[Closed] Engineering drawings - dimensioning

Don't bother. Just use a note.

davidtaylforth - Member

Don't bother. Just use a note.

To be honest, this was my first thought.
I can't remember having done that sort of alignment on any drawings/assys, but I can't think of a gtol that would encompass it easily.
Presumably the individual panels have some kind of surface profile gtol so they should be the same shape where you need them to be flush?

I'd probably put a x-section in with a note, but not having to do much sheet metal stuff, or surfacing type design I might be well wide of the mark.

Have you asked the supplier how they would normally achieve this & if they have a preference for it being annotated on the drawing?

This thread is useless without pictures.

Igmc

Don't bother. Just use a note.

Just had a quick scan through my copy of "Manual of Engineering Drawing" on my desk and can't spot anything that would cover it so this would be my guess too.

We would use the 'common zone' callout as below. The first symbol (the parallelogram) is for flatness, then your tolerance (0.5) and then CZ. I don't know if this is universal, but it's what we use for precision parts.

[img] [/img]

edit: [url= http://www.roymech.co.uk/Useful_Tables/Drawing/draw_geom_notes.html ]Roymech[/url] has a useful page; scroll down to Common Zone about 2/3 the way down.

Have you asked the supplier how they would normally achieve this & if they have a preference for it being annotated on the drawing?

The whole shebang is in-house so no supplier to ask. Currently we do this:

[img][url= https://c1.staticflickr.com/8/7476/27357904616_6d3aec9fc4_m.jp g" target="_blank">https://c1.staticflickr.com/8/7476/27357904616_6d3aec9fc4_m.jp g"/> [/img][/url][url= https://flic.kr/p/HFwpUC ]Capture[/url] by [url= https://www.flickr.com/photos/30105051@N04/ ]markywatton[/url], on Flickr[/img]

but this does not work as we use the gtol datum symbol although neither surface is a datum.

I'd probably put a x-section in with a note, but not having to do much sheet metal stuff, or surfacing type design I might be well wide of the mark.

Yeh, I reckon that's a good shout. Just do a scrap section and a note.

At the end of the day, aslong as someone can understand how to make it, that's what matters.

In the example above you could do it with a simple linear dimension from the part on the left with the Dutch fold to a tangential face on the other component.

Like Mowgli I would use a geometric tolerance. I usually do it by making one face a datum and then tie the second face to it with a flatness or parallel tolerance. If it is a large area then the flatness can be specified as X-mm per linear length

Yes, just a note. "set gap 0,50 on assy"

As above, geometric tolerancing. But that assumes everyone understands it!
If not, a note should do.

making one face a datum and then tie the second face to it with a flatness or parallel tolerance. If it is a large area then the flatness can be specified as X-mm per linear length

My first thoughts also, although I'm a little sketchy (moved into construction a few years ago).

making one face a datum and then tie the second face to it with a flatness or parallel tolerance. If it is a large area then the flatness can be specified as X-mm per linear length

this

have you considered asking the poor sod(s) who'll be asked to work out how to measure it?

you have a Metrology engineer/department, right?

(Metrology isn't a dirty word)

or, if you're not going to measure it anyway, why bother dimensioning it?

(there's a drawing on my desk right now, one feature is an internal diameter, 90mm, plus/minus 6 microns, 6! microns!. This feature is removed entirely before assembly, no-one knows what it's for, but there it is, plus minus 6 microns, then it gets turned into swarf)

ahwiles point is a good one. It depends how important the dimension is and what kit they have in inspection. Anything can be measured, given time and money, but is it worth it?

have you considered asking the poor sod(s) who'll be asked to work out how to measure it?

it is those poor sods who are saying it is dimensioned wrong as one surface has a datum on it when there is no datum, the 2 surfaces are only relevant to each other, either can be 'moved' during assembly to achieve the required flush so neither can be a datum.

are these poor sods of which you speak metrologists, or fitters?

sometimes, a datum feature is obvious, it's the big flat face that the part will seat on, and-or it's the diameter that'll determine the centre position.

but, othertimes, the datum feature isn't obvious, you just have to pick one and go with it.

without more info, your example sounds like the latter...

think of it as a reference rather than a fixed datum then, it still ties one part in position relative to the other

think of it as a reference rather than a fixed datum then, it still ties one part in position relative to the other

so how would you dimension it on the drawing then? That is exactly what the situation is but because the gtol symbol used is a datum is what causes the confusion. is there a gtol symbol that can be used that means reference rather than datum?

hang on, sketching...

or, i just found this:

[img] [/img]

is this roughly what we're talking about?

That's exactly what we're talking about

to be honest, I use Solidworks and insert a datum feature onto one surface which automatically allocates a letter of the alphabet. The second surface is then referenced to that with the gtol. This can be repeated on various areas of the drawing with a separate Letter allocated each time it's used. I have never had that questioned by a supplier. I assume they are saying that a datum surface should be an absolute fixed in space surface?

votchy - Member
That's exactly what we're talking about

limited as we are by this channel of communication, how about this:

next to the word 'gap' put a dimension/tolerance

next to the word 'flush' put a dimension/tolerance

tidy it up a bit so the extension lines don't touch the part, keep text off the part wherever possible, job done...?

(the concept of gap-and-flush is widely understood, the tricky bit is explaining how to measure it, and getting that process signed off)

I think you can call one of the surfaces a datum, but my GT British standards are at work, and I'm not.

Alternatively, if the datum symbol is the stumbling block then just add a note with a leader pointing to the reference face and reference the geometric tolerance to that. The note can be any reference number, letter or text

You'd have a drawing similar to Arwiles' sketch, but the two surfaces would be on nominal and each of the two panels would have a plus and minus tolerance on them. You would also supplement this with a note to specify any maximum step height and the direction of that step hight as it might be different in the two directions e.g. on aircraft you have an into wind step and a down wind step and you can have a much larger tolerance for the down wind step compared to an into-wind step.

I don't have an example to show, but you need to define the tolerance of each surface independently from a nominal and a max step between the two.

For all our assembly drawings all dimensions are reference and this would be handled in a note or in a manual...
The piece part drawings and associated tolerance stack would ensure that the parts ended up where they are supposed to be...
I fear that one of our shareholders also owns a paper mill and it makes me sad when we go to reviews...

wobbliscott,
That's interesting about the into wind and down wind. I've had the same sort of problem in the past as the OP and tried to do it with GT symbols (and failed), but are you saying you do it with notes?

Guys,

many thanks for all your inputs and ideas/suggestions. I appreciate its difficult to get things across on a chat forum but I am grateful for all your help, just need to get 'management' buy in to your ideas.

votchy