Help with Geometric Dimensioning and Tolerancing....

[Doc]

Hi all,

I am trying to teach myself GD&T from a textbook (Fundamentals of Geometric Dimensioning and Tolerancing - Krulikowski) and would like some feedback on a drawing I've created, please see attached.

Any help/criticism would be greatly appreciated.

Regards,
Doc

 

[Tom.G]

It's been a long time, but I will take a stab at it. Please verify my suggestions before implementing.

Specify the chamfer as at "All Edges"
The center hole is dimensioned at the outer edge of the chamfer, should be dimensioned as the hole diameter.
The drawing layout does not match the preprinted legend on the lower right corner of the page. (just right of where it says "SHEET 1 OF 1")
Maybe you should add an exploded detail (and explanation?) showing the corner chamfer to the isometric view. Alternatively, you may get away with specifying the chamfer as at "All Edges and Corners"; but it's not as clear that way. ... On further thought, probably doing both is best.

You may be interested in a free download of the Drafting Room Manual:
https://www.google.com/search?source=hp&ei=7iSjW5S0MoHV8AOulqgo&q=drafting+room+manual+download

Cheers,
Tom

 

[Doc]

Hi Tom,

Thanks very much for the feedback. I've fixed up all of the problems that you identified, rookie errors as they were (amended drawing attached). Could you please let me know what you think of the GD&T symbols I have applied?

Regards,
Doc

 

[Doc]

Okay next drawing. Please see sketch 01. Let's say that I want to clamp a laser pointer in a bracket and have the beam hit a target which is exactly one metre away. The maximum deviation that I am willing to permit the beam to have with respect to dead centre of the target is 0.5mm. So over a 1000mm I only want a pointing error of half a millimeter, see sketch 02. This equates to approximately 0.03 degrees. I want to tolerance my drawing with GD&T in such a way that the manufactured part achieves this accuracy.

I have attached the drawing (bracket.pdf), please tear it to pieces. Assume that the centre of the target is 47.5mm from the mounting surface which is also perfectly flat.

Thanks,
Doc

 

[Ranger Mike]

Been there done this. Martin Hamar pioneered machine tool laser alignment about 30 years ago. He started with a laser beam and ended up with a three plane laser scanner that projected 3 laser planes precisely perpendicular to each other.

A single laser beam does not tell you much. You must make it steerable so the projection is perfect to the mounting surface. Precise machining is costly. The photo electric cell or target will give you an X-Y displacement from the center point but not much else. Martin split the beam when it entered the target and could read center and angularity. This is very useful in alignment of a machine tool spindle as the tool path when drilling is very important to tool life. Beams take a long time to “buck in” or align to the machine guide way so the use of a laser plane makes the set up time very short. The squareness of the laser planes was very useful in checking the 21 errors in geometry of a machine tool ( less scale calibration) .

suggest you visit hamar laser web site
rm

 

[AZFIREBALL]

Re: water bottle holder
Check the perpendicularity call-out to surface C (not A)

 

[Doc]

Re: water bottle holder
Check the perpendicularity call-out to surface C (not A)

Ah yes, I see now. Thanks!

 

[Doc]

Been there done this. Martin Hamar pioneered machine tool laser alignment about 30 years ago. He started with a laser beam and ended up with a three plane laser scanner that projected 3 laser planes precisely perpendicular to each other.

A single laser beam does not tell you much. You must make it steerable so the projection is perfect to the mounting surface. Precise machining is costly. The photo electric cell or target will give you an X-Y displacement from the center point but not much else. Martin split the beam when it entered the target and could read center and angularity. This is very useful in alignment of a machine tool spindle as the tool path when drilling is very important to tool life. Beams take a long time to “buck in” or align to the machine guide way so the use of a laser plane makes the set up time very short. The squareness of the laser planes was very useful in checking the 21 errors in geometry of a machine tool ( less scale calibration) .

suggest you visit hamar laser web site
rm

Hi Ranger Mike,

Thanks for the reply. I'm not actually going to have this bracket manufactured, the purpose of the drawing is simply for me to develop my understanding of GD&T. I'm trying to choose simple examples, I realize now that the laser bracket is perhaps not as simple as I first thought.

Do you have any comments on the GD&T I've applied to the drawing?

Regards,
Doc

 

[Doc]

Re: water bottle holder
Check the perpendicularity call-out to surface C (not A)

Revised drawing attached.

