hexagon logo
Register Sign in
  • Replies 84 replies
  • Subscribers 223 subscribers
  • Views 801 views
  • Users 0 members are here
Options
Related

True Position in 3 axis???

1CMM4ME
Ok, I have read the post regarding this so called argument that my boss and I got into today. He told me that you cannot measure true position in 3 axis, as I told him he was wrong. I know it depends on the way the FCF calls it out, but I tried to explain to him that it is possible to measure TP in 3 axis.
I have seen the formulas, so I know it's possible. I just wish I could get one of you gurus to reply back explaining that it is possible and why. I see the picture but he does not. I do not like to get into pissing matches with upper management, but I have learned a h**ll of alot from you guys, and I know this is possible. So if one of you fine gentleman can back me up on this, I would very much appreciate it. Thank you very much.
Now I am going to get a cold beer.Smiley
  • Yes, BUT, there are TWO Axis thare are reportable. The IMPLIED ZERO is the second Axis. Also, the SHAPE of the feature does NOT necessarily dictate the shape of the tolerance zone. You do NOT have to use a Diametric tolerance zone for a hole if you do not want to. You CAN use spherical for a point if you want to.

    Sorry about the pic that I did not include. I will try to add it.
    Surface CAN be a feature of size, sort of, if you are dealing with stock thickness (yes, sheet metal).
  • Yes, BUT, there are TWO Axis thare are reportable. The IMPLIED ZERO is the second Axis. Also, the SHAPE of the feature does NOT necessarily dictate the shape of the tolerance zone. You do NOT have to use a Diametric tolerance zone for a hole if you do not want to. You CAN use spherical for a point if you want to.

    BTW, that is NOT a diametric True Position. That is an AXIAL true position called out Craig. If it were Diametric then the Diametric Symbol would be in front of the TP tolerance and it is not.

    Sorry about the pic that I did not include. I will try to add it.

    You are correct the callout is missing the diameter symbol, with the diameter symbol it would be an axis, the callout is bad without it. I'll edit it to make it legal. You are really reaching on the rest man lol. Your original position was what that you need three basic dimensions. That is what I am demonstrating, that you do not. Also the shape of the feature does in fact dictate the shape of the tolerance zone. you can only TP a FOS which a point is not so that would be illegal.
  • You are correct the callout is missing the diameter symbol, with the diameter symbol it would be an axis, the callout is bad without it. I'll edit it to make it legal. You are really reaching on the rest man lol. Your original position was what that you need three basic dimensions. That is what I am demonstrating, that you do not. Also the shape of the feature does in fact dictate the shape of the tolerance zone. you can only TP a FOS which a point is not so that would be illegal.

    Craig, it is not a stretch. We are just missing each other by words. I said BASIC Dimension but I meant BASIC AXIS LOCATION whether Dimensioned OR implied. You will find I do that on a regular basis. Tongue slippage man.

    The callout you posted as AXIAL is OK too. It could be a timing hole in a cam or other item. It could be that the radial component is all that we care about. It might be on a spinning item and need to be picked up by a sensor. Granted TP in that case is no different than a regular dimension but size can come into play on a round feature and we STILL only want axial. Just saying it IS possible and I have seen it enough and understood the rationale for it in those cases enough to believe it is OK. Function dictates the tolerance zone. Generally I believe you are right and that it is normal for a cylinder to have a cylindrical tolerance zone and a sphere to have a spherical tolerance zone. There are exceptions and you say they are illegal? Always illegal?

    Take for example a pierce point for a hole on a piece of sheet metal. Spherical TP for the pierce point has been used a lot by GM to determine fit to mating parts. Why can they NOT do that? Take for example a corner point. Why can I not use Sperical TP for that to control it's location in a sperical zone? Also, say I have a hole for an optical reflector that gets sensed when installed. I care about the Axial displacement to a tighter degree than I do about the radial rotation which can be within 3 or 4 degrees? Do I HAVE to change the hole to a slot to get the GD&T tolerancing that the function dictates?

    Also, please explain the criteria for reporting diametric TP when there are THREE basics controlling it's location. I am not asking because I know the answer and want to crack on you. I am asking because I know what I do and want to see what others do.

  • Also, please explain the criteria for reporting diametric TP when there are THREE basics controlling it's location. I am not asking because I know the answer and want to crack on you. I am asking because I know what I do and want to see what others do.

    As I posted earlier.

    If the FCF calls out three datums and one is the surface that the hole is normal to, then the axis of the hole (cylinder) must be within the TP tolerance zone. Based on ANSI Y14.5 you CANNOT just verify a point projected into the plane.
    Forget the CMM. Put a pin into the hole. As close as you can to where the pin enters the part indicate its position to the basic dimensions that control it's 2d location. Now, indicate the it's position on the opposite side as close to that surface (or move away from the surface the a distance equal to the depth of the hole.) You must maintain the TP tolerance at both those locations.

    In other words, you have to evaluate the AXIS of the cylinder, not just the center of it at one surface. This is just another way of controlling perpendicularity.

    I think I am agreeing with CMMGUY and VPT here.....
  • Just like John said. Three basic dimensions give the location for the axial feature, the first datum in the FCF is the feature to which the axis is to be normal. Then look at the pic I posted demonstrating this (it does not include the FCF sorry). As far as TP to a point it is illegal. It doesn't matter if GM does it or Ford or Boeing or whoever. You can't TP a point, per the standard it needs to be a FOS. Does this mean you can not make PCDMIS do it? No. Does it mean you are misinterpreting it and design intent is not being met? No. Is it a bad callout? Per the standard it is.

