What should i do? 5

[evilchickenking]

ok, so i work for a pump manufacturer that designs and builds large pump. currently im their draftsman. up until now all has been good there but lately they've been hounding me to find a way to catch more mistakes.

they've really, flat out, just not implemented anything that would remotely resemble a system in which my work is checked. what should i do?

today i was told that some job got botched because of something that should have been caught by someone and it's landed squarely on my shoulders.

what system do you guys use check your work and how does it function?

 

[davidbeach]

Well, I'd go so far as to say that you can never check your work as well as someone else could check it. The checker really needs to be somebody else, but if you have to check your own work, you need to get away from it for a while (several days would be good) between the time you did it and the time you try to do the checking so that you have to spend a bit of time familiarizing yourself with what it was you were supposed to have done and how that relates to what you actually did.

 

[jmw]

No ISO 9001 ?
No quality scheme that clients can audit?
No quality department?


JMW

http://www.ViscoAnalyser.com

 

[TheTick]

Make new title blocks that require signatures in blocks marked "Checked by" and "Approved by".

I worked for a company where the enginering manager said we should simply not mistakes on drawings because "we are all adults and should know how to make drawings." Not possible. It takes at least two minds and three eyes to properly check a drawing.

I could be the world's greatest underachiever, if I could just learn to apply myself.

http://www.EsoxRepublic.com

-SolidWorks API VB programming help

 

[tygerdawg]

Yup, what you've got is a bunch of nincompoops running the company. The direction you're getting is nothing short of a management cop-out.

"Back in the old days" of pencil & paper design, there was a hierarchy: engineer, draftsman, checker. There were probably some "leads" scattered in between. But with the advent of CAD, the role of checker has been lost.

It's simply bad to have to check your own work. But it's like the other guys said: force a manager to sign off on the drawing and assume responsibility, or the self-checking process adds a few days to every project. If a calm, analytical cost justification is presented to those clowns running your company, and the volume or value of work is sufficient, then perhaps you can convince them to pony up and hire a dedicated checker.

TygerDawg

 

[GregLocock]

Has anyone tried 'buddy checking' ?

Good excuse for a pay rise if nothing else.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[TheTick]

I worked at places where we did a lot of "buddy checking", since there were no checkers. This was considered a normal part of the process, and expected from the managers.

Again, it is vital to consider how "human factors" affect our work. It is simply human nature to not see mistakes that we have stared at for too long.

We are not machines. Our work is the product of our minds; we only assisted by our tools.

 

[PSE]

At a minimum, I would (hope) that the designer/engineer who created the item/assembly would review and sign off on it. For us, it is (if the procedure is fully followed), Design Engineer, Manufacturing Engineer, Systems Engineer or Program Manager. Engineering changes to drawings go a similar route for approval but the revised drawing itself is only signed off by the ECO originator (usually Manufacturing Engineer).

Regards,

 

[kchayfie]

Everything done in my department is checked by someone else in the department (which can be anyone in the department from new-taking-time-out-from-college student to been-doing-it-so-long-I-checked-Noah's-design-for-the-ark senior engineer). Then it gets approved by a senior engineer / department manager (which can't be the same senior who checked it). The approver usually looks at who checks it and he trusts them, he'll sign it without looking too hard and if he thinks the checker is a bit wet behind the ears and needs a bit more experience he'll check it himself before approving it.

That way, when things go wrong, there's at least 3 people to blame!

 

[JStephen]

Print it out, go through and check every last detail- highlight it as it's checked.

Generally, it's easier to spot something that's wrong than it is to spot something that you forgot to put in, in the first place.

If you check your own work, try to do things differently from how you did them in the first place. Dimensions that were just measured in autoCAD, derive geometrically. Check analytical integrations numerically.

 

[lostsailor]

evilchickenking,
Any of my clients that do any serious design work have a documented system in place for independent review/checking of drawings and calculations. It would be beneficial to everyone in your organization, especially your customers, to develop one.
Regards,
RLS

 

[YoungTurk]

Our system for aerospace engineering is formalized to include multiple checks, but things inevitably get caught on the floor anyhow.
We have:
designer do a self check (check BOM and concept),
lead do a layout check (conceptual check),
the checker for a drafting check (readability, standards followed),
the appropriate specialized engineer(s) for stress, mechanical, and/or electrical check (this is for compliance to appropriate regulations and eventual certification)
and then possibly also on to the production lead or manufacturing engineer, since they are downstream customers.

The drawing is then routed a second time, after pick-ups, to collect check, cert/stress, and finally lead signatures. Often times several of the checks will occur simultaneously for the simpler drawings, which can expedite the process but can also confuse things when it comes time for signoff since so many changes may be incorporated simultaneously.

