Starting on GD&T & Tolerances 5

[Algirdas13]

Hey everyone,

Recently I decided to learn and try to apply the GD&T & Tolerances in my designed products. For this reason I want to ask you guys, how did you started to learn those things? Perhaps you could share some insights from your experience where to start, what is important, in what way to think while learning, or even share some useful material from which would be useful to learn?

Thanks in advance!

 

[ctopher]

For me most of it is experience at different companies.
My current company paid for in-house 3 day training for engineering and inspection.
It's nice to have a refresher training.
It would be beneficial for your company to have training. To learn it on your own is good, but you can easily misinterpret it wrong.

Chris, CSWP
SolidWorks
ctophers home

 

[3DDave]

https://www.youtube.com/watch?v=Ui1HS0sHkos

https://www.youtube.com/watch?v=SdGSeD-KPgc&t=0s

 

[Burunduk]

First step should be getting to know which standard(s) on dimensioning and tolerancing you will be basing your product definitions on and getting access to it/them, whether it's ASME Y14.5 or ISO 1101 and its complementary standards from the ISO "GPS" set. Once you know which it is and get access to it, start familiarizing yourself with it. Standards may be too "dry" for some people though, so you can try some books/courses afterwards, but those too must be based on the standard(s) you will work to.

 

[Algirdas13]

Hey guys,

Thank you for sharing your experiences and insights! In this particular case I'm interested into ISO 1101.

 

[ctopher]

I don't know your experience. It also helps to understand machining practices.

Chris, CSWP
SolidWorks
ctophers home

 

[geesaman.d]

There are three things that matter to learn how to properly apply GD&T to your job:
1) How parts are held / mounted in operation, fabrication, and inspection - and how to specify datums that agree with these practices
2) What tolerances are critical to proper function of the part. Are these functional tolerances (preferred) or tolerances that come from the capabilities of the manufacturing process (not preferred).
3) How to convey 1) into GD&T datums and 2) into feature control frames.

Make sure you bring 1&2 with you to any normal GD&T training.

I've done a few GD&T trainings, and the biggest problem is spending time training GD&T methods that don't fit your business. It's good to know all of what's out there but most of us need GD&T to be working for us right now. My company makes small quantities of a wide variety of components, so functional gauging is almost never practical. Likewise, our component vendors mostly don't have CMM and we only buy parts in small batches, so highly clever tolerances such as profile of a surface are not available. So it helps immensely if you have GD&T training that focuses on the GD&T elements that are actually available to you.

I've never run a GD&T training class, but I have trained 3D surface modeling for a client, on-site. We couldn't get every employee for the full day of the three days, but that led us to a possibly better format - formal training in the morning and free study in the afternoon. The employees who had the time, also spent the afternoon applying the new skills to actual work projects and I spent my time supporting them and trying to resolve when they got stuck. We all learned a lot and got a lot done in the afternoons and I think the same format could be perfect GD&T training. Let the students spend a portion of the time re-tolerancing common parts and discuss amongst each other (designers, machinists, inspectors) how various GD&T methods can improve a part.

 

[Burunduk]

 1) How parts are held / mounted in operation, fabrication, and inspection - and how to specify datums that agree with these practices  

One doesn't have to know how parts are held in fabrication in order to select datum features. It's the other way around: those planning the fabrication process need to know which features are specified as the datum features on the drawing for deciding how to hold the part in fabrication.

highly clever tolerances such as profile of a surface are not available.

Then what replaces them?

 

[geesaman.d]

When you work in an industry where small lot sizes are the reality and lead times and costs need to be contained, knowing the readily available fabrication and machining practices will save you money. You can design your parts around available, affordable fabrication methods and your drawing tolerances align with them, you'll get the best prices and deliveries.

Of course as the designer you have the right to ignore manufacturing requirements. With high volume production it's insignificant. With low volume production, it's bad design.

What replaces profile of a surface? Well if it's not a doubly-curved surface, regular dimensions and tolerances. If it's an odd shape, such as a casting, you can use standard casting tolerances with a few overrides where it's necessary. If it's an odd shape that needs to be accurate (say, the driving face of a propeller blade) there are standard practices that work well and align better with the functional requirements of that shape than any geometric tolerance method.

GD&T is a solution, not a religion. It's often the best solution, but not as a rule.

 

[3DDave]

"GD&T is a solution, not a religion. "

It can be a religion for those who get defensive enough about it.

 

[Burunduk]

Geesaman.d,
There should be no illusion - a designer cannot always make everyone happy. The sooner this realization arrives, the better. In the ideal case when the needs of the customer, manufacturing, and inspection overlap, there is no issue and not much to think about. But what if certain features of the part are the interface for the assembly used by the customer, and other features are where the manufacturer wants to hold the part in his production process and measure from? Whose needs would you prioritize? Is it the customer who pays you or the vendor who gets paid by your company? That's when you need to make meaningful decisions and have a clear policy. Those that don't have a clear policy will always benefit the vendor, often at the expense of the customer. The part needs to be produced before it is supplied, so the vendor or manufacturing is always the first station that might rebuff your decisions. For sure, making convenient decisions for the vendor or manufacturer may reduce some costs and give you peace of mind. When it's done with the consideration that the design is for small, one-off batches, chances are that it will remain that way and the part will never be mass-produced. The customers will try the product and then move on to competitors that primarily integrate functional needs into the product definitions and can provide better-performing parts.

As for profile of a surface, it is not a "highly clever tolerance," and it's the only meaningful way to control the location of a feature without size from a datum reference. Often a surface plate, a height gage and possibly some gage blocks are all that is required for its verification. Another common device that might be relevant is an optical comparator. I doubt your vendors don't have that equipment.

 

[3DDave]

"a designer cannot always make everyone happy."

Preach.

 

[geesaman.d]

"GD&T is a solution, not a religion. "

It can be a religion for those who get defensive enough about it.

Indeed, I shall not explore that further.

To the OP, beware of GD&T trainers who lack the breadth of experience to consider the nuance of your specific situation. (If I had a dollar for every corporate trainer who is advised "yeah this topic is well outside of our situation; please skip over it", but does not skip anything and manages to fill the allotted training time with only boilerplate content, I'd be retired)

Our business is to deliver in-spec product, on-time, and under budget. You don't get to turn over your existing supply chain because you want to use better symbols on your drawings (I know that's not fair, but to anyone not making drawings, it's the first thought as soon as difficulty arises). GD&T definitely helps you make better product more cheaply. And poor GD&T makes it impossible. It's nice to have the winning academic argument in support of your drawing, but drawings are communication tools to get what the customer needs and we can't be so in love with our drawings to lose accountability to ensure the messages are being received.

https://www.youtube.com/watch?v=hEcuVR2vHwM

Customers don't make more money if you use GD&T eloquently and your supply chain fails to follow it. So get to know your supply chain and ensure the new messages are being understood and applied. When you add a new GD&T symbol, call the main vendors, show them the print, and ensure they understand its meaning and that it can be applied as you had hoped. Before long, your supply chain will become literate from on-the-job GD&T training.

I trust you will have the common sense to know how to tie together both your customer's needs and vendor's capabilities without giving away the business.