    There are no exceptions that I know of in regards to the shape of the tolerance zone. The shape of the tolerance zone is dictated either by the shape of the feature or by basic dimensions (i.e. profile).
  • Just like John said. Three basic dimensions give the location for the axial feature, the first datum in the FCF is the feature to which the axis is to be normal. Then look at the pic I posted demonstrating this (it does not include the FCF sorry). As far as TP to a point it is illegal. It doesn't matter if GM does it or Ford or Boeing or whoever. You can't TP a point, per the standard it needs to be a FOS. Does this mean you can not make PCDMIS do it? No. Does it mean you are misinterpreting it and design intent is not being met? No. Is it a bad callout? Per the standard it is.

    There are no exceptions that I know of in regards to the shape of the tolerance zone. The shape of the tolerance zone is dictated either by the shape of the feature or by basic dimensions (i.e. profile).

    You do not necessarily HAVE to have the feature normal to the first datum in the FCF, do you? Say you have a BASIC angle into which a hole is bored. Why is it illegal to use the BASIC angle to control the orientation or the normailty of the hole?

    I need to get the standard in my hand. See where it is called illegal. If all of this stuff is illegal then that is a mistake IMO and does not serve the community.

    Wait a minute. Page 60 of my trusty GD&T handbook Y14.M-1994 shows TP as a bi-directional control (single Axis) on ROUND features. A round feature should dictate diametric IF what you are saying is correct. I think it's OK man. It's OK to use single axis tp with round features.
    I will investigate the corner point further and may well conced it is illegal.
  • You do not necessarily HAVE to have the feature normal to the first datum in the FCF, do you? Say you have a BASIC angle into which a hole is bored. Why is it illegal to use the BASIC angle to control the orientation or the normailty of the hole?

    I need to get the standard in my hand. See where it is called illegal. If all of this stuff is illegal then that is a mistake IMO and does not serve the community.

    Wait a minute. Page 60 of my trusty GD&T handbook Y14.M-1994 shows TP as a bi-directional control (single Axis) on ROUND features. A round feature should dictate diametric IF what you are saying is correct. I think it's OK man. It's OK to use single axis tp with round features.
    I will investigate the corner point further and may well conced it is illegal.

    I don't believe it should be illegal to have an angle for the hole like you illustrated. Never said it was. Yes you can have a non-diametric tolerance zone for a FOS but you need to place the FCF in a manner that shows this. In other words the FCF can not be attached to the FOS but rather to the dimension lines. If it is attached to the FOS it is diametric. Your page 60 and mine are much different, I am in section 5.9 also demonstrated in illustration 5-42 (my page 60 is on datums). So yes you can do a non-diametric tolerance zone but where you place the FCF dictates the legallity of it.
  • I don't believe it should be illegal to have an angle for the hole like you illustrated. Never said it was. Yes you can have a non-diametric tolerance zone for a FOS but you need to place the FCF in a manner that shows this. In other words the FCF can not be attached to the FOS but rather to the dimension lines. If it is attached to the FOS it is diametric. Your page 60 and mine are much different, I am in section 5.9 also demonstrated in illustration 5-42 (my page 60 is on datums). So yes you can do a non-diametric tolerance zone but where you place the FCF dictates the legallity of it.

    Just curious as to why there would be two different ways to define that an FCF is diametric? There is the diametric symbol AND there is WHERE you attach the FCF? Why the redundancy? Just curious more than anything and since I realize you didn't actually write the spec I don't expect you to have all the answers here. Still, it doesn't make sense to have redundancy like that because it will inevitably lead to ambiguity and conflict.
  • Just curious as to why there would be two different ways to define that an FCF is diametric? There is the diametric symbol AND there is WHERE you attach the FCF? Why the redundancy? Just curious more than anything and since I realize you didn't actually write the spec I don't expect you to have all the answers here. Still, it doesn't make sense to have redundancy like that because it will inevitably lead to ambiguity and conflict.

    If your FCF is attached to a FOS and it is an axis then the FCF has to have the diameter symbol in it, you can not legally drop it. That is why I edited my drawing, it did not have it making it an incomplete callout. If your FCF is attached to anything that is not controling the axis all around or is controlling something other than an axis such as a plane you must omit the diameter symbol.

    Same would go for perpendicularity. It you tack it on a diameter it has to have the diameter symbol. If it is on a plane then you omit it.
  • OK. My illustration shows the FCF attached to the features and is called bi-directional (single axis) on page 60 of my handbook. Perhaps it is an incorrect depiction. I will get a picture of it and post it.

    Anyhow, when I was trained in GD&T (professional certificate course) one of the main rules was that EVERY Basic Dimension HAD to have a FCF associated with it or it was incorrect / undefined / uncontrolled etc. In the event that you have THREE Basic Dimensions locating a Diametric TP (Two Axis) you inevitably have an undefined / incomplete / uncontrolled Basic Dimension as ONLY two of them are significant. Which ones are significant and which ones are undefined are usually easy to discern. Sometimes they are not. I am curious as to what the people on here do when it is unclear which Dims are reported. One good example is when the feature is on a 45° angle. Which Axis is then used? Becomes ambiguous and you CANNOT use all three. In those cases I call the BASIC Dims as SET (perfect) and report the TP normal to the surface it is in. None of the dims on the print show up in the result. But, this does NOT truly satisfy the rules of GD&T - specifically that every BASIC Dimension must have an associated FCF. So, I deem it illegal as it violates the princinples of GD&T. Does that make sense?
© 2024 Hexagon AB and/or its subsidiaries. Privacy Policy | Cloud Services Agreement