Approval Signatures and Dates - Drawing vs. ECO

[CAD_Man]

Our Engineering department is having a debate regarding when to add Revision Approvals (Names/Dates) to a drawing. Keep in mind that we document both Initial Releases and Revisions on an ECO (Engineering Change Order) via Oracle ERP. Also, depending upon the contract, the customer may need to review and approve the drawing prior to Implementation of the ECO.

Opinion 1: Since approving the ECO IS our companies method for approving the drawings listed on the ECO, Approvals (Names and Dates) on drawings should match the approvals on the ECO, and the Date in the Revision Block should match the Implementation Date on the ECO (Which is also the date we publish the new/revised drawing). So after the ECO has been completely approved by all (including the customer), the drafter will add the ECO Approval names and dates to the drawing, create a PDF, and publish it.
-Pros:
--The Drawings Rev Date accurately represents when the drawing was published (released to production).
--No risk of not knowing if a drawing has been fully approved. If a print found on the copier has all signatures, you know it's released to production.
-Cons:
--Does not comply with ASME Y14.35M REVISIONS OF ENGINEERING DRAWINGS...
--Drawing is sent to customer without final approvals on them
--Risk of opening the drawing to add signatures and having geometry/dimensions change unintentionally (Welcome to 3D Parametric)

Opinion 2: Approvals (specifically the Rev Date) should represent the date which the Rev was rolled and the change was made to the drawing.
-Pros:
--Complies with ASME Y14.35M REVISIONS OF ENGINEERING DRAWINGS...
--Drawing is sent to customer with final approvals on them (because internal approvals happen prior to sending to customer)
--No risk of opening the drawing to add signatures and having geometry/dimensions change unintentionally.
-Cons:
--The Drawings Rev Date will not accurately represents when the drawing was published (released to production).
--Risk of not knowing if a drawing has been fully approved. If a print found on the copier has all signatures, you may presume it's released to production, but would never know.

Chris Wilson
Engineering Services Manager

Certified SolidWorks Professional
Certified Enterprise PDM Administrator

 

[3DDave]

The second method have been vetted for a long time. It is meant to manage drawing change control and if adhered to does that task well. Adding in secondary routing control is beyond a drawing.

To manage routing after it is out of Drafting/Engineering/CM hands, you can handle the 'released to production' problem with a watermark (previously a big rubber stamp that CM workers were given.)

If anyone takes a print from the copier and thinks that's what to use they need more training or a new career. Anything arriving outside a clear custody chain should not be taken seriously.

 

[CAD_Man]

3DDave,
Does your organization also state that by approving the ECO you are also approving the changes to the drawing(s)? Do the drafters add names/dates to the drawing title block after the eco is approved?
I wish there were some guidance in AS9100 or ISO on this topic so we knew the decision wouldn't come back to bite us on an audit.

Chris Wilson
Engineering Services Manager

Certified SolidWorks Professional
Certified Enterprise PDM Administrator

 

[3DDave]

Yes, I believe so - we used ECFs to document engineering changes and then ECNs to document the release of new drawings and incorporated drawings.

Once the engineering change was released it was considered part of the drawing package and suitable for procurement. If the drawing change was incorporated CM would distribute that on request. The incorporation was signed only by the checker and the PM. There were some drawings where the ECF didn't have enough detail to determine exactly what was supposed to happen; apparently because of everyone agreeing they all knew what it meant and they just released a piece of paper they thought was meaningful to them so some more research was required to handle the incorporation.

On some programs program managers got really lazy / stingy with budget; they would accumulate 50 pages on some drawings before being incorporated, usually the multi-sheet top-level assembly drawings.

Since suppliers were more interested in the deltas it made them happier to get ECFs than getting another big pile of prints to duplicated and distribute. I'm not sure if there were problems of different interpretations by suppliers than receiving inspectors over the ECFs; as mentioned, changes were thoroughly discussed and the bigger danger was a supplier thinking the phone call was all they needed.

As to the latter part, adding signatures: With ink on mylar this was never a problem, but in the CAD field it burned people a number of times. After a few cycles through the checker on a drawing incorporation, the checker would sign off on the latest check print and tell the drafter to add the name and date to the revision block. About 1% of the time the drafter would practically tinkle themselves and just open the file, zoom in on the rev block, and add the name and date, ignoring that they had failed to change some stupid setting, and make and submit a plot file to the file vault software where the CM guy would lock it down, signalling that all was good.

One of those stupid things was because the global setting of the model view state affected any views marked as Same As Environment. While tinkling themselves they would skip/miss changing the environment from Wire-Frame to Hidden-Line-Removed producing, on at least one drawing, completely garbage views. Reason for change on that particular change? Clarify drawing (some minor detail that was now hidden behind the scribble of the wire-frame)


Another thing that happened was the drafters would get so involved in the change-plot-checker-change-plot-checker loop they would forget to SAVE the changes at the end, so the next time the drawing came up many or all of the supposed changes were missing from the CAD file, so the incorporation should be repeated, but with only the latest ECFs deltas being checked. Sometimes a change that was done once would recycle a couple of times. Add a view and it would be on the release print, gone on the next release print, because the checkers only did 100% checks on new drawings. Sigh.

 

[ewh]

Technology is advancing... at my previous position any time there was a drawing change, a .cgm compare was done between the newly revised drawing and the previous release before the file could be approved. Any changes were highlighted and it was verified that only those authorized were present. If there were additional variations. the drawing was sent back to the detailer to rectify.
Now I am involved with checking model based definition/no drawings and it is early enough in this program that it has not yet become an issue; model compare has been used successfully (so far) to identify unintended changes. Approval process is entirely electronic and once approved any approvals and dates of approval are updated in the file immediately before release.

"Know the rules well, so you can break them effectively."
-Dalai Lama XIV