"Checker" Checklist... 4

[zissou]

Does anyone have a good starting point for developing a "checkers' checklist" for reviewing and/or approving new/revised drawings?

Thanks.

 

[ajack1]

I think we are actually agreeing here and it is summed up by the comment from ewh

“ If the drafters/designers/engineers would only LEARN and USE what the proper rules are (see the standards listed by MadMango), checkers would not be necessary.”

This is my whole point, whilst what Mad Mango says is also true that the wrong type of text can be misread, I would expect any half reasonable designer not to use that, does anyone use wingdings at work? ;-)

Still whilst companies employ people that cannot do their job and people who reject drawings for no good reason, again referring to the post by ctopher then life looks rosy for us small contract houses that can produce better quality for less money. I am still not sure that is good for the industry as a whole.

 

[ctopher]

I think we will always need some type of checking. Us designers, drafters & engineers can't check our own work. We need a second second set of eyes to look over our work. We all make mistakes. This doesn't mean we need to employ a "checker", just have someone each of our work.
“ If the drafters/designers/engineers would only LEARN and USE what the proper rules are..." I 100% agree with this.

Chris
Systems Analyst, I.S.
SolidWorks/PDMWorks 05
AutoCAD 06
ctopher's home site (updated 06-21-05)
FAQ559-1100
FAQ559-716

 

[zissou]

I agree with all points made. All good information and great discussion.

I'm thinking about recommending a "peer review" process. As we are a small and chronically late/behind company, I think that in lieu of employing someone to specifically check drawings, we could pass the drawings to another draftsman/designer/engineer to check the drawing for errors, manufacturability, conformance, etc...

 

[wes616]

At my company, a big part of the mistakes that are made are due to hurried productions schedules, and limited manpower resources...I probablly have 40% less time per design than I would like to have... so checking has become a part of our getting it right process. I was told, that when I started here, that on average, I should expect at minimum 2 sets of redlines per drawing I release, 1 from the drafting checker, and one from the stress department. I think I usually get 3 or so sets of redlines.

A peer review is the best way to check functionality. Other places I have worked that didn't have dedicated checkers, when a drawing was assigned to one engineer, another was assigned to be checker.

Wes C.
------------------------------
When they broke open molecules, they found they were only stuffed with atoms. But when they broke open atoms, they found them stuffed with explosions...

 

[ewh]

Peer checking is a good method given the restraints we are seeing today. The drawback is depending on your coworkers to know enough to recognize the mistakes. If you have coworkers who regularly make stupid drawing mistakes, how can you trust them to catch any you may make?

 

[pennpiper]

To some extent checking criteria depends on the type of drawing being checked and the industry focus.

Where I came from, work wise that is was the Process Plant engineering and Design field. We did things like refinery, chemical plants and power plants. The drawings that my group did were the finished piping plans and the piping isometrics.
Our checking criteria did consider proper line weights, line styles, drafting symbols and demensioning.
However, our drawings were a composite of the proper input fron all the many source documents. These included (but not limited to) process P&ID drawings, structural drawings, equipment drawings, vessel drawings, electrical drawings and instrument drawings.
The main thing about this kind of composite drawing effort was "does the finished product reflect the accurate interpratation of all the input documentation?"
The checker was required to compare a "certified" copy of the source document against the finished piping drawing to insure that there was a correct transfer of requirements.
So I say it is fine to check drafting conventions but also check for content, source and purpose.

 

[wes616]

ewh,

Agreed. Peer checking is usually produces a very poor quality drawing, because peers are generally just a busy as you are, and don't want to be bogged down with the minutae of checking drawings against standards.

It, however, is a fine way to get the details of the functionality worked out. Especially, when you have to present your design in small or impromptu "design reviews." We only have the luxury of doing this on bigger projects. At a previous company I worked at, we had these more formal reviews on Thursday afternoons (everyone in the office - it wasnt too big, about 15 engineers), and would have impromptu reviews (3 engineers necessary) whenever your designs reached a milestone....

It really helped the understanding of the product by the engineer to be constantly explaining how it worked... what it does, why it does this or that, etc... to peers... And the insight from others (if the environment is right to do this) is invaluable. Products turn out much much better.



Wes C.
------------------------------
When they broke open molecules, they found they were only stuffed with atoms. But when they broke open atoms, they found them stuffed with explosions...

 

[aardvarkdw]

Wes,

I agree wholeheartedly with you. Peer review, while not widely practiced where I work, is common amoungst those of us that work in the manufacturing Engineering dept. We produce a lot of drawings very quickly and on varying different projects, this makes it real easy to miss stupid things or to get so caught up in a convoluted method of detailing that we can't see a simpler way. In my experiance 80% of mistakes can be caught just by having someone else look at your drawing for five minutes. After staring at the same drawing for hours you just don't see things anymore nad sometimes all it takes is another set of eyes.

