Datum

[VD2108]

Hi everyone,

We have these triggers that are dimensioned and toleranced as shown in the attached image. Recently we have switched the supplier and received parts that are not good. The main problem is that the center of the ears (slot) does not align with center of the tab (0.43 dimension) at the bottom. The part is not straight. The goal is to have centers of slot and tab to have controllable alignment.

I am thinking to add Datum A to slot (0.416 dimension) and location/position tolerance to tab (0.43 dimension) referencing Datum A.

Am I doing this right?

Thanks,
Vadim

 

[3DDave]

PH - unless the slot is an interference fit, it doesn't control orientation. Since the primary datum (typically**) provides orientation control, the slot is not an appropriate choice for the primary. It would be different if the slot was an interference fit and the pin had a lot of clearance, but that's not been my experience with nozzle handles.

**I say typically as it's possible for a sphere to be primary and it controls no orientations, but it does control 3 location degrees of freedom.

 

[powerhound]

I agree with most of what you've said, Dave. My concern was that the cross hole was really short relative to the length and how far away the other width was. Considering the function, I didn't see any disadvantage to using the slots as primary since the datum would encompass more of the part, and the holes as secondary, with appropriate controls of course.

I don't know what you mean by primary datums typically providing orientation control. I've never thought of it that way. Can you elaborate on this a little? I see planes used as primary more than anything else. So to me, a plane is easily typical. Since my recommendation used a plane as primary, I'm at a loss as to what you mean.

John Acosta, GDTP Senior Level
Manufacturing Engineering Tech
SSG, U.S. Army
Taji, Iraq OIF II

 

[greenimi]

I am in Dave's camp in this particular issue. Anybody else is joining us? Or John's camp for that matter?
As I stated: one is driving and one is driven. In my opinion the upper width and the cross holes are driving and the bottom width/slot is driven into its position.

[COLOR=][/color]

 

[drstrole]

Not knowing more about the assembly than presented in this forum, I would choose the cross holes as primary and the width as secondary. The cross holes locate the handle on a pin in the assembly and would restrict degrees of freedom on two planes. The holes IMO are driving the design requirements.

Drstrole
GDTP - Senior Level

 

[greenimi]

Drstrole,
That was my thinking too. I said even the same thing in my previous posts.
So one more in my camp.

 

[powerhound]

I'm not actually arguing my position over yours green. I agree that the cross holes are the functional feature. I stated that my concern was that the axis was extremely short relative to the distance between the widths. I thought a more stable primary datum would be the plane between the two widths. Doing it your way controls the location of the .43 width perp to the holes and centered on the .416 width, in that order. My suggestion controls the perp of the holes relative to the plane established by both widths simultaneously. Both work. I feel mine is easier to achieve and inspect with no negative impact on the function but either way, I think both are acceptable solutions.

BTW, I thought you said it was Dave's camp...

John Acosta, GDTP Senior Level
Manufacturing Engineering Tech
SSG, U.S. Army
Taji, Iraq OIF II

 

[greenimi]

Re: " I thought you said it was Dave's camp.."

Yes , you are right here!

 

[3DDave]

I'm actually in the alternative camp which suggests locating the handle with a compound of the upper width, the lower width and some other anti-twist combination to orient the holes.

The problem with using the holes alone as the primary is that it's not possible to indicate the float allowed in the mating part without causing secondary problems.

If the holes are used as a datum RFS, then there's no float. If the holes are used as a datum MMC, then the hole diameter, which is used RFS by the pin, is not getting the benefit of the slop in the mating part.

OTOH if the width of the slot with an MMC callout (min size is the tightest fit)
+ the extended tab with an MMC callout (max width is tightest fit)
+ orientation targets (RFS)

then the pin holes can have the location and orientation tolerance allowed by the mating part.

If a user wants to take some tolerance away, then make all the datum references RFS.

This arrangement is the equivalent of putting the trigger into the handle and match-drilling the holes through the pivot hole in the handle.

Simple prismatic parts don't usually get this sort of treatment, but irregular parts like this can end up with more complicated datum schemes.

 

[powerhound]

I'm actually in the alternative camp which suggests locating the handle with a compound of the upper width, the lower width and some other anti-twist combination to orient the holes.

So this actually sounds exactly like my suggestion. I guess we're all in this camp together.

John Acosta, GDTP Senior Level
Manufacturing Engineering Tech
SSG, U.S. Army
Taji, Iraq OIF II

 

[3DDave]

PH - except you missed that the features you selected are unstable and need a third location to stabilize the part. And if your suggestion was familiar, it's because I made it first.

"I would make a primary datum of the an RFS with targets at the width of the tab, the width of the slot and some other location along the trigger."

 

[powerhound]

That's kind of weird because I don't see you posted at all before I made the suggestion. Maybe you didn't read the thread from the beginning. Also my suggestion completely constrains all degrees of freedom so I don't know where this idea of it being unconstrained is coming from.

John Acosta, GDTP Senior Level
Manufacturing Engineering Tech
SSG, U.S. Army
Taji, Iraq OIF II

 

[3DDave]

PH - to back up to a question you posed about orientation.

Let's say there's .125 nominal clearance with the slot width and .5 deep engagement.
Let's also say there's 0 nominal clearance with the pressed in pivot pin and .5 engagement.

