5 axis help using a trunnion

[Jeff2005]

The way you have "A" is correct.  That would be a limit of A-110.  "C" is the other way.  You have "C" showing Positive direction.

[So not a Guru]

On 8/6/2017 at 0:43 AM, : Chip : said:

Buy a KTM/Husky-seiki

For most, there is therapy. For the rest of us, there are motorcycles.

[gcode]

Jeff,

I've reviewed your zip2go and the machine definition looks kike it should from the photo you uploaded

I'm not sure about some of the tool planes in your file though.

I'm having a hard time wrapping my head around the A pivoting about the Y axis

I've worked up a simple drilling file with 5 different angled faces

Run it through your post and see what you  get

The block's origin is the pivot point of the intersection of the A and C axis

Post the file

Make tool plane A-70 C0 active and analyze the holes in that face 

The hole positions you get with A-70 C0 should match the posted code for that toolpath

That should be true for all 5 toolpaths.

https://www.dropbox.com/s/7b4a7osyv8uc93x/DRILLING TEST.mcx-9?dl=0

 

edit...  I forgot to mention that the post in the zip2go is encrypted and will not run on my sim

I can't test the file, but I believe it's correct judging by your phot and the machine definition in the zip2go

 

 

[Jeff2005]

Below is pics of the first program I posted, using my post. as you can see it is not correct, from what I can underdstand. plus the A axis would limit out at 30 deg.   It would not be able to go to A70

[gcode]

If you MDI the A axis to A0 is the platter parallel to the table

or perpendicular to it??

[Jeff2005]

Looking at the pic shows the trunnions A & C at 0 position.  Looking from the front of the machine as I stand taking the pic A is your cradle and C is your platter. A rotates left with a limit of -110 deg.  C if rotated CCW it will be a negative number.  Hope this clears it up.  LEaving for the day, so I won't see your comments till monday morning.

[Watcher]

This sounds like a wrong vector definition for the rotaries origins and their directions.

If I'm correct, you're pretty close to a fix. 

[gcode]

Maybe someone else can download and review my file to be sure I haven't dome something stupid,

but I'm thinking the A axis direction of rotation needs to be reversed in the post.

We could force this to post good code with the misc int switches but we shouldn't have to.

 

[crazy^millman]

It is a Postability Post. Shoot and email to Dave or Tim and I think they should be able to get this sorted out really quick. I wish you could get back in touch with the dealer to have them part of the process, but it is what it is. You are very close with the post and once you sort out this issue you should be good from a posted code side of the equation.

Need to also pay attention to these setting in the post.

orgshift       : 1      #Shift origin from Mastercam's programmed origin (0=Off,1=On)
ignoreorg      : 1      #Ignore toolplane origin, use WCS origin (0=Off,1=On)

That plays a big role in how your posted code is going to come out based off of your programming process. If you are moving everything to Zero then you are good to go, if you are not might try seeing what your code looks like with them turned off. I normally post code with them off, but I am old school putting ever plane back to the same point for each WCS. The new way of letting the post sort things out with different planes at different places, but tie it all back to one WCS is this switch process.

[Greg Williams]

Jeff,

I think you will need to have this one as well

one_revr       : 1      #Limit 3+2 rotary positioning? (0 = No, 1 = 0 to 360, 2 = -180 to 180, -1 = -360 to 0)

 

[Guest]

A Axis is around Y? 

Who decided on that? 

[Jeff2005]

7 hours ago, Foghorn Leghorn said:

A Axis is around Y? 

Who decided on that? 

If we had decided to turn it the other way, then we wouldn't have room on the table for our vise.

 

On 8/12/2017 at 11:34 AM, C^Millman said:

It is a Postability Post. Shoot and email to Dave or Tim and I think they should be able to get this sorted out really quick. I wish you could get back in touch with the dealer to have them part of the process, but it is what it is. You are very close with the post and once you sort out this issue you should be good from a posted code side of the equation.

I have already emailed them.  I have no response other than Dave is on vacation until the 21st. 

[Jeff2005]

Ok, Question of the day.  In our machine the "C" axis is the platter it's label to be 4th axis.  "A" is set to be 5th axis.  Is this correct, or should this be reversed. 

thanks,

[Watcher]

8 minutes ago, Jeff2005 said:

Ok, Question of the day.  In our machine the "C" axis is the platter it's label to be 4th axis.  "A" is set to be 5th axis.  Is this correct, or should this be reversed. 

thanks,

Think about it in terms of primary and secondary axes. 

There a good posts here about this, and some other blogs out there with details. 

But in a trunnion, usually the primary is C and the secondary is the other axis. 

I think your setup I correct in this aspect, otherwise the post would not output correct angles even for 3+2.

[gcode]

4 minutes ago, Jeff2005 said:

Ok, Question of the day.  In our machine the "C" axis is the platter it's label to be 4th axis.  "A" is set to be 5th axis.  Is this correct, or should this be reversed. 

thanks,

Typically the C is considered the primary in a situation like yours, but it depends on how the post is set up

I emailed my test file and a link to this thread to Dave

He is out of the office this week but said Tim would be in touch with you get this sorted out.

