Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations MintJulep on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
Positional Tolerancing
- Thread starter aldumoul
- Start date
pmarc
Mechanical
- Sep 2, 2008
- 3,248
Is there any functional reason that the 2 holes selected as datum features B and C are more important than the other two? If not, I would suggest assigning whole pattern of 4 holes as datum feature B and locate other holes relative to it. If you do this, datum feature C will not be needed at all.
If you are thinking of adding (M) modifier to B letter in positional feature control frame for 4 holes on the other side of the part, yes, you can do this, and as a matter of fact I think it is better choice than current lack of (M) modifer.
If you are thinking of adding (M) modifier to B letter in positional feature control frame for 4 holes on the other side of the part, yes, you can do this, and as a matter of fact I think it is better choice than current lack of (M) modifer.
CheckerHater
Mechanical
Aldumoul,
Your question is kinda “multi-layered” so it cannot be answered in one simple answer.
First, you already have bonus tolerance as you specified (M) requirement in your FCF for .573/553 hole.
If what you mean is “datum shift”, then you also have to specify (M) next to your datum reference as well.
On the matter of datum [C] there are different schools of thoughts:
Gage people and people who believe in “degrees of freedom” approach will tell you that you need your tertiary datum to fully constrain your part or to “clock” it.
My personal opinion is that your hole always has its X and Y position and, once you came to X and Y point, you have to pick direction to drill, so I am with 3 datum crowd.
CMM people who believe that you only have to put your part on the table and then you can measure anything will tell you that you datum [C] may not be necessary. So let's leave datum [C] for future consideration.
Now about datum .
I see your pattern of .573/.553 holes is dimensioned in relation to it. But is datum itself well defined?
It looks like it is free to move along your bolt circle as it pleases, because it is not dimensioned to any other reference on the part. Also, position of the center of your BC itself is not clear as well.
To be completely sure what the hole B is about I would also add perpendicularity requirement for hole B in relation to datum [A].
After position of datum is clarified the destiny of datum [C] may become more clear.
Also, this is not the only possible way to control your part. For example, you can use the entire pattern of 4 .666/.646 holes as a datum as well.
Naturally, my opinion is not the only one available and not necessarily correct one.
Your question is kinda “multi-layered” so it cannot be answered in one simple answer.
First, you already have bonus tolerance as you specified (M) requirement in your FCF for .573/553 hole.
If what you mean is “datum shift”, then you also have to specify (M) next to your datum reference as well.
On the matter of datum [C] there are different schools of thoughts:
Gage people and people who believe in “degrees of freedom” approach will tell you that you need your tertiary datum to fully constrain your part or to “clock” it.
My personal opinion is that your hole always has its X and Y position and, once you came to X and Y point, you have to pick direction to drill, so I am with 3 datum crowd.
CMM people who believe that you only have to put your part on the table and then you can measure anything will tell you that you datum [C] may not be necessary. So let's leave datum [C] for future consideration.
Now about datum .
I see your pattern of .573/.553 holes is dimensioned in relation to it. But is datum itself well defined?
It looks like it is free to move along your bolt circle as it pleases, because it is not dimensioned to any other reference on the part. Also, position of the center of your BC itself is not clear as well.
To be completely sure what the hole B is about I would also add perpendicularity requirement for hole B in relation to datum [A].
After position of datum is clarified the destiny of datum [C] may become more clear.
Also, this is not the only possible way to control your part. For example, you can use the entire pattern of 4 .666/.646 holes as a datum as well.
Naturally, my opinion is not the only one available and not necessarily correct one.
- Thread starter
- #4
CH,
The simultaneous requirement says "you do not". We do not even need datum "B" really. I am not a big fan of it myself but it is the "law of the land" if you are using ASME. What I do not like even more is playing by different sets of rules just depending on who is playing.
Really, I tend to more like pmarc's solution, except whatever mounts on the opposite side may actually only mount on it's own face which is lost here (if true), do all you guys tend to do that "for simplicity, so to speak"?
Frank
The simultaneous requirement says "you do not". We do not even need datum "B" really. I am not a big fan of it myself but it is the "law of the land" if you are using ASME. What I do not like even more is playing by different sets of rules just depending on who is playing.
Really, I tend to more like pmarc's solution, except whatever mounts on the opposite side may actually only mount on it's own face which is lost here (if true), do all you guys tend to do that "for simplicity, so to speak"?
Frank
- Thread starter
- #6
CheckerHater
Mechanical
Fsincox,
You confused the heck out of me.
The plate on the left view is under-dimensioned.
The features on the right view are dimensioned in relation to the hole which actually has no location, it’s dangling.
How exactly simultaneous requirement will help?
You confused the heck out of me.
The plate on the left view is under-dimensioned.
The features on the right view are dimensioned in relation to the hole which actually has no location, it’s dangling.
How exactly simultaneous requirement will help?
