ATA chapter/subchapter partnumbering

[Ozcar]

Hi everyone,

at my company i was asked to help to set up the pdm system that we'll use for our CAD files. We are designing an aircraft and should use the ATA chapter/subchapter numbering in our partnumbers.
The PDM system will use non-intelligent partnumbering, basically just sequential numbering, for internal identification, and therefore we are looking into adding the ATA partnumber in the part properties. That way we don't mess with the system's way of working and can still use the ATA based partnumber for searching and such.

One questions that arose is how are parts managed that occur in multiple ATA (sub)chapters? Say, a clip that is used in both the centerwing (57-10) and outer wing (57-20). It seems undesirable to have 2 different partnames for one part.

At the risk of opening a can of worms: how is that done in big companies?

Thanks!

 

[kontiki99]

I don't have any recent first hand experience with this. In the 80's when I was an airline technical writer, we used to talk about ATA spec 100. I think today its going to be IAW with something like iSpec 2200.

->

https://publications.airlines.org/CommerceProductDetail.aspx?Product=313

My posts reflect my personal views and are not in any way endorsed or approved by any organization I'm professionally affiliated with.

 

[verymadmac]

I have done this a couple of times once for an OEM and once for a mod shop.
For the mod shop we had the airframe model and matched the OEM's IPC section numbering that the mod was used on, so end up with something like XX-MOD-737-5720-XXX.


For the OEM if I had to do it again. I would use numbers and letter & avoid dots and dashes, as its gets a lot more items for a set number length of digits (because the stores guys will complain about typing in all the digits if they don't add any value) for assembly's and single use items. For ATA numbering I would base the numbering on the FAA's code for assembly's which is modified from ATA in the subchapters.

https://av-info.faa.gov/SDRX/documents/JASC_Code.pdf

For anything that comes in multiply sizes (shear clips, duct fittings etc etc) I would just rip off the Boeing numbering system, so for a rivet BACR15CE (this is equivalent of NAS1097), the break down is (I guess, never having worked at Boeing)
BAC Boeing Aircraft Company,
R15 the descriptor starts with an R (Rivet)& 15 is the 15th descriptor starting with R.
CE next available numeral sequence.

While some people like the simplicity of random numbers, a system allows mistakes to be more readily identified and not all operators have a automated stores system. Also operators stores systems often have the descriptors in the system inconsistently, so trying to find stuff that you don't know the exact PN of can be a real slog through all the possible descriptor combinations.

 

[Sparweb]

Try not to overcook the bird.
Two systems I've seen on two aircraft I work on regularly:
1) The entire aircraft structure was broken into 8 major asssemblies, and several dozen subassemblies under that. This took care of the first 3 digits in the part number. The rest of the digits (2) were sequentially assigned as the drawing numbers were assigned. Then come the dash numbers for each part on each drawing. The system allowed for somewhat less than a 100,000 drawings.
2) The aircraft was subdivided by 2-digit ATA number, and subassemblies get a top-level number (two) loosely based on the ATA subsection numbering. The rest (3) are sequential, for a total of 7 useful digits. Actually, the drawing number is 8 digits long, and the OEM prepended an "8" to all of the drawings - because it was their model number 8, which should give you a hint what plane I'm talking about ("Dash 8"). This system allowed for more than a million drawings.

 

[RoarkS]

My honest opinion...

First off, I personally dislike intelligent part numbers. When I have my way it's a born on part number: 031720231937. Prevents a need for a master list for sequence.

The place I'm at currently we have a 4 digit contract code, then a 3 digit identifier. the hundreds place is group. and tens, ones are sequence. Works pretty good... but we have more -after than I can shake a stick at cause well... intelligent part number systems break down over time. you can't plan for what you don't expect.

Aside from that...

In my opinion the two parts need to be separate part numbers. You may need to revise one, modify it, maybe an AD or SB gets issued that effects one but not the other.

 

[verymadmac]

I worked for a short while for an Oil & Gas component manufacturer (subsea high pressure valving & Jumpers), they kept no drawing register all just a project register. Each valve was got a project number followed by 4 digits. Its was great from an administration point of view except they end up with a number of quite complicated units that were identical except for an orifice plate & needle, each having completely separate drawing stacks that under went divergent evolution as improvements weren't carried across all drawing stacks and then required one to make a substantial effort to read new qualification tests across all the units.

They also probably could have reduced their time spent drafting by 10% if they had used a couple of table drawings instead of having to draw every single straight & single bend length of pipe & giving each one a completely unrelated part number.

Whatever system you use make sure it allows for pragmatism.

 

[Ozcar]

Thanks for the great answers, very helpful!

Will take (need) some time to digest it all, and might post some follow up questions...

 

[WKTaylor]

The part numbering drawing scheme You are looking-for 'resembles' the system for F-16 drawings. The drawing numbers were so easy/intuitive to 'read', that after a couple of years field experience, I could look at a drawing number and practically guess what the drawing was about... or VV... could be working in area... and intuitively know major Assy numbers, etc... where I needed to start looking for 'details'.

These numbers also incorporated 'a letter' for the major element/area... so there was no need to guess. IE F = Fuselage, W = Wing, S = Stabilizer and so-on.

ONE CAUTION about drawing numbers: characters and lengths make a difference. IF YOU are not already aware, drawing/drafting/CAD conventions are defined in ASME Y14.** specifications. There is a discussion of developing drawing numbers and part numbers on drawings... and conventions for drawing titles and parts. BE CONVENTIONAL!

Also, A relatively new ASME spec tailors the Y14.** for aerospace: ASME AED-1 Aerospace & Advanced Engineering Drawings

AND OH YEAH... whatever You do... develop a concise drawing tree with each [all] major Assy/Install and sub-tier drawing(s) listed. You won't regret this road-map.

Regards, Wil Taylor
o Trust - But Verify!
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation, Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", HBA forum]
o Only fools and charlatans know everything and understand everything." -Anton Chekhov