Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

Interface Control Drawing Content (Aerospace)

Status
Not open for further replies.

Jste

Mechanical
Mar 31, 2021
17
0
0
US
I've skimmed through some posts and haven't found a great outline for what an interface control drawing actually contains, especially for aerospace.

I get the base idea behind it, and here is the ASME 14.24 definition for reference

An interface control drawing should: a) establish and maintain compatibility between items having a common boundary b) coordinate and control interfaces between interconnected systems c) communicate design decisions to participating design activities

When I read this, I think of a reduced dimension drawing of a single components with more text than usual to help explain important interfacing features, or maybe keep out zones. Some posts I've seen talk about actually showing the fasteners and other mating components, in my mind that now becomes an installation or assembly drawing. My working idea of an interface drawing is it shows the minimum requirements in order for another designer to make a mating part without extra constraints.

Any examples out there?
 
Replies continue below

Recommended for you

Jste,

I have prepared them and I have used them. I would be less concerned with the standard, and more concerned about the information that goes out of house.

I am designing you an ACME SUPER DUPER WIDGET, and you need to know how to design the platform to install it on. Particularly in aerospace, you need to know the mass, and all the dimensions, definitely including mount holes. You need to know keep-outs for connectors and cables and for ventilation. You may need to know grounding points. Maybe the colour matters! I regard interface control drawings as mechanical, but electrical pin-outs and loads may be simple enough to fit on the drawing. It also depends on how well I can work with my electrical department.

Security [—] the interface control drawing goes out of house and out of my control. I would not rely on a Non-Disclosure Agreement[ ](NDA) to product my IP, or national secrets. I would carefully ensure that there was no sensitive information on the drawing. If Kim Jong-un, Michael Moore, or Elon Musk show up at our front door with the drawing, no harm is done. The interface control is a separate drawing with a separate number from the assembly drawing and BOM. I regard the assembly and BOM as proprietary information, that I do not share willingly. If you ask for an interface control and I send you a drawing clearly labelled sheet[ ]3 of[ ]4, you will ask for the other three sheets, and someone may send them to you.

Format [—] When my drawing goes out of house into an uncontrolled environment, I assume the other guy has an 8[‑]1/2[×]11 black and white inkjet printer and/or a laptop with a 1366[×]768[ ]screen. Anything not readable on this equipment is potentially not readable. A C[ ]size or A2[ ]size drawing with 1/8" or 3mm lettering is readable. A D[ ]size and A1[ ]size drawing with 5/32" or 4mm lettering are marginally readable. E[ ]size and A0[ ]size are not readable.

3D[ ]CAD [—] one of the benefits of it is that I can generate an interface control drawing the moment I have an enclosed model, which I can send to you. If you are not happy about the drawing, you can come back and argue with me, and I can actually do something about it. I don't need to wait for everything to be finalised.

--
JHG
 
Consider an Edison base light-bulb. The interface is the screw thread and the bulb envelope. No internal details are required to define the limits. It can be LED or tungsten or some other technology. An interface control drawing essentially documents what is required to design the item whether that design is in-house or outside. Unlike a detailed design, the interface may include clearances that aren't places where solid material is or motion volumes as keep-outs.
 
I agree. It's called an ICD because it's just that, dimensional for it's interface. The dwg indicates outside dim's, and product info, that are useful to a customer. Sometimes aerospace customer will dictate what is needed on the ICD.

ctopher, CSWP
SolidWorks '19
ctophers home
SolidWorks Legion
 
Good to hear from all 3 of you again!

drawoh,
Thanks for the outline. One thing I would appreciate some more detail on in how you manage data when you create a separate dwg # for the ICD. In my mind, doing this doubles up your workload in the event of a config change, plus the added potential of missing the ICD and not updating it at all! Do you just rely on PDM and PLM interactions to remind you there is a 2nd drawing of your part?

You make some great points about not just including it as a sheet on the main dwg, I'm just getting nervous about doubling the amount of dwgs I need to keep track of.


All,
Any insight on if the ICD should ONLY be of components and not assys? We have some confusion on what counts as an installation / assy dwg or what is an ICD. The ICD seems to be component level to me. Would any of you be able to share an example of a mechanical ICD? I'm failing to find some
 
Jste,

In your PDM, you have an assembly model with two drawings attached to it. Ignoring CAD for the moment, your assembly has multiple drawings attached to it. You have an assembly. You may have an external BOM. I prefer external BOMs. Maybe you have a wiring diagram. You may have some form of functional schematic, like an optics drawing. Process instructions, test procedures, and the user manual anyone? The interface control is just one more document in the package that your PDM must track. For each assembly, you need a list of top level drawings. From the point of view of CAD management, I don't see a problem with two drawings attached to the assembly model. Perhaps your PDM group needs to be more flexible.

Once you finalise your assembly and its drawings, you do not change form, fit and function. There is a long list of reasons for this. These are the very things your interface control specifies. Any change affecting your specification control requires a new part number for the assembly, and probably a new document tree. If the people managing your design and updating your drawings are disorganised, anything can happen, almost all of it bad.

Think about it. If someone proposes to modify your assembly, your interface control shows you all the stuff you must not modify.

Conceivably, I could tabulate your assembly, creating a second version, with a different interface. Hopefully, I can explain this clearly on the revised specification control. Otherwise, we will need an additional specification control, and there will be three drawings attached to the assembly model. How much complexity can you manage?

--
JHG
 
You need to only expose as much intellectual property as necessary to define the interface(s). Ideally this would be a single dumb solid defining the part envelope.

"Know the rules well, so you can break them effectively."
-Dalai Lama XIV
 
ewh,

I have actually not been there. I have been handed 3D[ ]models of product, and seen way more information than needed. I see more than the vendor needs me to know. I see way more detail than I want in my large, complex assembly model. I hate it when people download screws from McMaster Carr's website.

We can always create a configuration of the assembly that suppresses most of the internal detail.

--
JHG
 
drawoh - yeah. Got an assembly that brought the workstation to its knees. Started looking at what was taking memory. 2/3rds of the memory for a biowarfare habitat model was one electric motor which was built with both the armature and field windings. And brushes. And springs. And every other detail. Because in an assembly it was really important to have all those in a model of a sealed motor.

Also a shared dislike for the McMaster-Carr models. Just random crap with random layers (IIRC) and none of the correct parameters or materials. But boy it saved a whole 4 minutes over modeling the screws or looking for existing ones. /s
 
The purpose is to define the interface(s) and only the interface(s).
There are CAD programs that will aid in this, giving a basic "wrap" of an assembly. This model would still be associative to the actual assembly model, but would lack all of the unnecessary detail. You can still determine the interface from the model, but any internal part details are absent.

"Know the rules well, so you can break them effectively."
-Dalai Lama XIV
 
MintJulep,

I have created interface controls that showed the field of view of the optics. The drawing should show all the critical external features that concern the end user.

--
JHG
 
3DDave,

I don't think McMaster Carr models save time. It takes a few hours to model something like a metric cap screw, with configurations for each size, with correct material parameters, with correct BOM entries, and without unnecessary detail. Particularly with metric screws, McMaster Carr's data tables can be copied and pasted into your design table. At this point, all your screws can be installed quickly and efficiently.

I had a pleasant surprise the last time I did this. The driver size is part of McMaster Carr's specification. I added this to the assembly BOM.

--
JHG
 
Status
Not open for further replies.
Back
Top