CheckerHater
Mechanical
Aldumoul,
This is the basic idea how the entire pattern may become your datum
Here [X] and [Y] are temporary datums you use to establish datum framework using plane [A] and pattern
After that you can dimension the rest of your part in relation to [A] and (no [C] at all)
This is the basic idea how the entire pattern may become your datum
Here [X] and [Y] are temporary datums you use to establish datum framework using plane [A] and pattern
After that you can dimension the rest of your part in relation to [A] and (no [C] at all
)
pmarc
Mechanical
- Sep 2, 2008
- 3,248
aldumoul,
Not that I do not want to help you, but I guess you first need good GD&T handbook and Y14.5 standard to extend basics. The way how you switched your datum features from features on one side of the part to features on the other side makes me thinking that you have not identified functional relationship properly yet.
Frank,
I agree - most likely another datum feature on the opposite flat face should be defined and chosen as primary for positioning of 4 dia. .563 holes.
CH,
I believe Frank will explain his intent better than me, but I guess he was refering to your comment whether 2 or 3 datums are needed and went even further saying that only single datum may be defined and simultaneous requirement will do the rest (of course assuming that every feature in this part is oriented/located to A through geometrical tolerances). This approach is shown in one of Tec Ease's tips and has been discussed quite often in different threads here.
Not that I do not want to help you, but I guess you first need good GD&T handbook and Y14.5 standard to extend basics. The way how you switched your datum features from features on one side of the part to features on the other side makes me thinking that you have not identified functional relationship properly yet.
Frank,
I agree - most likely another datum feature on the opposite flat face should be defined and chosen as primary for positioning of 4 dia. .563 holes.
CH,
I believe Frank will explain his intent better than me, but I guess he was refering to your comment whether 2 or 3 datums are needed and went even further saying that only single datum may be defined and simultaneous requirement will do the rest (of course assuming that every feature in this part is oriented/located to A through geometrical tolerances). This approach is shown in one of Tec Ease's tips and has been discussed quite often in different threads here.
CheckerHater
Mechanical
Pmarc,
Yes, I remember those “discussions”; unfortunately majority of people on this forum do not fully comprehend the idea of using degrees of freedom to correctly define the datum framework.
As you are one of the smartest people around, I was hoping to discuss it with you sometime, but probably not today.
As a parting gift I leave you the following statement to think about:
“Position is a fixed related tolerance when the referenced datums lock all non-redundant degrees of freedom for the tolerance zone.”
Yes, I remember those “discussions”; unfortunately majority of people on this forum do not fully comprehend the idea of using degrees of freedom to correctly define the datum framework.
As you are one of the smartest people around, I was hoping to discuss it with you sometime, but probably not today.
As a parting gift I leave you the following statement to think about:
“Position is a fixed related tolerance when the referenced datums lock all non-redundant degrees of freedom for the tolerance zone.”
- Thread starter
- #11
Thank you CheckerHater. The idea of using the pattern as a Datum is still confusing. If I were to inspect the .563 holes how do I relate back to a pattern? Unless one hole in the pattern is Datum B? Through my inspection experience I was always taught to align the part on two holes (Datums).
I generally prefer to assume the missing dimensions are left out on purpose, as the standard often does this and states it will.
Yes, pmarc, I agree all features would need to actually refererence the same framework, that was kind of the point I was making, if you simply it that far why not go all the way!
Yes, pmarc, I agree all features would need to actually refererence the same framework, that was kind of the point I was making, if you simply it that far why not go all the way!
CheckerHater
Mechanical
Well fsincox, when it comes to dimensioning everything from the same datum framework, I am all for it.
Unfortunately one single datum does not constitute “framework” and overusing simultaneous requirement principle may be a bit of a stretch.
Please read the following quote from T-E tip carefully:
Whatever happened to GD&T being process-independent?
You see, tec-ease sometimes allowed itself 70% margin of error (as you can see from enclosed graphics)
On more serious note, I also highlighted couple of statements that tec-ease probably doesn’t want you to see anymore.
To conclude, it appears like sim. requirement technique is more suitable for “look how smart I am” situation rather than real-life application (sorry, Tec-ease). At the very least you can say that even tec-ease itself does not recommend the technique as universally acceptable.
Unfortunately one single datum does not constitute “framework” and overusing simultaneous requirement principle may be a bit of a stretch.
Please read the following quote from T-E tip carefully:
Tec-ease said:
Whatever happened to GD&T being process-independent?
You see, tec-ease sometimes allowed itself 70% margin of error (as you can see from enclosed graphics)
On more serious note, I also highlighted couple of statements that tec-ease probably doesn’t want you to see anymore.
To conclude, it appears like sim. requirement technique is more suitable for “look how smart I am” situation rather than real-life application (sorry, Tec-ease). At the very least you can say that even tec-ease itself does not recommend the technique as universally acceptable.
CH,
I am not a fan of a single framework unless it actually works that way, in case you can't tell.