I'd like to check my understanding with the perpendicularity tolerance I've applied here: if the part is manufactured to a tolerated thickness of 20.1mm and the hole axis can deviate from the theoretically perfect orientation by 0.1mm, the maximum angular offset I can accept from the manufacturer would be approximately 0.29 degrees: is that correct? Not that you would physically measure an angular offset when inspecting the received part, but just for understanding conceptually.

Another question: isn't the tolerance I have specified on the position of the hole, and the perpendicularity of the hole contradicting each other here? I'm saying that the position can be off in a diametrical region by half a millimeter, BUT I'm also saying the maximum perpendicular tolerance can be no more than a tenth of a millimeter. So either I need to relax the perpendicularity tolerance, or tighten the position tolerance of the hole: am I understanding this correctly?

 

[Tom.G]

re waterbottle holder rev01:
The notes have lost the title of "NOTES:"
The orientation legend next to "SHEET 1 OF 1" has is now missing a circle
And @AZFIREBALL caught the perpendicularity call-out to surface C (not A) that I missed

re bracket:
I see some practical problems with actually using this design for its stated purpose. Since you are asking about diminsioning standards rather than mechanical design, I will leave it to those with better familiarity with dimensioning to continue the discussion here. I will continue to watch this thread, however, in case there is anything substantive I can contribute.

Cheers,
Tom

 

[Doc]

:
re bracket:
I see some practical problems with actually using this design for its stated purpose. Since you are asking about diminsioning standards rather than mechanical design, I will leave it to those with better familiarity with dimensioning to continue the discussion here. I will continue to watch this thread, however, in case there is anything substantive I can contribute.

Cheers,
Tom

I'm a very new engineer so I am more than happy to hear criticism on mechanical design if you're willing to give it.

Thanks,
Doc

 

[Tom.G]

...I am more than happy to hear criticism on mechanical design if you are willing to give it.

Well, it's somewhat off-topic from the thread title so it may get split to a new thread, but here goes.

Now that is a problem (in post #4 above) that is underconstrained! Is the Laser beam required to be perpendicular to the target? Since only the linear dimensions are given, the Laser could be positioned anywhere on a 1-meter radius hemisphere and aimed at the target.

Just to simplify things, let's assume that perpendicularity is required and the target is appropriately fixed to a beam or bed on which the Laser is also mounted.

Here are some things to consider in mounting the Laser:

1. Keep in mind that you do not have control over the how the Laser diode is aligned within its package. How will you handle this?
2. What is the reference surface for Laser aiming?
3. The four mounting holes in the bracket seem to be sized for clearance of a M6x1 machine screw and 'preferred' tolerance. Be aware that the mounting screws have both a diameter tolerance and a builtin clearance allowance. Thus both the exact position and angular rotation of the bracket on its mounting surface is uncontrolled.

These are things that must be either controlled or compensated for at assembly or calibration time. There may be more.

SIDE STORY:
I once worked at a company that made Laser levels for the construction industry. They were used for leveling pipelines and such with worstcase accuracy better than 1/8 inch over 100 feet (3mm over 30m). The owner designed a new, lower cost, model with the same advertised accuracy. Well, he forgot one degree of freedom in the internal Laser alignment, and the manufacturing folks were spending an awful lot of time getting the darned things calibrated. We had to drag him out on the manufacturing floor and get him to try calibrating a few before he realized that a "Version 2" was kinda needed!
END STORY:

Cheers,
Tom

 

[Doc]

Here are some things to consider in mounting the Laser:

1. Keep in mind that you do not have control over the how the Laser diode is aligned within its package. How will you handle this?
2. What is the reference surface for Laser aiming?
3. The four mounting holes in the bracket seem to be sized for clearance of a M6x1 machine screw and 'preferred' tolerance. Be aware that the mounting screws have both a diameter tolerance and a builtin clearance allowance. Thus both the exact position and angular rotation of the bracket on its mounting surface is uncontrolled.

1. Without being able to adjust the pointing of the laser bracket, I would initially quantify the pointing error of the laser with respect to the package: this error would then need to be accounted for in the positional requirement which would of course flow down to the manufacturing tolerances. However, in real life I would just mount the diode on a two-axis adjustable mount.

2. I was just going to use a hypothetical bulls-eye.

3. Yes correct on the M6x1 clearance holes. With respect to controlling angle and position, I would likely use three dowel pins to act as reference surfaces.

Thanks,
Doc

 

[Doc]

Can anybody comment on the apparent contradiction of my tolerances?