Change Authorization Documents

[CAD_Man]

Our current practice is to have the Drafter/Designer electronically add ALL signatures and dates (even for a new release) after the ECO is approved. There is nothing stopping them from making last minute edits (intentional or not). Then they create a PDF and publish the drawing.

My strong opinion is that an ECO only "Authorizes" the change to be made. The drawing still needs to be signed and dated, proving that someone verified that the approved changes where implemented correctly.

Can someone show me an ASME/ISO/AS Standard which states that if a drawing change is documented and approved on a "Change Authorization Document" (ECO, ECN, DCR, Etc.), than the drawing(s) does not have to be signed (physically or electronically)?

Thanks so much

 

[KENAT]

ASME Y14.100 & ASME Y14.35 are the relevant documents - not sure they go into enough detail for you though.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[CAD_Man]

Thanks for the response Ken, but I'm looking for something a lot more specific. I've found nothing in any of the ASME Y14 Series, DRM, Etc. to support the idea that a Drawing does not need to be signed if it is being approved on an ECO.

 

[3DDave]

Companies do what they want. And get the results they get. Last place I worked, they had the drafters print a copy of the drawing, which everyone signs. After that they ask for a PDF for Doc control, which may or may not match -at all- the paper that was signed. I think they were running 10% of released docs had incorporation errors. The best part was no one checked the CAD (that was controlled with a PDM system,) so it was even more likely that the CAD didn't match the signed drawing or released PDF.

It worked because they ran their shop off of unrelated shop process drawings which weren't approved by engineering, so it didn't matter what the drawings were like. Some drawings had parts called out that weren't in their ERP system, parts in their ERP that weren't on the drawings, et al. When required, the shop people would do their best but were not given the engineering drawings.

 

[CAD_Man]

Dave,
Even though companies may do things differently, doesn't make it right. The reason we have National (and International) Standards, is to establish a baseline, defining a proven process by which everyone should comply to. The more we get off track, the harder it is to return.

Even when I see posts where people are saying things like, "we can't release drawings at REV - because our PDM/PLM/MRP/ERP doesn't recognize a "dash". Also, we can't skip I,O,Q,Etc. for the same reason." it infuriates me. It's obvious to me that they changed there process to match the software. Software should be built to support (at the least) Common Industry Practices.

We have to stand for something, or we'll fall for anything!

The ASME Y14 Series has been my bible, and I tend to be quite a bible beater. LOL

Chris Wilson
Engineering Services Manager

Certified SolidWorks Professional
Certified Enterprise PDM Administrator

 

[KENAT]

CAD Man sadly I've seen different definitiaons of ECR V ECN V ECO ... and what they explicitly cover.

I agree with your assessment that the typical 'ECO' is detailing what the change is & getting approval for the change etc. - it's not explicitly approving the drawing.

Places I've worked that took this seriously still only had one signature/initials go on the actual revised drawing though - the drawing checker. Idea being the drawing checker was 'checking' the changes were incorporated properly.

Don't see this kind of thing details in Y14.35 or Y14.100 though.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[JNieman]

You sound like you're way too high strung on how to do things.

Which industry practices should software serve? All of them? Some of them? What about when some conflict with others?

ASME is meaningless in some circles. In your world and in my world, it means a lot. More in yours than mine, I'd say. Regardless, neither your world nor my world /is the universe/. It's kind like religions (especially relevant to you, a self-professed bible beater) and who can eat ham. Luckily (imo) I subscribe to rules that don't forbid the eating of those delicious mud-loving critters. I don't care if someone else's book forbids it. More power to 'em; just doesn't apply to me.

ASME doesn't try to tell you how to connect the dots in many cases, either, when it comes to document control. Not every /business/ /should/ conform to such heavy-handed practices.

If your company hasn't instituted their own controls to ensure conformance to effective practices that eliminate unacceptable risk (if there /is/ any unacceptable risk right now) then what makes you think they'd just flip over a new leaf because it was written in someone else's (your) "bible"?

 

[CAD_Man]

JNieman,
You actually got me pegged pretty well. I humbly admit to being a high strung perfectionist.
And you are also correct that I shouldn't have to fix a problem that may not exist. Everyone else IS perfectly happy and we have never been sited on an audit, but we also have no record of having been audited for Configuration Control.

As far as I know, the only nationally recognized Drawing Standard is ASME Y14.100. All our drawing state "INTERPRET DRAWING IAW ASME Y14.100".

So the answer I am looking for is probably buried in ISO-9001 or AS-9100.

Regarding software, yes... All of them! It should be flexible enough to simply automate whatever process you used before.



Chris Wilson
Engineering Services Manager

Certified SolidWorks Professional
Certified Enterprise PDM Administrator

 

[drawoh]

CAD_Man,

Are you trying to comply with a standard, or are you trying to solve a problem?

