Pattern of features

[Gubo]

Hi all,
i have a tolerancing problem similar to this example:
a. 5 holes on a flat surface, along a line;
b. the distance between each hole should be 10 +/- 2
c. the distance between the fisrt and the last hole should be 40 +/- 5

With dimensional tolerancing only the 2 tolerances don't stack up but I need them to be exactly 2 and 5, not bigger nor smaller.

Ideas?

thansk

 

[ringster]

Disregard your tolerances as stated and determine what the mating feature will allow tolerance wise.

 

[Gubo]

That's what the mating feature allows +/-2 for the hole and +/-5 for the overall length. For manufactugin purposes I can't tighten the +/-2 to match the +/-5 overall tolerance

 

[ewh]

I would use a "non-cumulative" approach, referencing the distance between holes and using a toleranced hard callout on the overall pattern, such as "(4X 10 =)40±5".
I am interested in learning how others would approach this.

"Good to know you got shoes to wear when you find the floor." - [small]Robert Hunter[/small]

 

[Gubo]

Can't then the part be manufactured with the distance between holes outside the +/-2?

 

[ewh]

No, the max between holes would be 1.25 (5/4 spaces).

"Good to know you got shoes to wear when you find the floor." - [small]Robert Hunter[/small]

 

[ringster]

Heck,

I would just specify match locate at assy and let it go.

 

[drawoh]

Gubo,

Apply your four hole to hole dimensions at ±2. Apply a dimension from the first to the last hole with the ±5 tolerance. The tolerance provides a requirement, so this is not a reference dimension, even though the dimension value is redundant.

How do you control your holes moving side to side?

You should think through your requirements a bit. I cannot visualize why you would want tolerances like that. If you explain to us what you are trying to do, I can probably explain why it will not work.

This may be a job for a composite positional tolerance.

JHG

 

[KENAT]

Wouldn't using positional more or less solve the problem? Ok it'll effectively restrict the overal length tol tighter than +-5 but that wouldn't conflict with stated requirement.

As Ringster says though, I'd be interested in what function/tol schem on the mating part makes this necessary.

KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at posting policies:

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

What is Engineering anyway: faq1088-1484

 

[Gubo]

Thanks for the suggestions!

The general problem is that for manufactugin purposes i can't tighten any tolerance.

The mating part is similar to a part with 2 pins that mate with each hole 10+/-2 aside. Then i need the mating part to be in an area 40+/-5 in length.

In reality the problem is more complicated, i have 48 spiral features, that's why if i use a reference dimension between each feature the tolerance becomes too tight. Or the other way around, if i tolerance each feature the overall length can vary too much.

thanks

 

[drawoh]

Gubo,

Your tolerancing scheme does not work.

Let us arbitrarily select one hole to locate the two parts together. How about the first hole?

There is a very limited need to locate accurately the first hole in each part. You need the second holes to keep the second part rotating on the first. These holes have to be close enough that your bolt will pass through both pieces. You could slot them.

Each and every additional hole must be located accurately in two dimensions. Probably, you need to enlarge your clearance holes. Slotting might work, depending on what causes your accuracy problems. Maybe you only need two fasteners!

This is the scenario addressed by composite positional tolerances. The hole pattern must be accurate enough to locate the two parts together. Possibly, this is not very critical. The holes within the pattern must be accurate enough to accept your fasteners.

JHG