LDD drawings

[[Je...]]

How are you all handling the push towards LDD drawings? On a programming and reporting level. How are you satisfying the cumbersome general profile requirement for unspecified dimensions? This is a new concept to us. We have a few approaches but no one seems to have the bugs fully worked out for any method.

I've used space points under a min/max filter in reporting. I've used extensive collections of curves to report profile. All of it seems clunky.

Curious how the rest of the programming community is handling the task

[[Ca...]]

I work with them quite a bit. I mostly use space points, as the customer also requires point plot presentation printouts. I use MIN/MAX Result Elements to filter the points.

[[De...]]

I have a similiar question in the off topic area - regarding DPD, PDD, and PMI based parts, but so far have gotten limited response.

LDD seems like its pretty much the same thing as DPD and PDD - ie a print with critical dimensions along with a model to complete the definition of the part.

We have been checking profile all over on a number of parts for years now, however the following issue is always there and so far we have not seen any great solutions from the customers asking for this.

The biggest problem for us has always been, "Where do you draw the line on the statement 'profile all over'?"

It is certainly possible to check profile in most places to a defined datum reference frame and we generally do that by checking each plane and cylinder, in some case we use multiple 3d curves to check profile using multiple line profile characteristics in places such as toroidal and hemispherical shapes.

For us at least, the problem comes in when you consider small internal corner radii and chamfers and lead-ins in many places, especially those with many interuptions in them. These surfaces are inherently small and therefore difficult to check on a conventional CMM, especially when you consider the very limited value added there is to checking what are essentially non functional surfaces for profile only due to the desire to offload a one time engineering expense and instead create a long term manufacturing expense.

Once a determination is made regarding what surfaces are to be checked on the cmm, all profiles are then checked back to the same datum reference frame using the same evaluation settings and constraints for all features.

Once this is done we generally create a maximum result element that determines the maximum profile for all of the results and can then report it. This has been our method for some time now and works well however now some have begun asking for results on every surface individually with every surface of every model bubbled with its own unique identifier and its own line on the inspection report.

With complex models this can get ugly fast, with some models having many hundreds of faces on one model and requiring pages and pages of views to create bubbles for each face, not to mention the excessive number of lines added to the inspection reports.

[[Aa...]]

Please sign in to view this quote.

Amen. Problems happen when the lines are blurred between product specifications and inspection instructions. I think it's a perfectly legit way to specify a product, but it can leave inspection a little open-ended.

There's no such thing as 100% inspection of an all-over profile, because you're never going to contact every point on the surface. Just because I take a few points on each nominal geometric element of the surface, still doesn't prove anything.

For me, it comes down to a few questions. Who is the customer, and how picky are they? (Will anyone raise a big stink if something is later found to be a little out, and you didn't catch it?) How was this form generated, and how much faith do you have in the methods/equipment/operators? How does the surface finish compare to the tolerance?

At the end of the day, that all-over profile is usually the same as the generic +/- tolerance blocks for untoleranced dimensions on old-school prints. The engineer didn't spend much time thinking about the tolerances of each dimension. It's just a ballpark on how close it needs to be.

If they wish for more thorough (though less accurate) inspection of the entire surface, there's an emerging market of optical 3D scanning equipment.

[[Je...]]

I definitely agree with the decision on difficulty drawing a line between a confident inspection vs overly intense inspection. I was mostly curious how everyone was handling the inspection then handling the reporting to the customer. It seems we are in the same boat as everyone else. Evaluate the customers needs and history and proceed accordingly.

As far as the statement made about a customer asking for specific results per feature relative to the general profile call-out. Time for a requote.