponent when he is making a mistake..." - Napoleon Bonaparte
www.sprut.co.uk
Go to the top of the page
 
+Quote Post
ka67_72
post Apr 1 2015, 01:33 AM
Post #3


Member
**

Group: Members
Posts: 22
Joined: 31-May 13
Member No.: 3,531



Sorry for the slow responses. I'm not getting notifications. The machine just has tool length and diameter offsets. No TCPM. The postprocessor is a slightly modified Fanuc. What I'm getting is the correct shape but exaggerated. B seems to be rotating about twice the angle it should be.

Thanks,
Kevin
Go to the top of the page
 
+Quote Post
Sprut_UK
post Apr 1 2015, 05:47 AM
Post #4


Advanced Member
***

Group: Administrators
Posts: 1,091
Joined: 12-December 07
From: United Kingdom
Member No.: 4



Hi Kevin,

That does seem a bit strange. Have you checked the B axis angles in the CLData displayed in Simulation mode against the B axis angles output by the postprocessor? If they don't match then the postprocessor needs looking at.
Are you able to provide a project with postprocessor for analysis?
Does it work correctly when index machining (3 + 2)?

Dave

PS. there was a problem with the e-mail notifications on the forum. Hopefully this has now been corrected.


--------------------
"Never interrupt your opponent when he is making a mistake..." - Napoleon Bonaparte
www.sprut.co.uk
Go to the top of the page
 
+Quote Post
ka67_72
post Apr 1 2015, 12:31 PM
Post #5


Member
**

Group: Members
Posts: 22
Joined: 31-May 13
Member No.: 3,531



E-mail notification worked.
It works great in 3+2. I'll check the CLData vs output when I get home tonight.

Thanks,
Kevin
Go to the top of the page
 
+Quote Post
ka67_72
post Apr 1 2015, 10:47 PM
Post #6


Member
**

Group: Members
Posts: 22
Joined: 31-May 13
Member No.: 3,531



Dave,
B is fine. It's X and Y that are out to lunch. In the attached pic, what's milled in foam should just be one of the bellmouths of the piece next to it. I also posted a screen shot comparing the CLD to Post.

Thanks,
Kevin
Attached File(s)
Attached File  5d_CLD_vs_Post.jpg ( 440.64K ) Number of downloads: 14
Attached File  IMG_7441_50.jpg ( 1023.17K ) Number of downloads: 5
 
Go to the top of the page
 
+Quote Post
Sprut_UK
post Apr 3 2015, 02:14 PM
Post #7


Advanced Member
***

Group: Administrators
Posts: 1,091
Joined: 12-December 07
From: United Kingdom
Member No.: 4



Hi Kevin,

Yes, the X & Y moves from the postprocessor aree not matching the CLData. Also the Z axis movements are missing completely.
It looks to me like the postprocessor MULTIGOTO routine is the place to start searching for the fault The problem could also be related to a postprocessor sub-routine.

Hopefully this will give you some idea's as to the cause?

Dave


--------------------
"Never interrupt your opponent when he is making a mistake..." - Napoleon Bonaparte
www.sprut.co.uk
Go to the top of the page
 
+Quote Post
ka67_72
post Apr 7 2015, 12:41 AM
Post #8


Member
**

Group: Members
Posts: 22
Joined: 31-May 13
Member No.: 3,531



Dave,
Here is a screen capture from my postprocessor. Looking through all the CLD from my previous 3 axis and 3+2 projects none of them use MultiGoto. 3+2 works great. Does any of this look abnormal. I just need G1 X,Y,Z,A,B.

Thanks,
Kevin
Attached File(s)
Attached File  MultiGoto.jpg ( 237.89K ) Number of downloads: 9
 
Go to the top of the page
 
+Quote Post
ka67_72
post Apr 7 2015, 12:43 AM
Post #9


Member
**

Group: Members
Posts: 22
Joined: 31-May 13
Member No.: 3,531



This is a screen capture of the settings I was asking about in the first place. Do I need to have any of these check boxes checked? What about LCS and Euler settings? 3+2 works fine. I'm also not getting e-mail notifications again.