 

[ctopher]

I have worked on projects for years creating a couple hundred dwgs. After that much time, I lose track of some aspects of the design. Peer checking is the only way my errors were caught

Chris
Systems Analyst, I.S.
SolidWorks/PDMWorks 05
AutoCAD 06
ctopher's home site (updated 06-21-05)
FAQ559-1100
FAQ559-716

 

[wgchere]

I AM a checker where I work. I check for:

**Format (line/lettering conventions, arrowheads ((some guys like BIG ones and some like small one that at are hard to see on a B-size print of a D-size dwg)), drawing borders ((some of our engineers and designers think borders are flexible and like to change elements to suit their personal tastes)), etc.)

**Tolerances. I just checked a dwg from an engineer of a fairly-simple-but-extremely-accurate-in-many-places part that had datums A thru H! I reduced it to A, B, C and after explaining it to him, he agreed. I also check to see that tolerance stack up doesn't interfere with the assembly and that the GD&T is toleranced correctly (tolerance is distributed for fixed fasteners, etc.).

**I check the drawing and the way it is drafted on the "paper." For instance, that same dwg referred to above was 7 pages. After consolidating views (there were, for example, 3 top views, each with different dimensions on them--they all fit on one view now, and the dims for a hole or feature are not on three different sheets) it is now 5 sheets and there is only one detail from another sheet on sheet 5. It took me 2 hours just to determine where the datums were and where they were applied! The drawing is now readable.

I feel my job is value added.

 

[wgchere]

I also reduced the Sections/Details from A thru J, to A thru D.

 

[KENAT]

Interesting thread.

I’ve worked both at a place without dedicated checkers and at one with.

From that limited experience I would say that having a dedicated checker, if the size of organization permits is the better option. Obviously that checker needs to be good with the ability to teach/communicate not just an eye for detail or a long list of unjustified drawing prejudices.

If you can’t justify a dedicated full time checker then I’d suggest that designating one designer/draughtsman/engineer to do all checking (except his own work which you’ll need a second person for), preferably one with a good mix of experience and eye for detail, is the best alternative. This is what we moved toward at the place I worked without a dedicated checker just before I left.

Sourcing all drawings through one person (or for larger organizations a few people) better ensures consistency, especially if they have a good set of standards, both Industry, company and if applicable customer to check to.

Having everyone check, although better than none isn’t good from a consistency point of view and some people just aren’t good at checking even though they may be an OK engineer/draughtsman. Certainly my having to check drawings when I had less than a years experience was a joke, you could have slipped almost anything past me.

Design reviews are good from a top level/functionality point of view but from my experience even the most exhaustive one I’ve been to; which had around 10+ attendees from various departments and lasted for a very long afternoon reviewing a design with only around 20 drawings; didn’t spot all the problems in the drawing pack. They aren’t good for detail drawing checking or things like tolerance analysis or interference analysis.

As to checking your own work, even some really experienced guys I’ve worked with, who knew the standards really well (“ If the drafters/designers/engineers would only LEARN and USE what the proper rules are...") made mistakes which I could sometimes spot, let alone all the ones I missed. Having a second pair of eyes is always better in my opinion. There’s no point having standards/rules etc if someone doesn’t monitor compliance to them and enforce them, even with the best of intent we all make mistakes or under pressure will cut corners etc.

As to what checkers check, some companies have different levels of checking. For instance the most basic might just be to check that the title block is fully populated and there are no really obvious mistakes, missing dims etc. The next level might be to check this and if it fully complies to standards. The highest level would also check tolerances, clashes/interfaces and basic functionality. There could be other levels in between.

“Frankly someone who thinks it is okay to dimension using thick phantom lines would be looking for another job from day one, I expect certain standards from qualified employees.” Unfortunately it seems that really good draughtsman (and I don’t claim to be a really good one) are a dying breed. The course at my university was very brief and from some of the interns and apprentices I’ve worked with I haven’t found any that had really good drawing training, plus a lot of time seems to be spent learning what the CAD can do, not necessarily what you should do with it. Unless they learn on the job, where having checkers is invaluable, they may never learn and eventually there wont be enough. There seems to be a school of thought that just knowing the CAD software is enough, no real knowledge of either drawing/design communication or engineering is needed, seems kind of dumb to me.

Sorry my first post was so long,

I’ll try and keep it short next time but this is a hot topic at my place at the moment, so I’ve been thinking about it a lot.

 

[ewh]

KENAT,
Welcome to the group! New voices are always appreciated.

 

[KENAT]

Thanks ewh,

Zissou, realised I didn't really answer the question in my diatribe.

If you have a membership the below on Drafting Zone is a pretty good list

http://www.draftingzone.com/tips/tip31/

Having looked at it I was a bit wrong on my levels of check in my original message but hopefully you get my point. Should have been

Level One, Format Check
Level Two, Design Check
Level Three, Design Review

Below is a link I found if you don't have a membership.

http://meweb.larc.nasa.gov/amsd/refs/dwg_ckng.html

Hope it helps.

 

[thundair]

I have worked at places with no check and it caused you to step it up a little to make sure there was no mistakes.
I would actually run prints and put my checker hat on and markup the drawing. It worked for me.

That being said we no longer use drawings as we are Model Base Definition and it is funny to look back at all the arguments we had over first angle projections and line width and even the use of an open 4 lol

There is a MBD check list that we use and it can be checked by any staff engineer.

Cheers