Using repeating FOS to locate patterns

[prdave00]

I have a base plate to which several components (motor mounts, pillow blocks, etc.) will be assembled onto using locating pins.

Each component is located with pins and then fastened down with socket head cap screws. Some components have the same hole patterns while others do not, but the locating pin is common to all.

I want the inspection to be straight forward using our contract shop's Cordax 1810 CMM with 3D Workshop (v. 1.0.1.5). I fear I may have inappropriately ruled out composite FCF's and went with the SEP REQ annotation. Am I doing myself a disservice here? Can I get away with the INDIVIDUALLY annotation even though the patterns are not identical?

I'm sure there are other "issues" with the prints but wanted to focus on the hole/bolt pattern strategy.

 

[drawoh]

prdave00,

The standard locating pin design is for a mating hole and mating slot. The alternative is to have two rounding mating holes and for one pin to be diamond shaped so that it locates rotation, only. If you do it this way, your primary concern with tolerances is to locate each of your components. I see no benefit to locating the individual patterns.

You can locate round holes over your round dowel pins. The quick and dirty process is to assign a positional tolerance of 1/2 the clearance around the dowel. I think this would be way too accurate for fabrication.

A composite tolerance allows your fabricator to use a template of some sort to drill hole patterns. This could be very useful in a welding shop or for someone working with a big punch. I do not see a machine shop benefiting from it.

JHG

 

[MechNorth]

I was going to ignore the use of multiple pins/holes vs pin/hole with pin/slot. Drawoh went there, so enough said.

As for SEP REQT & INDIVIDUALLY ...
The Principle of Simultaneous Requirements indicates that each feature or sub-pattern of features which bares the same datum reference frame (identifiers, sequence and modifiers) becomes part of a single pattern that can be inspected simultaneously; i.e. they are all related to one another. The invocation of SEP REQT breaks that overarching pattern but does not free the features within the sub-pattern; note that a (sub-)pattern is established when a quantity is specified for a given feature. OK, so for Detail A, you have 4X the c'bored hole which creates a pattern of 4 c'bored holes BUT because of the Principle of Simultaneous Requirements, that pattern of 4 holes is also tied to the 4X tapped holes in Detail D, which by the quantity of 2X on the detail, also ties the second grouping of 4 tapped holes together ... it's WAAAAAY nested. If you don't need those sub-patterns to be related, then break the pattern with SEP REQT otherwise their relationships will have to be verified by inspection.

As for INDIVIDUALLY, that's only used when you have a repeating pattern of sub-patterns wherein a datum is established within and for the control of each INDIVIDUAL sub-pattern. That may be applicable for your Detail D's, where the 4X tap holes may need to be controlled more precisely to the central pin hole rather than to the labeled datums A,B,C.

Two different intents and two different functionalities.

Jim Sykes, P.Eng, GDTP-S
Profile Services

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[prdave00]

Drawoh,

I was (halfway) hoping my use of a single locating pin would go without notice. I actually spent an hour convincing my coworker that a round-diamond pin locating methodology was superior and then proceeded to spend another hour convincing him to go back to a single round pin. I convinced myself that a single pin could be used to accurately locate the bracket and the then the shafts that run through the bearings/bores in the brackets would "force" orientation instead of the fixed orientation of the bracket biasing the shaft. I also convinced myself that the location and orientation of the hole to accept the diamond pin is not trivial. In one direction it still needs to be toleranced such that it does not force an orientation that might not be ideal for bearing life. So instead of having 2 highly toleranced holes per component (e.g. 1 hole in the base, 1 in the bracket), I'd have 4 (e.g. 2 hole in the base, 2 in the bracket). Is it just a matter of assigning a loose locational clearance fit to the mating hole in the component to accept the diamond pin?

I fear I have made this drawing too complicated. Would it be advisable to just assign a location tolerance to each hole without trying to get clever with composite tolerancing, invoking SEP REQ, etc.

 

[drawoh]

prdave00,

A rotating shaft does not sound to me like a good locating device.

The whole point of diamond pins is that they contact only on the sides of the hole, and they control only rotation. The distance between the two dowel holes is not absolutely critical. You could be in trouble if your machine shop cannot get the holes perpendicular to your plate, but if you are having this problem, it is time for a new machine shop.

You can calculate the positional accuracy of your module based on the clearance hole and the positional tolerances of both the dowel and the clearance hole. This is equivalent to a screw assembly as per ASME Y14.5. Your rotational accuracy also is calculatable based on the positional tolerances and clearance holes of both dowels.

If orientation is very much more critical than location, perhaps you could use parallel spacers to position your modules.

JHG

 

[KENAT]

On the location pin topic, another option is to use 'pins' with some flexibility such as a roll or similar spring pin. Then you can use simple 2 hole pattern. It may not be suitable for all situations, since if nothing else it's not quite as rigid, but worth keeping in mind.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​