[Jeff2005]

What i am thinking about is even when i post the program in 5 axis, it is not completely correct.  It tries to crash.  In our simulator it DOES NOT show the crash.  Would this be something in the machine causing this to happen?

[gcode]

7 minutes ago, Jeff2005 said:

What i am thinking about is even when i post the program in 5 axis, it is not completely correct.  It tries to crash.  In our simulator it DOES NOT show the crash.  Would this be something in the machine causing this to happen?

absolutely..

the x axis number in that sample file are coming out completely wrong too

I'm thinking the post has the distance from the c/l of A to the C axis platter built in

 

[Jeff2005]

1 hour ago, gcode said:

Tim would be in touch with you get this sorted out.

Tim has emailed me and set up a time for a phone call.  I will keep you guys updated.  thanks

[Watcher]

28 minutes ago, Jeff2005 said:

Tim has emailed me and set up a time for a phone call.  I will keep you guys updated.  thanks

I think you guys started with the right foot in regards the post processor, although it didn't seem to be like that in the 1st moment. 

Many companies don't realize the importance of obtaining the post from a reliable and experienced provider.

Now you have all the information the post writer needs, will get this to work in no time. It's just a configuration issue I'm sure, maybe due to your 'exotic' setup. Maybe the post/machine definition that was delivered to you was configured to assume you would use a trunnion in the regular way/orientation.

Often companies that save money in a post end up in a dead end. That's not your case here. The post you ordered is provided by world class professionals and you will experience this very soon.  

[gcode]

1 hour ago, Jeff2005 said:

What i am thinking about is even when i post the program in 5 axis, it is not completely correct.  It tries to crash.  In our simulator it DOES NOT show the crash.  Would this be something in the machine causing this to happen?

There is something wrong with how the post is set up

it s outputting strange Z values and the X axis values are wrong, even when accounting for the reversed axis.

For the time being, I would not spend any time on your machine simulator.

The first order of business is to get the post outputting code that drives the machine properly

Dave told me your machine simulation is not hooked to the post.

That means it's running on Mastercam's internal NCI file and does not reflect what posted NC code will do on the machine.

Once your post is dialed in and your fully understand how to program the 4th and 5th axis it can be a useful tool,

but it is not Vericut.

As you already know, it is showing my sample file is safe, yet the posted code would wreck the machine

 

 

 

[Newbeeee™]

11 hours ago, Foghorn Leghorn said:

A Axis is around Y? 

Who decided on that? 

Okay - probably a stupid question...

I understand the need to have the rotary on the machine the way it is for best machine utilisation.

But would it not have been best to reassign the machine control parameters in the first instance, to match the 'standard' AX, BY, CZ  format? Would the post/mac/control def etc have initially worked better???

[Jeff2005]

14 minutes ago, newbeeee said:

But would it not have been best to reassign the machine control parameters in the first instance, to match the 'standard' AX, BY, CZ  format? Would the post/mac/control def etc have initially worked better???

Probably, but when you're new to trunnion/5axis programming you tend to lean to the person that seemingly knows what they are talking about.  This would have been great information going into buying the machine. They left the programming group out of the initial buying process all together.  Smart, right? I don't think so, but still we wouldn't have even known to ask this question about positioning of the trunnion unless we had experienced this problem before.  

[Newbeeee™]

8 minutes ago, Jeff2005 said:

Probably, but when you're new to trunnion/5axis programming you tend to lean to the person that seemingly knows what they are talking about.  This would have been great information going into buying the machine. They left the programming group out of the initial buying process all together.  Smart, right? I don't think so, but still we wouldn't have even known to ask this question about positioning of the trunnion unless we had experienced this problem before.  

Jeff - I see where you're coming from and please don't think I was being cocky in asking the question.

The reason I asked is when we got our first vertical with a 4th, the machine control was assigned as 'B' and the 4th was sat on the table aligned with X.

No biggie thinks I, we'll just do a post mod and change all 'A' output to 'B'.

In my case, a little knowledge was a dangerous thing, and we quickly learnt to re-think this and changed one parameter in the machine control from 'B' to 'A' and everything was then hunky dory with mastercam.

Obviously this was just a 4th, and what you have going on there is a whole lot more complicated.

[Guest]

X, Y, Z

A, B, C

I, J, K

U, V, W

 

A rotates around X

B rotates around Y

C rotates around Z.

 

I relates to X

J relates to Y

K relates to Z 

 

U relates to X

V relates to Y

W relates to Z

 

If your machine has anything other than that it is configured/addressed wrong and you're going to have trouble talking to knowledgable people to get information that is vital.

 

 

[Newbeeee™]

^^^ unless it's a Haas.

Or a Bridgeport Hori of a customer of mine where the table rotates and it's a C instead of a B...

Standardisation is a wonderful thing. If only people would do it.