Let's ignore standards, and look for problems you may need solved.

Your Engineering Change Requests (ECRs) allow people to request changes to documentation, either to fix spelling errors, or to fix problems with the hardware. This makes a to-do list of stuff engineering needs to review. It is the start of an orderly document change process.

Separate Engineering Change Orders (ECOs) allow you to assemble a drawing update task out of several ECRs. The ECO can be a straight work order, excluding all sorts of editorial content that was in the ECR. Your finalized documents ought to be stored read-only somehow[ ](PDM?). They are authorized for release when there is an active ECO.

When you update the document, you should be checking it before you release it to production. What are you concerned about? Is it essential that the instructions on the ECO be followed to the letter? Presumably, the CAD[ ]monkey is an idiot, and the person who wrote the ECO is an engineer. Is it essential that CAD[ ]monkey solve the problem described in the ECR, somehow? They will be held responsible, of course.

The problem with a CAD administrator typing your initials into the CAD file is that you may not have authorized this. Does your PDM have a workflow that allows the people involved to check and approve the release, leaving an electronics trail. Do you want a physical or digital signature applied somewhere? Do you have a means to do this?

[ol a]
[li]We have a process called design checking.[/li]
[li]I used to live next door to Elvis.[/li]
[/ol]

Both the above statements are true. Are you really doing design checking?

--
JHG

 

[swertel]

Sounds to me that you have a Configuration Management issue, not a drafting standards issue.
There is an Engineering Configuration Management forum on eng-tips that this would be a great question for.

I will point your to ISO 10007, EIA-649, and GEIA-HB-649 as a starting point.
My real recommendation, though, is that you take a training class in CMII from the Institute of Configuration Management (

http://www.icmhq.com).

NOTE: I am a CMII-P certified practitioner from ICM, so I'm a bit biased. But I do practice what I preach which is why I recommend it. Feel free to use me as a referral if you sign up or ask other questions on the CM forum.

--Scott
www.wertel.pro

 

[CAD_Man]

drawoh,

Our ECO process is:
[ol 1]
[li]Our Highly Talented Aerospace Designers (or as you put them "CAD monkey's") are assigned an ECR, which they use to create an ECO in our ERP System (Oracle) and makes changes to Models and Drawings in our PDM System (SolidWorks PDM Pro).[/li]
[li]A PDF is made of the drawing, and given to the Engineer to review (no checker).[/li]
[li]When Engineer agrees that the changes are correct, he approves the ECO which sends it for parallel approval to all other functions.[/li]
[li]Once all approvals are received, the designer is notified. The Designer then adds approval Name/Date in the Revision Block (Adds for all function in title block on new releases).[/li]
[li]Designer then puts new PDF (with approvals) in a network folder which only designers have access to (this is where MFG gets them).[/li]
[/ol]

Yes... we do list several drawing changes on one ECO which I believe is already in violation of ASME Y14.35.

Yes... our PDM system CAN have a workflow that allows the people involved to check and approve the release, leaving an electronic trail. This would create a digital signature in PDM and on the drawing (Title/Rev Block). The CAD files would be read/only during the approval process, and approvers would have to enter their network password (which they have SOLE CONTROL of).

But we don't actually do this because the ENGINEERs claim that it's rhetorical. The culture here is that if you have approved an ECO, you have simultaneously approved the changes made to every drawing called out on the ECO.

I've always been told that an engineering drawing is a contract. But our "contracts" (drawings) are NEVER signed. So this is like signing a paper that says you approve of changes made to 10 other contracts. Then somebody types your name on those contracts.

Chris Wilson
Engineering Services Manager

Certified SolidWorks Professional
Certified Enterprise PDM Administrator

 

[CAD_Man]

swertel,

Thank you. I agree that it is related to a Configuration Management issue. But everything I research on CM always relates to software development. I have not found anything that defines or demonstrates CM as it relates to Engineering Drawings.

I will have to look up the Engineering Configuration Management forum and possibly even repost this question there.

Chris Wilson
Engineering Services Manager

Certified SolidWorks Professional
Certified Enterprise PDM Administrator

 

[cowski]

I've found the following book to be very helpful:

https://www.amazon.com/Engineering-...ds=engineering+documentation+control+handbook

 

[drawoh]

...

But we don't actually do this because the ENGINEERs claim that it's rhetorical. The culture here is that if you have approved an ECO, you have simultaneously approved the changes made to every drawing called out on the ECO.

What the heck does "rhetorical" mean?

By "CAD[ ]monkeys" I mean CAD operators who have minimal skill and authority. The engineers may or may not have access to the CAD[ ]model. They examine the ECR, and PDFs of the drawings. They mark up the PDFs and they write the ECOs. The CAD[ ]monkeys receive the ECOs and they do what the heck they are told. It sounds like your engineers are very confident about this.

