ECO (Engineering Change Order) Revision - Letter change or part number change?

K

kcmomde

We have 2 conflicting opinions on what constitutes an rev letter change to a PCA versus changing the part number. We had some boards assembled, but the bill of materials that we provided was not correct, thus some resistors were not installed. One camp states that the part number should change for example 123456-00 Rev A to 123456-01 Rev A. But the other camp states that it should only be a rev letter so it would become 123456-00 Rev B. The second camp makes a good point that we should only change the part number when the bare PCB is changed, but the first group states that adding the resistors changes the boards form, fit or function. Of course I'm in the middle of the whole thing so I wanted to get some other folks opinions.

Thanks,

Kym:confused:
 
F

fireonce

I prefer the second method,which is more convenient especially in system.
Of course, if the changed part is fully different from the original part.
 

CarolX

Trusted Information Resource
Hi Kym,

I see this method of changing part numbers more and more often. It tends to occur in companies that are using Oracle for their MRP system. Since I know nothing about Oracle, I can't really comment why it is happening -

We have a customer that changes part numbers everytime they make a small revision to the part (i.e. move a .250" dia hole by .500"). Unfortunaely for us, when we receive an order for the new/revised part, it triggers our prototype process instead of our revision process.
 

Steve McQuality

Quality Engineer
Hi Kym,

...We have a customer that changes part numbers everytime they make a small revision to the part (i.e. move a .250" dia hole by .500"). Unfortunaely for us, when we receive an order for the new/revised part, it triggers our prototype process instead of our revision process.

Hi Kym:
Carol's example above is a little extreme from a customer perspective(although I know it happens). We've had this same discussion recently at my company. I would offer the following;

Considering this is a PCB change (components only to correct in internal error w/ the BOM), it would depend on how you identify the parts and what kind of traceability you need for that change. Do you have P/N and Revision screen printed on the PCB? If so, then the Rev change may be traceable and you will be able to segregate the parts and not mix them up in production.

However, if you don't have the Rev printed on the boards you'll need to decide the degree of the "what if". That is to say, "what if" the new and old PCB's get mixed up (How much time & money would this cost you?)? What about returned products - can you easily identify which board is in a returned product several months down the road. Are the parts interchangeable (I assume not, since the addition of components was made to fix an earlier omission)?

The reason I quoted Carol is that, I would assume (and maybe Carol can comment from a Supplier perspective) that if you contact your supplier of the PCB and indicate that the P/N change is only for traceability in your facility and that the board will not require new artwork, Gerber files, etc., then the Supplier will be able to "circumvent" the automatic trigger for additional cost that comes with a brand new P/N. ...just the addition pennies to add a couple more resistors.

...I hope this helps.

-Steve
 

CarolX

Trusted Information Resource
The reason I quoted Carol is that, I would assume (and maybe Carol can comment from a Supplier perspective) that if you contact your supplier of the PCB and indicate that the P/N change is only for traceability in your facility and that the board will not require new artwork, Gerber files, etc., then the Supplier will be able to "circumvent" the automatic trigger for additional cost that comes with a brand new P/N. ...just the addition pennies to add a couple more resistors.

Hi Steve and thanks for your input - I can see how this may work to an advantage in other industries. I work in sheet metal, completely different animal!!!!

And yes - it would be nice if someone would pick up the phone and call us, but, alas , we are at the bottom of the food chain.:lmao: :lmao: :lmao:
 
J

John Nabors - 2009

Kym and Carol-

I just did a quick review of ASME Y14.35M-1997 ("Revision of Engineering Drawings and Associated Documents"). The standard does not directly give guidance for when the scope of the design change requires superceding the drawing; however I would think that in the examples you both cited it would be better to rev up the drawing than to supercede it just because you would have the revision history right there on the drawing to reference the evolution of the design.

Here's a thought to throw out - suppose one adopted the philosophy that a change in a design feature for purposes such as correcting an error results in a drawing revvision but a change in the design intent is cause for a new part number? What do all of you think?

Also - Carol stated...

We have a customer that changes part numbers everytime they make a small revision to the part (i.e. move a .250" dia hole by .500").

...Does the customer reference the superceded part numbers as specified in paras 4.4.1 and 6.1.3 of ASME Y14.35M? If the superceded drawings are retained for reference purposes are they labeled in accordance with para 4.4.2 of the standard? Just wondering.

Regards -John
 

CarolX

Trusted Information Resource
...Does the customer reference the superceded part numbers as specified in paras 4.4.1 and 6.1.3 of ASME Y14.35M? If the superceded drawings are retained for reference purposes are they labeled in accordance with para 4.4.2 of the standard? Just wondering.

Regards -John

HAHAHAHAHA - I don't think any of my customers follow that standard.

In fact - FWIW and :topic:

I asked our young, recent college graduate engineers what they were taught in their college drafting courses, and they both replied, we didn't take drafting classes - not a graduation requirement for BSME.:lmao: :lmao: :lmao:
 
J

John Nabors - 2009

I asked our young, recent college graduate engineers what they were taught in their college drafting courses, and they both replied, we didn't take drafting classes - not a graduation requirement for BSME

That... ...is truly... ...apalling....
 

Steve McQuality

Quality Engineer
I asked our young, recent college graduate engineers what they were taught in their college drafting courses, and they both replied, we didn't take drafting classes - not a graduation requirement for BSME

That... ...is truly... ...apalling....

Not really... I mean, it's up to us BSME's to design the parts; it's up to someone else to try and figure out how to build it! :cool:

Those "new college grads" are probably the same ones that put +/-.0002 tolerancing on a 12" x 12" steel plate! :bonk: It's all in what you're taught... or how many machinists you've had to deal with throughout your career :whip: :lol:

-Steve
 
J

John Nabors - 2009

:topic: :topic: :topic:

Those "new college grads" are probably the same ones that put +/-.0002 tolerancing on a 12" x 12" steel plate!

That is just one little facet of a much wider trend. In so many fields you see the education and training process try to build the top of the pyramid without ever building the bottom. Another example is the now-common practice of training people on CNC equipment who have never so much as touched a manual mill or lathe and whose background on a surface plate consists of leaning against one drinking coffee while an inspector tells them if their work is any good or not. The days are almost gone when a reasonably trained machinist could walk into any reasonably equipped machine shop, take a drawing and a batch of raw material, and see the job through to the point where the product was ready to be packed up and shipped to the customer. I would say that probably 3/4 of the CNC operators know how to load and unload parts, keep the fliuds topped up,
check a few attribute gages, and hit the big red button if they hear an unaccustomed noise. And maybe, just maybe, punch in a tool offset if it is necessary.
 
Top Bottom