On the flip side, I used to work as a research engineer at a lab. No drafters, checkers, or specialized engineers. I took it upon myself to do a self check (never the same day I finished the drawing) and circulate check copies (clearly marked as such) to my supervisor and the appropriate downstream customers (typically the machinist, mechanic, and/or electrician). Though they didn't always have the time or desire to do a full check, that accomplished two things. First nobody could ever say they didn't get a chance to provide input or pick-ups; second, although the ultimate responsibility for mistakes was mine, it was much harder to point the finger at me when the inevitable mistake was found. It also opened the door to get feedback from all involved, which some of the production guys really appreciated. I learned a lot from them and they gained an appreciation for what I brought to a project.

 

[ewh]

Many good suggestions here; I think davidbeach gave the best, most realistic solution for your situation (which is far from ideal). I also think that it is important that there is an "approved" sign off, as well as other sign offs on the drawing, and that they are recognized for what they are. This can only help spread the responsibility. It is also a very good idea to get feedback from your customers before release.

 

[aardvarkdw]

I feel your pain. I am in a similar position, being a drafter that produces drawings for others with little to no checking of my work.

There are a couple of things that I do to help limit the number of mistakes. One, find someone else to at least look at the drawings before you submit them. Don't wait for someone else to implement a system of checking, find someone else to look at it, and try not to let it be the same person that gave you the assignment. Often times the engineer doesn't see the mistakes either because they are seeing their design, not your drawing. Second, if you are drawing in a 3d program like solidworks or inventor, take your finished print and try to redraw it in 2d based on the dims on your drawing. This can be tedious and time consuming but it helps me to catch my mistakes. It also helps you justify your dimensioning scheme. If you can't redraw the part without trouble you need to rethink your dimensions, and it also helps you find things you missed.

David

 

[KENAT]

These threads talk about drawing check and my posts on the dates below give some of my thoughts.

thread1103-151604 (1 Aug) & thread1103-151962 (June 12th)

My personal preference is dedicated drawing checker for primary check with technical approval from relevant engineer (stress etc) and preferably final sign off from some kind of manager (technical not HR or someone).

Doesn’t sound like this is necessarily an option at your place. Definitely try and get at least a second sign off on the drawing. If you must rely on only self check then like others have said, print it out and come back to it a few days later, but I wouldn’t be happy with this.

Others who’ve read my posts on checking will know I can go on for ages but I’ll leave it there for now!

 

[ctopher]

There should be a drawing checker. The drawings should have been checked by the design engineer, production engineer, and quality. It should never be assumed a drafter checked the design.
Also, it is up to the drafter to do a spell check. (your first post)

Chris
Systems Analyst, I.S.
SolidWorks 06 4.1/PDMWorks 06
AutoCAD 06
ctopher's home (updated 06-21-06)

 

[evilchickenking]

thanks guys. those are great tips.

so far I've proposed that:
1.) a comprehensive book filled with data that pertains any position you happen to have at the company,
2.) we implement iso 9001 standards.

i'm fairly sure this company can do it, but i'm not sure they will go all the way. when i told them about the idea they asked "who was going to do it?"

are there companies that specialize in setting up a quality assurance system?

 

[KENAT]

Yes, can't name any of the top of my head but there are consultants that help with this stuff. I think possibly some of the auditing companies may even help, we used Lloyds at my last place in the UK.

To solve your immediate problem you don’t need full ISO 9001. You just need to set up a checking process, document it in a procedure and have a policy that all designs/drawings go through it.

 

[ajack1]

This is always a problem and basically boils down to the fact that humans make mistakes.

The first thing I would say is do not assume that ISO 9001 will solve your problem, if your manual states all drawings will be checked by the designer then you are back to square one. All 9001 does is check that systems are in place and implemented, a well-written and implemented internal system will work better than a badly written and implemented external one.

It is very easy to say that many different people and departments should check any work and whilst this is true and certainly the best method it is not always practical depending on the value of the product. It is very easy for a company to go broke trying to produce the perfect washer.

Have you been making more mistakes than “normal” lately? There could be a perfectly good reason for this, overworked, there may be a new baby in the house and you have not had good nights sleep in weeks, that kind of thing. A good boss will discuss these types of thing.

Whilst no one likes other peoples mistakes and less so our own they do happen, it would be my advice to try and get people around the table to come up with a working solution, as well as reducing errors you may also increase production with input from different departments.

Good luck.

 

[JStephen]

On checking drafting and engineering details, the approach required will need to vary according to the circumstances. If you you're spending a $1000 worth of drafting time to design a $1,000,000 product, it makes a lot more sense to invest another $1,000 in checking it. If you're spending $20 worth of drafting time drawing up a $50 part, it doesn't make so much sense. If the drawing(s) will be used repeatedly or distributed widely, it makes a lot more sense to make sure that every last detail is correct.

Don't just assume that the same standards of care used on the space shuttle should be applied to miscellaneous steel fabrication.