I am not familiar with ASME[ ]Y14.35. I don't see why an ECO cannot do multiple drawings and even multiple ECRs. You have the data trail. I have also seen places that do not use ECOs. The ECR does everything. This process is not as flexible, but it is simpler. Remember the KISS rule.

--
JHG

 

[KENAT]

CAD Man - ASME Y14.35M-1997 section 5.3 appears to allow multiple changes if I understand what you mean correctly.

APPROVED Column. Authorized signature(s), name, or approval indicator, as required, shall be entered to indicate approval of the changes(s) made to the drawing.

Implies to me that the drawing should have some visible indication of approval beyond reference to the ECO but I suspect "or approval indicator" could be interpreted as mention of the ECO.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[CAD_Man]

KENAT,
You're close. An "Approval Indicator" is something like a Quality Stamp, which is issued to and under full control of that person, but could also be the approvers initials, as long as your system can prove that person was the one who put them there. A good PDM System can support this.

If a designer is putting an peoples names and dates on a drawing, there is no proof that nothing was changed.

Chris Wilson
Engineering Services Manager

Certified SolidWorks Professional
Certified Enterprise PDM Administrator

 

[swertel]

Another "approval indicator" is "see electronic record."

I'll share an anecdote. We do our very best to follow the rule that after a document has been approved in the data management system, NO ONE, not even Document Control, should modify the pdf file. (Yes, we still consider the PDF file the master.) This is not always possible with the nature of some requirements. This bit us in the behind during a recent internal audit where the implied release date on the face of the document didn't match the release date in the PDM system. Of course it doesn't, we never know how long it'll take for the list of approvers to review and sign off on the revision, therefore we usually type in the date as the day we started the routing workflow. That little inconsistency was considered a finding and now we either have to A) modify a released document to update the date or B) somehow clarify that the date on the face of the document is not the release date, but is closely related.

Why is this relevant to this thread. Because even before this audit finding I've been looking for a way to meet all the ASME and other requirements, not to mention convenience of being able to see approvers names on the document, without having to modify a document after release. It appears Chris is having the same problem.

@drawoh, I think "rhetorical" should instead say "redundant."



--Scott
www.wertel.pro

 

[3DDave]

The ideal is for the PDM system to stamp the PDF at time of user retrieval with the relevant information, including the date it was retrieved. CAD guy checks in PDF, which is locked, then reviewer review the in-PDM version and approve it. When last approval happens PDM software marks drawing record released.

Best way to foul it up - have someone print a copy and hand that around for signatures and then have drafter go back and generate a new version of the drawing but now with the release date. There is a 100% chance that someone will publish a blatantly wrong drawing that won't be caught until a supplier gets a copy and makes a call or an audit occurs. Then everyone gets a lecture on how important it is to never make a mistake rather than removing this obvious source of error that is carried over from ink on linen days.

 

[CAD_Man]

@swertel,
Not sure what PDM system you're using, but we use SolidWorks PDM Pro, which can add names and dates to the drawing in real time during approval without any users having write access to the files.
This is done by the PDM System driving Custom File Property values which are linked to text on the drawing.

I have the process down cold, and would be happy to write an article on the process, but have never done anything like that before, and wouldn't know where to start.

What I'm struggling with, is that our drawings are NEVER signed or filled out by the person(s) approving them. They feel that approving the ECO, is the same as approving the drawing(s). So after the ECO is released, the drafter adds the names and dates, and creates a new PDF.

Chris Wilson
Engineering Services Manager

Certified SolidWorks Professional
Certified Enterprise PDM Administrator

 

[swertel]

I have two answers to that specific issue.

1) The Drawing the ECO are two separate documents, therefore each requires their own sign off. Period.
2) We route the change order and the drawing with the same workflow, so when an reviewer approves his task in the workflow, both documents are approved simultaneously. One cannot live without the other.

#2 only exists because we create both documents at the same time and route them within the same workflow. The approvers are supposed to be reviewing both the ECO and the drawing during their step in the workflow. If, at any time, the documents aren't created or routed at the same time, then we would require two approval cycles.

[tt]Because our CAD data does not live within our PDM system, we don't have any linking of properties. Therefore, after the routing is complete the draftsman must manually enter the names and dates on the drawing, create a new PDF, and save that PDF in the "vault" as well as perform other file management tasks. At any time during the file maintenance (final release tasks AFTER the document has been approved and therefore technically released), too many hands touch the documents and we run the risk of inadvertent modification. Regrettably, this is the best process we have with the tools available to us. I am constantly looking at ways to improve the process or the tools in order to eliminate this potential source of error. It is very bad Configuration Management practice to allow this to happen. You may find the justification you need within the CM specs I listed above. Otherwise, see if you can get some CMII training; feel free to use me as a referral.[/tt]

--Scott
www.wertel.pro