Flagnoting every print dimension - bad drafting practice? 3

[McLeod]

Some of our manufacturing and inspection staff are requesting that all prints in the company have every single attribute (dimension, note requirement, etc.) individually and sequentially labeled with flagnotes, so that they are more friendly to their inspection forms. Instinctively, the drafters, designers, and engineers in product development know that this is a Bad Thing, primarily because of the resulting clutter on the drawing and the inevitable revision nightmares. But are there any particular parts of ASME Y14 which directly or indirectly preclude such madness? I've been all over it and can't find anything which specifically addresses the rampant over-use of flagnotes or similar topic. Has anyone seen this put into practice? What were the engineers smoking when they agreed to it?

 

[Gianfranco]

Dear Friends,

I'm seeing the "walls of castle" of the Engineering Departements growing up more and more
McLoad, you have lifted the "bridge" "for a moment": and this is not a good new for me.

Please, take into account the following facts.

The use of CIM and TQM (to increase the competiveness of the company when there are extendend suppliers-customers chains and nets, as on the automotive and serial production), requests one special attention to all characteristics, for variables and for attributes.

Somebody-somewhere in the company needs to stabilize the flow. This stabilization has an added value for the final customer (increasing the flow does the product more competitive).

Somebody-somewhere decided, that for stailizing the flow, it is necessary to reduce the variability of the output of all process having an impact on the production/company flow.

Somebody-somewhere decided to measure these variability, in order to do specific actions to reduce the variability and incresing the production/company flow. How to measure this variability? Measuring, inspecting all those characteristics that, in the past, or that could be "blocking" the flow of CIM's processes (all the processes! Have you intented what I mean).

In the past, the controlling methods used were not effective. The statistical sampling was sometimes ineffective with CIM and TQM.
Sometime somebody underextimated the importance of one characteristic, sometime somebody forgot an other characteristic. After those facts, somebody-somewhere decided to putting under control all the operations/processes, that have or could be a critical impact on the flow of the CIM processes, using the simplest methods known: check all the characteristics linked to the CIM processes (production, informative, logistic, managerial processes!).

How? Somebody-somewhere decided to perform an in-line and off-line measuring controls during the time. For these tasks somebody-somewhere decided first to numbering all the characteristics sequentially and, second, to use these numbers to taking under control the variability of the specific characteristic (--> database).

So, somebody-somewhere does, by hand, the circling and numbering sequentially the characteristics. How is the costs of this ancillary activities? Every guy, working in the engineering/time study departements, can gives the answer. That guy, may be or not, knows what his boss knows: the value lost caused by variability on the production flow.

Let's imagine now that your boss in charges you of finding where allocate this activity and to propose the best way to performe it. He requests you not only an opinion but an evaluation of economics, times, resourses, for different solutions.

This is - Hush, Curmudgeon, MadMango, ivymike - the original question posed by McLeod, but under the perspective of the Production Flow. This, as you see, is not an ancillary matter.

The ball is yours ....

Gianfranco.


N.B.^3: Friends, it seems a game but, who has worked in the automotive field in Japan, Europe and in some company in US, knows, more or less, the answer, and knows that the questions is very serious for the boss. Why?

Please, first replay!.

 

[vti]

Hey MeLeod,
I think there has been a little misunderstanding of your problem and I may have a solution. The requirement for numbering each dimension on a print for inspection generally comes from the requirement for traceability, which is common in aerospace work. You must be able show that each dimension was inspected. This is one reason why aerospace "stuff" is so expensive.

But the problem has been beat to death here. How about a solution. I struggled with this a few years ago and never came up with a good solution. But I had some cranial flatulence over this past weekend and came up with the idea of using a drawing viewer with redlining capability. This would put the work in the Q.C. department's hands but leaves engineering drawings alone. Now it's been awhile since I tried using redlining in a viewer but I am going to go play. MeLeod, if you try it, let us know how it works.

 

[SnowCrash]

I would flag these notes with this symbol:
( * )

Unless you think that is too subtle? Perhaps a simple "L"
would do.

Crashj 'happy to help the boss out' Johnson

 

EVERYONE is missing a few important points here!
I became VP of quality for a $40 mil aerospace machine shop, not because I was experienced in "quality control", but because I was experienced in every other aspect from CNC programming, set-up, justifications, estimates, building planning etc.

For the FIRST ARTICLE REPORTS:
The Quality Dept should do their own numbering for two reasons, it's a very good review of the drawings, and the Quality Dept is who will inspect the parts, so let them number the dimensions in a sequence that makes sense to them!

