New Drawing

[looslib]

We create initial prototype drawings using an alpha revision designation. In the title block, we have 3 to 5 signatures for the initial revision A release. These signatures do not change as the drawing is revised through the prototype stages, revision B, revision C, etc.
When we decide the part is ready for production, or a as-built configuration, we will release the drawings with numeric revisions, so the first one would be revision 0. We will remove all alpha revision signatures and descriptions from the revision block. The question of the day is: Do the signatures in the title block also get removed and replaced with new ones for the revision 0. To ask it another way, is the revision 0 drawing 'considered' a new drawing with fresh signatures across the board? How does your company handle this?

"Wildfires are dangerous, hard to control, and economically catastrophic."

Ben Loosli

 

[TheTick]

That is 100% up to your organization. Whatever your policies say.

 

[ctopher]

That is the opposite to what I have seen my entire career. I have always seen numerical as prototype, alpha released.
But, we have always removed the prototype revs when prod release.
No use taking up room. As Tick wrote, it's up to your org.

ctopher, CSWP
SolidWorks '17
ctophers home
SolidWorks Legion

 

[looslib]

I figured it would be up to management and that is why the question. I wanted to get ideas from others.
Every place I have worked has always treated rev A (prototype) and Rev 0 (production) as a new drawing and the titleblock signatures have been erased along with all revision data from the prototype revisions.
What we found here was that it has been done both ways with nothing in the written procedures to define which way to go. We will revise the procedures to address this in the future.


"Wildfires are dangerous, hard to control, and economically catastrophic."

Ben Loosli

 

[3DDave]

So many things hang on to manage problems long handled by computers. I've seen revision blocks with as much info as went into the engineering change request in the first place (though not the reason, which is the important part) enabling a person to "unincorporate" if they wanted to restore the original. Why? Now it's trivial to compare drawing revisions and emphasize the changes. I've used Photoshop and can highlight in under a minute a change the size of the period at the end of a sentence - much better at finding unnoticed or unintentional changes that can propagate through parametric parts and assemblies.

 

[drawoh]

looslib,

It depends on company policy, and your requirements for accurate information, and ass covering.

The signatures on the title blocks state that a bunch of people approved your drawing. When you do revisions, you ought to document what you changed, preferably on the revision block. Signatures on the revision block indicate that changes have been approved, and that someone can be held responsible. If someone completely rearranges your drawing and they do not document what they changed and the person signing off on it is, in your opinion, a complete idiot, you probably want your signature removed.

A lot of this based on your trust in your co-workers.

--
JHG

 

[3DDave]

All of that is predicated on having a single original piece of paper or mylar that is the only source document, one that is altered by a difficult process and has to pass through many hands.

As soon as it's a computer file, all bets are off.

However, one company I worked with tried to hide this problem. They required printing out each CAD drawing revision, which then was manually signed in all blocks, then scanned for distribution and each revision original stored away. Unfortunately the reviewers could not and did not perform a 100% inspection to ensure nothing else had changed, with hilariously stupid changes in areas away from the one tiny change they were focused on. How the FDA did not discover their high incorporation error rate was a mystery to me.

 

[drawoh]

3DDave,

That's what PDM is for. Files are systematically checked into PDM. Anybody needing the information gets the document out of the PDM. If changes are required, the documents are checked out of the PDM, modified then checked back in.

You cannot work around undisciplined people.

I see no difference between files copied into people's file folders, and prints stored in people's desk drawers. One of my early drafting board experiences was tracking changes to a drawing that was xeroxed, marked up, xeroxed again from the original, xeroxed from the marked up first copy, etc. Back in the day, I loved blue/white print machines. I loved the smell of ammonia. It was the smell of [—] controlled documentation!

--
JHG

 

[3DDave]

The process was highly controlled. The company-DM (document management) is where the scanned versions of the signed, defective drawings were stored. Basically, to make a change one had to get the CAD drawing, make all the changes in CAD, make a plot file/PDF, submit that for approval, and then, when it was approved, create the new revision in the CAD-side PDM, open the CAD drawing AGAIN and make the changes AGAIN. They did this so that the approvers of the change could see exactly what they were approving, but that could not be against the CAD-side PDM document until the change was approved; they did not want to leave the new revision in the CAD-side PDM system.

They could not use a single PDM system because they had multiple CAD software systems that don't play nice with each other, particularly WRT PDM. OTOH they had trouble getting each to work with their dedicated software - UG behaved badly connected to Team Center, for example.

All highly disciplined and under FDA supervision.

Back in the day (different company) we had a clever program manager have mylar repros made, put the repros back in the vault in place of the originals, and then reworked the originals to make a derivative drawing package by erasing the old drawing numbers and revision blocks because he did not understand how the erasing chemicals worked on repros and scraping the emulsion from the repros left no tooth to ink or use plastic lead on. I think his drawing vault access was curtailed after that.

 

[CWB1]

My various employers the last decade have all used PLM. Tracking approvals electronically eliminates the need for signatures and lets you have various release status for each rev. IOW, you approve rev A for research, you and the boss approve it the second go-round for proto, and the full team approves the final A for production. You also end up with a bunch of side benefits - cleaner/easier to read title blocks, the plant calling the correct engineer rather than the guy who hasnt worked there in five years but has his name on the print, etc.

 

[looslib]

We have been using PDMLink since 2007. But we don't transfer the electronic signatures back to the drawings. We would need some additional coding in the release signoff if we did it all electronically. Not a bad idea.

"Wildfires are dangerous, hard to control, and economically catastrophic."

Ben Loosli

 

[3DDave]

The sad thing is that Windchill doesn't just do this.

 

[looslib]

Right, which is why I said it would need additional coding. I am not a Windchill JAVA coder.
I do know a few companies that have customized Windchill to populate the drawing field parameters from Windchill and republish the 'signed' files.

"Wildfires are dangerous, hard to control, and economically catastrophic."

Ben Loosli

 

[3DDave]

I was saying no one should ever have to customize it to generate this result. But Windchill is basically a separate company from Creo and I think the two groups hate each other.

More than that - the information should be dynamic, not "republished". It should be able to mark the drawing with who is making the request to see it, when it is being opened for viewing from Windchill, and any other information - directly and without using a Creo license to do it, so that this could be done for any file of any origin.

No sure that the Windchill group cares. It should have been a day one task when ComputerVision was the owner.

 

[CWB1]

Apologies if I was unclear, but our prints don't have names on them of any kind. Paper print copies have become rather rare so noncritical info like names/telephone are all kept separately within PDM as separate attributes. It might seem a pain but really isn't. Lots of folks find extracting info from a database easier/faster than opening individual prints, many of whom need your contact info and limited attributes but not access to the entire print.