1 minute ago, Colin Gilchrist said:

Then your best bet, would be for your company to invest in a Catia Design license. That's going to be the easiest route to take. (Although also potentially very expensive.)

Otherwise, rebuilding the model might be your best option, or using a 3rd Party tool like SpaceClaim.

I appreciate your input Colin.  it will help me with other customers to know that about Catia's tolerance defaults.

41 minutes ago, Colin Gilchrist said:

If you can get them to change the value, I'd suggest 0.002mm Export Tolerance. That would get you under 0.0001" for face/edge gaps. 

 

it would be easier for me to talk to Abe Lincoln. if something is going to change it will have to be on my end.

15 minutes ago, gcode said:

beej sent me a file and I attempted tp repair it in SpaceClaim

between SpaceClaim and SolidWorks Import Diagnosis, I was able to turn it into a watertight solid

but the model is no longer accurate. Those are some nasty models beej has top work with

You'd think it wouldn't be a problem in this day and age

thanks for trying, Tom.

1 minute ago, crazy^millman said:

If is the for any of the Tier ones that fly aircraft and proper DPD/MBD process have not been followed and there is ever an audit then it could lead to some major issues. Part of the correct MBD process is a model integrity check. If that is not being done and followed from Japan then they are in violation of Boeing D6-51991 if it is for them. Other Tier ones have similar integrity checks. A lot of the newer MBD files I have seen are in 3D PDF with a Step 242 File saved in the PDF you extract.

its automotive. I didn't really want to give the maker, because I'd hate to complain about an otherwise GREAT customer.

10 minutes ago, JParis said:

What format did you receive it in?

step and parasolid.  if you open up the step file in Notepad, you can see it was created in NX. the parasolid file was created in Solidworks,  But I do not think either are the native files.  When I researched the Manufacturer, I saw that they use Catia and that there was also an article about them creating their own cad software.

2 minutes ago, JParis said:

What format did you receive it in?

I have seen Catia models bounced between several formats end up ugly as sh!t at the receiving end

I think that is what is going on.  It gets converted several times before it gets to us.  But I'm betting the original file was Catia, which has been a conversion problem for as long as I can remember.  We always gitter done.  but it adds a lot of time.

6 minutes ago, gcode said:

Spaceclaim…  it's model repair utility can sometimes do amazing things

thanks for the ideas, guys.  I never heard of Spaceclaim, I'll look into that.  and also try to find out the native format of these files.

6 minutes ago, JParis said:

It's always best if you can to have the native CAD format...

Years back there was a chart that showed the parasolid format was +98% correct on translation and that was the highest....step 214 was +96% all other formats fell lower down on the %

If data integrity is crucial, native is the way to go, even if that means investing in the plug-in to allow to open.

 

that's a good point. I'm not sure what the native format even is.  it's not easy to get that kind of information the American purchasing agents who don't seem to know a lot about their Japanese counterparts.

I"m wondering if anyone has had any luck using tolerances to help with importing solid models.  we always have problems with solid models that come in from Japan. They always convert in as surfaces or open sheet bodies and the edges are all over the place  keeping them from knitting together as a closed solid.    There are too many layers of insulation between me and the people generating these solids to get any help from Japan.

just wondering if there are tricks to help myself, that I'm missing.

thanks for all the input guys.  A lot of really good ideas here.  More than one way to skin a cat, I see. not that I'm into cat skins and such.

wire edm a blade and core pin and silver solder them together.

 

then go slap the designer silly and tell him to use a narrower, off the shelf ejector blade the next time.

some Good ideas, guys, I appreciate the input.

I've been kicking that  wire edm idea around some too.

I can't blame the designer too much on this. These aren't ejector blades, they are punches on a Trim die.  Punching flash through a pocket on a casting.  there are lots of different shapes and sizes beside the one that I showed in the pic.

You could also try ramping to keep your tool engaged.

 

Edit: You don't say which size endmill you are using. The smaller the better when trying to reduce chatter. Less surface contact will reduce chatter.

we have tried ramping and constant z cuts. about the same result both ways, but I thought, as you did, that ramping would be the way to go.

I have been using a 3/8 dia cutters with 5/16 shank.  That is bigger than I would like but, the overall length of the pin keeps me from going much smaller.

 

 

If you really want to stand it up to cut it, leave more stock for the finish pass and finish it in shorter sections. For example, rough and finish .25" lengths at a time. 

that's a good idea, I'll give that a try. standing it up is the only option I have at the moment although it may come to getting a lathe with live tooling or a 4th axis

I'm having problems cutting the thin rib on the end of this pin. the pin comes pre hardened 50Rc. the rib is .088x.275 it stands up 1.75 tall. Straight walls (no draft).  I'm stuck with the design. I only have 3 axis machines. We've been able to make them functionally right, but the customer is complaining about the chatter marks on the finish.

so far I've tried using Millstar Backdraft cutters and reduced shank endmills to try and keep the chatter down but so far not much has helped. 

I attached a jpeg file so you can see what I'm talking about.

Any ideas or processes that have helped you?

why? do I look familiar?

I am seeing the same thing shazaam, If i use all the same settings (step over, feedrates, etc) x7 takes twice as long as X6.

Maybe it handles the last little bit of material, before the break out better. but the overall time in the machine is 2x as long.

we have all okumas here. I had it happen to me once on my okuma cadet mate. the problem ended up being caused by the excessive chips built up under the table. the chips compacted, put excessive pressure on the Y axis and the machine did a little bounce back action that made it look like it was taking off into the middle of the part. the problem was up under the way covers. over time chips had gotten blowed up under there. kind of a poor design.

Thanks, Tony.

Hardmill, just workin, and then in my spare time...chasing messicans

hey fellas, long time no see.

in previous versions, if you copy arcs from one drill path to another it would ask you if you wanted to "Retain Drill Depths". It was gone in X5 and now it's back in X5MU1. is there any way to make that message go away?

It should be THAT easy on your Okuma too. All I do is hit cycle start, and then send the program.

In fact I don't have to reboot the okuma to switch back and forth between running a buffer program and DNC-B program. I just flip the switch and go.

There is something that just isn't setup right yet.

johnner did you, just previously, do an MDI function? If you did you have to hit reset. because the machine thinks it's already in the middle of a program.

I think that the dude might be feeding you a line.

I can drip feed with the version of Cimco that comes with Mastercam.

although we bought a stand alone version just to give myself some options. The version I have on the DNC computer is Cimco version 5. And it is not DNC MAX.

tell me step by step, how you are trying to send it. Maybe there's a problem other than Cimco.

I have DNC-B on one machine and use Cimco Editor to drip files

mine is setup like this:

Standard Serial Protocol

Port information

Com2 7600 7E2

standard ISO mill

make sure that both switches are in the up position on the side of your mill. There should be a switch for communication and another one for Buffer. For drip feeding they should both be up.

I think they're getting close to having it figured out.

Dapra and millstar are pretty good for that.

Wow! nice post, scott.

That is exactly how I have my computer setup. My verify is still real dark but managable. And ever since downloading the latest driver I have a drift issue with dynamic rotate in verify, no where else, just in verify.

I have been working around the dark issue by changing my lighting every time I verify and changing it back when I'm done so that I can see my shaded solids outside of verify. It's a pain in the butt but it works.