For the in process inspections:
Make simplified drawings - even sketches with your critical or controled features labled in a sequence that makes sense on the production floor, to the poor bastard who's going to be measuring the features! That means, get engineering, inspection, quality, production, and SHOP FLOOR people to review what's going on! Engineering - "this is important to control because it ..... and if this is controlled, it follows that this other thing is controlled because it uses the same process...." Quality " looking at the processes, if we have this level of control over process abc, then we only need to measure it according to schedule x, and measuring these features is the way we can monitor that process", PRODUCTION - "Geez, boss, if you want me to measure that at this point in the operation, I'm going to have to stick my head into the machine and get coolant dripped on me each damn time." (He's saying that he'll measure it after it comes out of the machine - humans must be treated with some respect.)

THEN make a SIMPLE SHOP FLOOR DRAWING with as LITTLE INFO as possible on it! Numbered features, with a sketch of the measuring tool in place to measure the feature is a GREAT IDEA. I hired an ARTIST / ENGINEER ("designer) to do it. That 1000 words = one picture is VERY TRUE. It's especially important with "instructions" because we've all been taught that real men don't read instructions, but we like cool "Popular Mechanics" style cut away sketches. Further, if ENGLISH is a SECOND LANGUAGE for anyone on your shop floor, they can still understand a simplified sketch.

Did I make it clear yet that ENGINEERING DRAWINGS do NOT belong on the shop floor for production jobs???

Measure feature #1 as shown in sketch #1a and #1b, record dimension on line #1. Compare to limits on line #1, and adjust according to TABLE #1 if required.

Their sheet just has the lines and limits, but their station has the sketches and adjustment tables.

If you think one document will suffice you're nuts! Getting the damn thing made is done by folks who you have to enable to do a good job by telling them what it is you want them to do! If you get really creative you can have another set of SKETCHES that explain what the part does in some other assembly and why certain features are important.

You'll get many brains looking at the problem with sharp eyes. You might be surprised what happens.

VPQ

 

[Umhoefer]

Hi VP of Quality,
I agree that the Quality Dept. should definetly do there own numbering. It is typicaly the Quality department who performs dimensional and capability studies, and handles the initial flack from a out of tollerance dimensions. So it just makes sense that they number the dimension that they are going to measure. And I also use very simular work instructions with limited dimensions and easy to understand instructions for measurement by floor inspectors (created by Q.E.). Although it helps that most of our Quality Engineers are proficent in CAD and have had some design expirence. As of late we are now employing actual digital pictures of the parts with the dimensions (as needed)in an attempt to make it even clearer for the floor inspectors. So far this has been productive but we automaticly stamp every printed document with an un-controlled label, and require all Work Instructions to be re-printed at every start up. This leaves us liable only for the controlled documentation on the network, which is easily updated with rev. changes.

Umhoefer

 

[dieseldown]

Hasn't anyone ever heard of inspection processes? Engineering drawings define the requirements and not how to achieve the requirements. That is what manufacturing and QC engineers are for. The QC engineer should be the one to come up with the inspection process, which dimensions are to be inspected and how. Of course, the basis for the inspection process is the engineering drawing, but the process controls the inspection and not the drawing. Part of the inspection process sheet could be a version of the CAD drawing with the process included but this is a seperate document that stands on its own. The engineering drawing is not the do all for either the manufacturing or inspection processes. Many small manufacturers make the mistake of trying to use the engineering drawing for everything. Some sort of seperate document needs to exist for inspection of parts. Changes to engineering drawings are expensive even in small companies ($1500 per ECO is a common guideline). Trying to use the engineering drawings for process control documents results in too many ECO's.

 

[awol]

I know I am entering this late, but DPUTNAM states another correct view. MOST engineers departments want to supply design (end item) information in the clearest method possible. I have worked with these "inspection forms" . The better forms list each dimension, their acceptable tolarance range, and the location (A4 or 2B3 as examples) they appear on the drawing. The inspector, then knows which 1.500 dimension he is checking for the line items on his form. Engineering drawing ARE engineering drawings. If inspection (and Manufacturing for that matter) want their own set of drawings to inspect to (or machine to), they need to create inspection drawing (or manufacturing) drawing using their own drafters. Of course that won't happen.

The pitfall you are trying to avoid: Say you add all this inspection stuff to the drawing. Then inspection methods change due to newer equipment or new inspector. Now a change is required to the ENGNIEERING drawing NOT because the product is being improved, BUT because it is inspected differently. Engineering is now using resources to maintain Inspection standards. I can't see MGMT going for that.--awol