Which is the more stable control?
Which affects the path of the trigger as it turns?

However - using the pin hole as a datum prevents the convenient transfer of clearance in the mating part as tolerance to the hole location.

If the handle is fixed in a frame of reference such that it clears the mounting boss, fits into the track, and is not twisted in the frame, then the cylinder of the pivot pin has limits on its location and orientation relative to both parts, limits which can be applied as orientation and location tolerances on the pivot pin holes.

 

[powerhound]

To be specific about what I've proposed, it's not just a single width that is used as the primary datum, it's both widths referenced as a compound primary datum feature. The secondary datum feature would be the coaxial cross holes, and the tertiary is just some feature to stop rotation around the axis of the cross holes. I agree that there will be clearance around both widths so referencing the A-B plane at MMB is certainly in order. Since the rest of the handle still needs to be controlled (the edges and curvature, perhaps with profile of a surface) I think it would be easier to control the outside profile parallel to a plane that extends from one end to the other, rather than perpendicular to a short axis at one end of the part. All that is left to do is apply the appropriate perpendicularity control to the cross holes relative to the compound primary datum.

Which is the more stable control?

They are both stable. I think we have different ideas of what stable really means. Allowing datum feature shift does not make for an unstable DRF.

Which affects the path of the trigger as it turns?

Definitely the pin. That has never been in question.

It's hard to ignore the fact that you're back to the "pin as primary" argument. This is your second flip flop in the same thread. You're just arguing to argue aren't you?

John Acosta, GDTP Senior Level
Manufacturing Engineering Tech
SSG, U.S. Army
Taji, Iraq OIF II

 

[pmarc]

Am I the only one who thinks that there is not enough information about how the trigger interacts with the mating parts (pin and frame) to constructively discuss about defining proper datum features, in proper order, and at proper material boundary conditions?

 

[3DDave]

PH - your method is over-constrained.

A pin constrains 4 degrees of freedom. a plane does 3. So that's 7 of a possible 6 required. Then you add a third point, so there are 8.

To 'work,' some of the controls need to be ignored.

Width, first constraint:
Z u v

Pin, secondary
X Y u v <- ignore u and v, the angles the pin is the primary control for, or ignore the initial u and v, which requires instability in the primary constraint)

Point/line, tertiary
w.

So the pin, the feature that controls the installed orientation of the part, when used as a secondary, is ignored as the primary orientation control that is used to prevent the tab at the other end of the trigger from dragging along the side of the frame.

I have two equally useful positions - either the pin interface is primary, as it is the feature that controls the most degrees of freedom or it isn't used as a datum at all and is derived from the functional requirements for the positioning of the trigger.

In the fabrication of the trigger, and the particular problem here, there is apparently a failure to appropriately locate the holes following the forming operation. This suggests it is best to create the holes after all other features are formed, which prevents using the holes as a datum reference.

 

[pmarc]

Dave,
It is interesting what you are saying about degrees of freedom, because if I were to follow this logic in case of figure 4-2 in Y14.5-2009 for example, it looks like 9 DOFs would be constrained. I must be missing something, right?

 

[powerhound]

You've completely lost me now, Dave. Never mind, I give up.

John Acosta, GDTP Senior Level
Manufacturing Engineering Tech
SSG, U.S. Army
Taji, Iraq OIF II

 

[3DDave]

pmarc - the apparent difference is a side effect of the standard using precise prismatics for many of its examples.

Same method:
1st datum: Z, u, v
2nd datum: X, v, w; ignore the redundant v because, unlike the pin, there's nothing indicating the part will be oriented with the side.
3rd datum: Y, u, w; ignore both u and w because, unlike the pin, there's nothing indicating the part will be forced into alignment with the smallest side.

9 degrees - 3 ignored = 6.

Note the lack of explanation in the standard for what happens to the redundant controls in secondary and tertiary, et all constraints.

Now imagine that the real part being constrained had all surfaces convex and that the second surface was known to be the primary orientation source, but the large surface was selected for being conveniently large. Then some of the primary constraints have to be ignored for the secondary ones to be meaningful.

 

[pmarc]

Dave,
If I understand your last two posts (reply to powerhound and to me) correctly, you seem to be sure that the holes play the biggest role in orienting the trigger in the assembly. I am not saying you are wrong, I would just like to understand why you think this is the case. Because, as far as I see, apart from knowing that the width of the opening in the trigger is .416 +.015/-.000, we have not been provided with any meaningful information about dimensions and tolerances in the interface area? Do we know the size and tolerance of the holes in the trigger? What is the size and tolerance of the pin that goes through those holes? What is the size and tolerance of the width of the tab that mates with the .416 opening?

Is this standpoint simply based on your experience with nozzle handles?

 

[3DDave]

Pmarc, I didn't reply to your inquiry about being the only one ..., because it's an obvious observation that doesn't affect how various requirements might be analyzed. I did reply to your remark about Figure 4-2, but see no hint of its consideration in your latest inquiry.

Based on the apparent surprise the OP had at how misaligned the holes in the failing part were, following my suggestion about their affect, I'd say the hole alignment was of previously unrecognized importance.

Is your experience that high duty-cycle pivots are very sloppy and depend on secondary guides?