Rotational Pattern of Irregular Features

[ruscag]

I need to define two patterns of shapes (slot and vane) that fit together for welding. The engineer wants to keep the clearance to a minimum.

After doing some homework, I thought to use composite profile of surface tolerance to define a PLTZF and FRTZF framework.

My problem is I don't know how to figure the upper and lower values and weather or not to use RFS or MMC and LMC.

In ASME Y14.5-2009 page 173, a two slot linear pattern is shown. But I can't seem to find an example of a rotational pattern of irregular features.

 

[pmarc]

ruscag,
I do not think you need composite profile tolerancing in your case unless you want to control perpendicularity of slot / vane relative to B tighter than location of those elements relative to A and B.

Whole RFS, MMC, LMC thing strongly depends on how your part functions. For sure per Y14.5 you are not allowed to use (M) or (L) modifier in a tolerance portion of profile feature control frame.
You can however apply (M) or (L) right after A (assuming A is a cylindrical datum feature subjected to size variation), but like I said it depends on function, especially on mating conditions of the cylinder vs. something that is mated to it.

As for profile values, if the requirement is that at min. worst-case there is no gap between slot and vane surfaces and at max there is 0.15 gap, just adjust profile values specified in feature control frames together with radii values to accomplish this, e.g.: the radii values could be that the nominal difference between them is .075 (R1=.5; R2=.575) then the equal bilateral profile values for slot and vane would be .5*.075=.0375. This would give you 0 min and .15 max.

 

[ruscag]

pmarc,

Thanks for the reply!

I thought I'd need composite since the two parts will be manufactured by two different shops. Also to keep datum A and datum C as concentric as possible I wanted to locate the pattern from the bores. Then the two are welded, finished machined and mounted on the shaft.

The challenge is six irregular shapes on the two disks with small clearance (actually 0.015" not 0.15").

The perpendicularity issue is important in the slots and near the top of the vanes where they meet. With about a tolerance of .004 (see below) too much angle and the two parts will not assemble.

We can throw modifiers out since datum A is not the final turn dia. That comes after welding.

Using your equations I get .015/2=.0075 then .0075/2=~.004.

Do you still think I can avoid composite?

 

[pmarc]

ruscag,

First, let me clarify some things:
- I was too hasty in saying about perpendicularity to B when in reality lower segments of profile composite FCF control parallelism to A and C respectively;
- I somehow did not notice that FCF for slot uses A and FCF for vane uses C as primary datum features, and therefore whole play with profile tolerance values is not so simple, because mutual relationship (coaxiality) between A and C has to be taken into account. I am afraid profile values will have to be even smaller after that.
My apologies!

But I still think you can live without composite - of course if it is acceptable for you to have parallelism of slot relative to A and vane relative to C controlled with the same value as location of those features relative to corresponding datum reference frame.

By the way, some dimensions defining form and location of slot & vane are still missing on the print. Any idea how to deal with this?

 

[ruscag]

pmarc,

Again thanks for your time and attention.

In MY haste, I think I confused datums A and B. I had it in mind to make the slots and vanes perpendicular to what is now called datum B and D respectively. But that's a minor point.

The coaxially between A and C is important for the sake of even flow thru but not for assemble-ability.

As for living without composite...The accuracy of the location and shape of the vanes is as it touches upon their function in the operation of the part is less important (tolerance wise) than the ability to assemble the two parts for welding.

If we end up with vanes that are more tightly toleranced than needed for operation so that they can be assembled...that's fine.

You wrote:
"if it is acceptable for you to have parallelism of slot relative to A and vane relative to C controlled with the same value as location of those features relative to corresponding datum reference frame."

And it is acceptable. So that means I can use one frame with a tolerance of .015*.5= ~.007?

As for missing dimensions, there are more basic X/Y/Rad sets to define each arc but they would have just clutter my example sketch.

 

[pmarc]

And it is acceptable. So that means I can use one frame with a tolerance of .015*.5= ~.007?

Yes, I would say you can use single segment frame for both profile controls, but, like I already mentioned, values for those callouts would have to be smaller due to the fact that a tolerance between A and C is having an impact on each slot/vane assembly.

Actually, without knowing the tolerance value (coaxiality?) between A and C, I am pretty sure that performing any stack-up calculation for finding a possible gap between slots and corresponding vanes is rather impossible. As far as I understood this is the only tolerance that will tie both parts together.

 

[fsincox]

I believe, a composite tolerance might be appropriate on one or both if you were also trying to also going to specify a rougher location to another radial datum feature that did not mate, like these simultaneously to each other but somehow in an upper level installation.
Frank

 

[MechNorth]

Composite profile definition changed a bit in '09. It now acts as a pattern relationship control as well, as position always has. So, if you need the vanes within the pattern / slots within the pattern controlled wrt each other (and it looks like you do) then composite profile is perfect. As for the tolerances, a look at a Limits & Fits standard (ASME, ISO, Machinery's Handbook) would be a good though not absolute reference. Suggest you map out the tolerance zones in CAD to establish the worst case relationship is ok, and tweak as necessary.

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

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[fsincox]

Jim,
How do you feel it changed? My impression has been they were the same. I would base that on figure 6-26, ASME Y14.5-1994, although I would prefer they used the word orientation to parallelism in the description text.
Frank

 

[MechNorth]

Frank,
Standard didn't support the use of composite profile for controlling a pattern before '09. It was indicated as a single feature control. The extension to pattern control brings it back on a part with position in that respect, and enhances the value of profile even more.

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

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com