If you are going to simplify/compromise, where do you stop. You are really saying I am just simply going to be led by “what the shop wants” and we are all back in the same position we started at before GD&T, IMHO.
I am not a fan of a single framework unless it actually works that way, in case you can't tell.
If you are going to simplify/compromise, where do you stop. You are really saying I am just simply going to be led by “what the shop wants” and we are all back in the same position we started at before GD&T, IMHO.
CheckerHater
Mechanical
Understood.
About function - I assumed OP attached datum A to mounting surface.
Then I was thinking also of manufacturing.
You definitely want to machine all the features on one side, then flip it over and do the other side using first one for fixturing – there are not so many options.
So, flat as your primary, hole for location, hole for clocking (or the entire hole pattern at once) – cannot go wrong with that.
About function - I assumed OP attached datum A to mounting surface.
Then I was thinking also of manufacturing.
You definitely want to machine all the features on one side, then flip it over and do the other side using first one for fixturing – there are not so many options.
So, flat as your primary, hole for location, hole for clocking (or the entire hole pattern at once) – cannot go wrong with that.
- Thread starter
- #17
Pmarc... extend the basics? As far as using the entire pattern as a Datum, one should look at inspection in real-time... Is it a good idea to take the average of 4 holes and establish a Datum? If you try to construct an X, or Y axis on a CMM using 4 holes, it wants to establish a plane perpendicular to what you are trying to achieve. I can see the theory but in real-life... this is painful on a CMM. I prefer step by step instructions... The average of the pattern's holes is in the new ASME standard. Unfortunately, it can be difficult to apply, but I'm working on it.
CH,
Not the first time Don (Tec-ease) and I don't see eye to eye on things. He has a lot of good stuff. Remember, from my experience, people who make a living selling the standard sometimes may have interests other than just advancing pure theory.
I am not a big fan of the simultaneous requirement, rule #1, and a few other ASME concepts, but, it is the law of the ASME land and I REALLY don't like when people play a game and do not require everyone to play by the same set of rules.
Not the first time Don (Tec-ease) and I don't see eye to eye on things. He has a lot of good stuff. Remember, from my experience, people who make a living selling the standard sometimes may have interests other than just advancing pure theory.
I am not a big fan of the simultaneous requirement, rule #1, and a few other ASME concepts, but, it is the law of the ASME land and I REALLY don't like when people play a game and do not require everyone to play by the same set of rules.
pmarc
Mechanical
- Sep 2, 2008
- 3,248
aldumoul,
My comment to the extension of basics was more about your sudden decision of switching datum features from one side of the part to another, and not about using whole pattern of holes as datum feature B. I did not meant to sound offensive, my apologies. Please correct me if I am wrong, but from what I see you are rather trying to choose datum features to make your CMM inspection process easier and not to grasp real functionality of the part. I firmly believe it is not the best approach.
As for whole rectangular pattern of 4 holes being datum feature B, there is one very important distinction that has to be made and which has been very often dissembled by CMM operators. Two datum planes (horizontal and vertical) crossing at the center of the pattern shall be derived from 4 theoretically perfect datum feature simulator cylinders of certain theoretical size and certain theoretical distance apart, and not the holes themselves. This would actually make datum planes establishment insensitive of what happens with the holes.
Unfortunately your both sketches do not provide sufficient information in order to be able to explain my statement above in plainer words. I would have to assume some tolerances in order to clarify this. If you are interested, I can try to explain it - just let me know.
My comment to the extension of basics was more about your sudden decision of switching datum features from one side of the part to another, and not about using whole pattern of holes as datum feature B. I did not meant to sound offensive, my apologies. Please correct me if I am wrong, but from what I see you are rather trying to choose datum features to make your CMM inspection process easier and not to grasp real functionality of the part. I firmly believe it is not the best approach.
As for whole rectangular pattern of 4 holes being datum feature B, there is one very important distinction that has to be made and which has been very often dissembled by CMM operators. Two datum planes (horizontal and vertical) crossing at the center of the pattern shall be derived from 4 theoretically perfect datum feature simulator cylinders of certain theoretical size and certain theoretical distance apart, and not the holes themselves. This would actually make datum planes establishment insensitive of what happens with the holes.
Unfortunately your both sketches do not provide sufficient information in order to be able to explain my statement above in plainer words. I would have to assume some tolerances in order to clarify this. If you are interested, I can try to explain it - just let me know.
- Thread starter
- #20
So, the 2 planes would be from the theoretical/perfect position of the 4 holes? I think I just broke it down to make it easier to inspect (step-by-step) fashion. I've learned this approach from a guy that has been an inspector for 25 years and I've been a QE for just 1.5 years. Looking thru a recent ASME book you are supposed to inspect the holes first, then establish the average planes for datum B. which is it? It is driving me nuts lol. I have broken down and decided to go with the 4 pattern approoach as it is to the new ASME standard...
- Status
Similar threads
- Question
- Locked
- Question
- Locked
- Question
- Locked
- Question