A question regarding the right way of defining a set of datums

[eli28]

hello everyone,

I am new in this forum, so hey to everyone.
I would like to consult with you regarding the right way of defining a set of datums in the case described in the attached pdf.
Link

On the first page you can see the general assembly for understanding what the design intent might be.

On the 2nd and third pages I suggest options of defining a set of datums to the main central mandrel.
I would like to hear what you think - which option is right, and maybe none of them and you have a better suggestion.

thanks!!!

 

[Burunduk]

If I understand the design intent correctly, despite of the relatively tight H7 tolerance on the slots (what is the reason for that?), the flats of the mandrel are going to be the main interface features at assembly, constraining degrees of freedom. Therefore I tend to recommend the option on page 3, only that I would probably make B and C a combined datum B-C (secondary) for the position control(s) on the slots. By the way, why don't you treat the 8 slots as a pattern?

I am also interested in other opinions on this.

 

[eli28]

Hey and thank you for replying.
As for your 1st question - the tight fit is for restraining each part/section of the mold laterally. If we try to analyze it, each section is laid on the main planes of the mandrel, restrained longitudinally by tightening it to datum A and finally reatraining it laterally by the slot.

I also thought of defining datum B-C but after reading the ISO standard I found that this is possible only if B and C are coplanar.

I would be happy to hear What is the advantage of defining the group of slots as a pattern.

 

[greenimi]

I also thought of defining datum B-C but after reading the ISO standard I found that this is possible only if B and C are coplanar.

Could you, please, provide some references here? Which ISO standard did you read where it is stating that the "compound" B-C have to be coplanar?

 

[greenimi]

This article (found on linkedin) does not seem to support your stance.

https://www.researchgate.net/public...mples_of_common_datum_imposed_by_ISO_standard

 

[eli28]

 

[greenimi]

Eli28,
Which ISO standard is that?
And (important) what year?

 

[eli28]

ISo 5459

 

[greenimi]

Which edition?

The one I got shows that even a non planar feature(s) could be used as one common datum....
Also the planar features don't have to be coplanar (but any random basic angle)

Am I correct?
Where did you read otherwise?

 

[eli28]

Can you please attach here the souurce that you mentioned?

As I said befoee I saw it in ISO 5459.
I can't find out the edition it was taken from since I took it from a presentation in which rhe edition isn't mentioned but only the standard.

Thanks

 

[Belanger]

I also thought of defining datum B-C but after reading the ISO standard I found that this is possible only if B and C are coplanar.

Just because a standard shows that example doesn't mean that it's the only allowable way to hyphenate two datum features. What we're all looking for is a statement in that standard that says that the only way to create A-B is if A and B are coplanar. As far as I know, that statement doesn't exist.

The "enhanced definition" of a common datum linked to by greenimi is interesting. But I think they push things a bit too far by trying to hyphenate two features that are nominally 90º apart! (I haven't yet read the entire article.)

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

 

[SeasonLee]

I think the attached figure from OP is 5459-2017 edition, page 80 and 81.

Season

 

[eli28]

Thank you all for taking part in our discussion. Does anyone can say with confidence what is the right answer?

By the way, I agree that as long as any source doesn't claim that A and B have to be coplanar I can't be sure doing otherwise is forbidden.