超级版主
注册日期: 04-03
帖子: 18592
精华: 36
现金: 249466 标准币
资产: 1080358888 标准币
|
when to release prints
when to release prints
greetings-
yesterday during a design review there was a heated debate with the manufacturing engineer about when to release prints. to make a very long story short, this manufacturing engineer wants all documents release as soon as they go out the door for fabrication, so that ways when they come in he's building to released drawings. my argument with him is i don't release prints because they are prototype and if i need to make a quick change, i don't want to be bogged down with the eco process. once the parts are in and asembled and tested then i release. in my past experience, most of the changes i do are not design but mostly document changes, such as an added detail or sectional view, or missing dims. so i don't think its wise to have changes such as these always needing an eco.
any suggestion or comments?
eng-tips forums is member supported.
the "rule" here is that, for quotes or prototype, a drawing does not have to be formally released, but does have to be controlled by ddc (document data control), so as to maintain part history.
for production, an unreleased file may be used for quote purposes, but any purchase orders must refer to formally released files.
i am looking into implementing a "pre-release" scheme, where we have more control of the data, but don't have to expend the resources to prepare and release detail drawings for development parts.
the edge... there is no honest way to explain it because the only people who really know where it is are the ones who have gone over.fff"> - hunter s. thompson
usually there is a difference in production documents and prototype documents. regardless of what you call them, they need to be revision controlled as ewhfff"> states above. i have been at some companies where prototypes are defined with numeric revs and production is controlled with alpha revs. some companies have added an "x" prefix to the prototype part numbers, while others only reference the project/program number in the rev field instead of an eco/dcn number.
how ever you do it, you should be controlling the revision of your prototypes, especially if they ever fall into anyone's hands other than your own. they do not need to go through a formal eco process, but the standard "release" discipline should be applied.
when not ready for release, always red-stamp with "for reference only" or "preliminary". then initial and date.
this will hold the prints at their rev's
chris
solidworks/pdmworks 08 3.1
autocad 08
you have manufacturing represented in your design reviews - which is good.
manufacturing's sign-off of a design review should mean "we can make this, and there is enough information here to allow us to make it."
so you need your management to broker a deal. the costs of any ecos needed to add additional details, sections or missing dimensions at the request of manufacturing after manufacturing has signed-off come out of the manufacturing budget, not the engineering budget.
otherwise, go buy yourself a big red "prototype" stamp.
londonderry,
how difficult is your ecr/eco process?
it it were me, i would be make this as convenient as possible, and insist that you and everyone else follow it. heck, i would even follow it!
once a new version of your drawing leaves your office, it must be uniquely identified from any previous versions. you are getting away without recording revisions.
the term "release" means different things in different document control organizations. where i am, an ecr can only be placed against production release drawings. if someone has a problem with preliminary drawings, or figments of my imagination, there is no formal reporting mechanism.
jhg
i've seen it done both ways.
in my opinion your manufacturing guy is more correct butfff"> to work this requires a nimble/quick/simple/low labour eco process for drawings that aren't yet in full production.
at a previous employer in the uk initial revs were alphabetical, production revs were numeric. they were all under eco control, however, the eco for prototyping only required about 3 signatories, all of them in engineering located within 20 yards of each other. this system ensured that design changes etc were properly tracked but was nimble enough to not cause schedule delays or waste labour. at full production release we didn't usually have to go through and change all alpha revs to numerical unless the contract required it, we'd usually put them to rev 1 just whenever the next change was required, sometimes years.
at my current employer, prototype revs are numeric full release alphabetical. there is no real configuration control for prototypes, it relies on the relevant engineer keeping track of changes. before release drawings must be changed to rev a. this system sucks. not only do we lose configuration control during the prototypes so changes that should have been made sometimes aren't but also this lumps all the releasing of files right at the end of the project when all the other last minute panic stuff is going on. our eco process is incredibly cumbersome, something like 10-15 approvals and can take weeks, even so called 'fast tracks' can take several days and require to you walk around the eco.
kenat,
kenat,
quote:
...however, the eco for prototyping only required about 3 signatories...
this one bothers me. why is it necessary for a company that is all at one location, to have multiple sign-offs on ecrs? the process is excessive where i work, too.
take the case that the consolidated dominion widget company has an ecr form that must be signed off by the project/product manager. the new ecr shows up on the manager's desk. the manager has a bunch of options.
he can call a meeting of all the people affected and discuss whether to sign it or not.
he can walk the ecr around the office and ask the affected people for their opinions.
he can wait for the weekly (monthly?) ecr meeting to discuss it.
he can sign it and demand to have the updated system on the loading dock by 3:00pm $$#&^^%$ it!
if you don't give people the option of getting things done, they will work around you.
if you have a large company, with scattered locations and marginal communications, you need extra sign-offs. everyone who wants to be on an ecr sign-off list should justify themselves.
jhg
kenat & drawoh
so now the long story i guess. we use autodesk inventor with the vault, soon getting product stream as well. basically our eco process goes as follows. we open the native 2d drawing convert it to pdf and then send it off to review using adobe's features. it gets reviewed and sent back with redlines or not. next we assign a number to an eco form all so in pdf format, attatch the redlines. then i go back to the native document make the suggested changes based on the redlines, and convert it pdf. then i go to the eco for send this out for approval and then the drawing out for approval as well. all of these files are then placed in a folder somewhere on the network where someone loads them into a web based server to for company wide access.
the system is very unpractical in so many ways. for example, engineers control the native documents via vault and pdf's are placed on a web server for everyone else to see.
our whole eco process is bottlenecked by 2 reviews one of which allso approves documents and the major bottle neck is the person that has to load them to the web base server. if i send a document out for review the eco form the propsed changes are not included. so the reviewer does not know if the drawing is being initial release or going through an eco, and if its going through an eco where is the eco form and markup to compare the new and old drawings?
rev control is as follows, prototype documents are assigned a x rev and controlled through the use of our cad tools and vault, mostly by engineers. once a document is ready for release its given a alpha letter and set out for review using our pdf system.
its all of these reasons that i get into screaming matches with the manufacuturing engineer. i disagree with prototype drawing getting a full product release until parts have been made, assembled and tested, such as how s/w id more or less done.
did i mention that this system was created by a s/w engineer?
seems like your company needs training in configuration management.
chris
solidworks/pdmworks 08 3.1
autocad 08
londonderry,
is it safe to approach management and tell them that the process sucks, and that you have work to do?
whis is where you need a senior manager to sit in and referee discussions, and either fix the things that do not work, or work out a compromise between good process, and getting design done.
jhg
__________________
借用达朗贝尔的名言:前进吧,你会得到信心!
[url="http://www.dimcax.com"]几何尺寸与公差标准[/url]
|