.nc1 output of channel shape does not match part

[markhubrich]

Pretty messed up and I have my arms up. 

Part of a stringer channel, about to be processed on a plasma robot.

The issue comes about that Tekla rejects the nc1 file.  I'm surprised actually that Tekla picked this up! Look what the part looks like in the nc file viewer.

Anyone happen to know how I can repair this? There's a small handful of these with the issue but no audit or model check is showing anything wrong.

[markhubrich]

If that's not enough, the radius corners are getting flipped on cuts on angles too 😞 

[ChristianBlei]

Hi Mark,

in such a case I would try to vary the modifications, if possible and look if that changes the outcome. If it is a bug you should talk to Emy Nestor, he can dig into that and is extremely helpful. And you should report it. It will get fixed with the next update or upgrade or .......

@Emanuel.Nestor 

[Emanuel.Nestor]

Hi @markhubrich 

Sorry to hear about the issues you have with NC files.

Would it be possible to attach the model and the generated NC files from your side ?

I would like to have a look, to test on my side.

Best regards,
Emy

[markhubrich]

Thanks for looking into this. Attached is my model and .nc1 output for the stair currently in view. The red parts are the ones throwing the errors. 

If you unhide everything you'll see the other stair to the north that has the angles at the bottom landing that have inverted radius corners. This one loaded into Tekla without error I think but the cnc output was incorrect. (a1010.nc1, a1009.nc1)

Thanks again! I'm the one preaching around here for years to automate more and now I'm a bit embarrassed. Not because of the additional shop/field bolt lists I need to provide because the kiss file doesn't contain bolts. Because they're out in the shop with brand new machine and my files have errors:|  

[markhubrich]

I pulled the angle up on Raptor to make sure my freebee nc file viewer was not misleading me. It was not. It actually looks worse.

[Emanuel.Nestor]

Hi @markhubrich 

Thank you for the model and generated NC files.

Hm, at a first look ... the NC files seems to be correct, at least from DSTV-NC Standard.

Let's start with the NC file for angle, because it is easier.

First, it is helpful to generate also the NC-DXF files from Advance Steel.

These DXF files are actually a graphical representation of the NC file. They are very helpful in investigating the NC file. We can construct each AK contour from the NC file directly, but it is slower 🙂

When open the DXF file for angle a1010, it looks exactly like in the model.

Even more, when I use the viewer "Steel CAD Project Viewer" (https://www.steelprojects.com/en/), it looks also correct, as in the model (I do like this Viewer to double check, very easy to use, and really good).

I opened your files a1009 + a1010 with ProCAM Lite also, and they look correct.

Version ProCAM Lite version 2019.1.1 385.

Do you have the same version ?

Best regards,

Emy

PS: separate post for the small channel 

[Emanuel.Nestor]

Ah, something happened and the image was not loaded.

I have attached it here.

Emy

[Emanuel.Nestor]

Hi @markhubrich 

To continue now with the small channel - s1012.

The situation is tricky, it seems to be a bit outside of "normal" expectations for NC viewers and NC machine.

Where is the "unusual": the 2 shorten features from each beam end are intersecting each other, and cut the bottom flange completely.

I am not even sure if a NC machine knows how to handle this, if the two cuts can be done by the machine. It might worth having a discussion with the NC machine fabricator.

When I check the NC file, I see that the web contour is correctly described. Also, if you open the DXF file, it looks correct. (see attached the DXF file).

When open the NC file with the ProCAM lite, it shows a small "something" from the bottom flange (see image). I do not know why, it does not make much sense.

When I open the NC file with Steel Project CAD viewer, it shows the entire bottom flange, in a different color:  

Perhaps it knows a bit more on how to handle this situation.

Instead of a conclusion, I can only say that:

- the NC file seems to be correct, from the DSTV-NC Standard side.

- I do not know if this piece can be executed in reality. It might be a very good idea to discuss with the NC machine producer.

Please let me know if it helps, and if there is anything else I could assist further.

Best regards,
Emy

[Emanuel.Nestor]

Hi @markhubrich 

Please let me know if you had the time to review the information I sent, and if there is anything else I could help you further in this topic.

Best regards,
Emy

[markhubrich]

I have hopes to get back to looking at this tomorrow. Don't worry you didn't spend your time on this for nothing 😛 I'm just spread very thin lol.

[gatordesign]

@Emanuel.Nestor Was there ever a solution to this problem with channel flanges not being correct in the NC1 files?

[Emanuel.Nestor]

Hi @gatordesign,

Thank you for your message.

I am unsure what issue you are referring to inside the NC1 file.

The previous investigation in this thread did not reveal an issue there ("- the NC file seems to be correct, from the DSTV-NC Standard side.").  A possible issue might have been on the NC viewer side.

If you encounter an issue with the NC1 output, maybe you can create a separate post to investigate further. You could also attach a test model and the NC file for investigations.

Best regards,
Emy