Thanks,
Kevin
Attached File(s)
Attached File  5axis_params.jpg ( 295.94K ) Number of downloads: 3
 
Go to the top of the page
 
+Quote Post
Sprut_UK
post Apr 7 2015, 10:32 AM
Post #10


Advanced Member
***

Group: Administrators
Posts: 1,091
Joined: 12-December 07
From: United Kingdom
Member No.: 4



QUOTE (ka67_72 @ Apr 7 2015, 01:43 AM) *
This is a screen capture of the settings I was asking about in the first place. Do I need to have any of these check boxes checked? What about LCS and Euler settings? 3+2 works fine. I'm also not getting e-mail notifications again.

Thanks,
Kevin


There is no definitive answer that I can give you regarding these setttings because it is dependant on various things: your machine configuration within SprutCAM (MatsuuraMC600VF) and maybe the postprocessor as well.

The SprutCAM machine configurations can affect the output of the CLData within a SprutCAM project. This can be for such things as the 'axes definitions' which relates to the 'axes definitions' used within the relevant postprocessor. It can also relate to items such as TCPM and the like.
Some of these items can optionally be set via the selected machine configuration.

The quickest way to resolve this would be to contact your SprutCAM reseller and ask them to have a look to see what the likely cause is. They may need to see your machine configuration / postprocessor / project.

If you cannot resolve it, send me a PM via this forum.

I am receiving reply notification e-mails here. Let me know if this persists and I'll ask them to check it again.

Dave

Edit: you can control forum reply e-mails via your forum control panel: 'My controls - Email settings'.


--------------------
"Never interrupt your opponent when he is making a mistake..." - Napoleon Bonaparte
www.sprut.co.uk
Go to the top of the page
 
+Quote Post
ka67_72
post Apr 8 2015, 02:27 AM
Post #11


Member
**

Group: Members
Posts: 22
Joined: 31-May 13
Member No.: 3,531



Dave,
After five hours of up and down the stairs between the garage and bonus room tonight I figured it out. The first thing I did was remove the blue section with "NeedKorD1" from the MulitGoto. My post is a Fanuc one that I modified for my machine. This took care of the lack of Z axis. Then I started checking boxes, generating code and testing. When I had exhausted all possible configurations with little to no effect I figured out my B was still running backwards. I had just added the 5d steps to a previous 3+2 project that was made with the B running backwards. I reloaded the machine and problem solved. My test in foam looks good. I'll post pics when I get up the nerve to put some metal in it.

Thanks,
Kevin
Go to the top of the page
 
+Quote Post
Sprut_UK
post Apr 9 2015, 09:09 AM
Post #12


Advanced Member
***

Group: Administrators
Posts: 1,091
Joined: 12-December 07
From: United Kingdom
Member No.: 4



Hi Kevin,

Thank you for the update. It's good to hear that you have made some progress.
The Needkordl looks like it is used for the output of the tool length offset command e.g. G43 H.....
It shouldn't be a problem as long as you don't start using 'Approach' moves which use the Multigoto moves. If the first move is a Multigoto and the Needkordl section is disabled, you might not get any G43 H... commands.
I'm still confused as to why the X & Y figures in the CLData did not match the X & Y output from the postprocessor.............they do appear to have some modifier applied to them.

Let me know how the actual cutting trials progress.

Dave
Attached File(s)
Attached File  Approach.png ( 30.64K ) Number of downloads: 7
 


--------------------
"Never interrupt your opponent when he is making a mistake..." - Napoleon Bonaparte
www.sprut.co.uk
Go to the top of the page
 
+Quote Post

Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



Lo-Fi Version Time is now: 16th July 2019 - 08:04 AM