Recommendation? CAD GD&T for knife makers

REK Knives

REK Knives

Moderator
Knifemaker / Craftsman / Service Provider
Joined
Apr 3, 2011
Messages
5,791
As with most things we don't need to know every aspect of gd&t (geometric dimensioning and tolerancing) ... So what portions should we be concerned with and learn? I'm especially wanting to know how to specify surface finishes (form). Any recommendations on videos or training for this?
 
Suggest you pickup

Engineering Drawing and Design​

By David Madsen

It’s a book I use 30 years ago when I was first a Mechanical AutoCAD draftsman

Any older edition will be very useful to you
 
HSC /// said:
Suggest you pickup

Engineering Drawing and Design​

By David Madsen

It’s a book I use 30 years ago when I was first a Mechanical AutoCAD draftsman

Any older edition will be very useful to you
Click to expand...
Thanks will do. Is there a way any specific part of the book that would apply more to knife makers?
 
REK Knives said:
Thanks will do. Is there a way any specific part of the book that would apply more to knife makers?
Click to expand...
Well you mentioned surface finish... so that information exists in the book

You need a general understanding first and then you can identify what’s important to you.

Specifying surface finish generally means something is machined or ground and you are communicating to the fabricator what you are expecting on a finished part
 
As far as surface finish goes I see mostly everything in industry called out in Ra micro inches. 63 Ra seems like a pretty common go-to finish. I’ve milled some parts to 16Ra or better and that can be a little tricky to maintain without any kind of post polishing process. I think most knife finishes would be 16Ra or better. The smaller the number the smoother the finish.
 
my experience is with RA and RMS, where its a measurement of peaks and valleys on a machined surface. I think it would be more practical to provide a sample of the surface finish you want and forget about the technical jargon.
 
REK Knives said:
As with most things we don't need to know every aspect of gd&t (geometric dimensioning and tolerancing) ... So what portions should we be concerned with and learn? I'm especially wanting to know how to specify surface finishes (form). Any recommendations on videos or training for this?
Click to expand...

CNC Cookbook has a really good page on Surface Finish. Form and finish probably should not be looked at as the same thing. You can have a dead flat surface (form) with a 60grit finish, just as you can have a wavy surface with a 1200grit finish.

Just a thought, if you want to know what finish you want, get some Norax or Trizact belts where you can determine the grit in microns, run some flat steel lengthways past a contact wheel and you will have a good idea of what a surface ground finish of that micron level will look like.

Seems to me GD&T is only important if you are making batched mid-techs where you will be assembling scales to tangs, or in folder batches were you want parts to screw together and have their profiles match up. In these cases you want to understand Datums, Position, Flatness, Parallelism, Profile (maybe Surface Profile too) and maybe Perpendicularity. If you are having folder parts made you could add nearly all the others. Then you use Theoretically Exact Dimensions (boxed dimensions) in conjunction with the geometric tolerances, rather than using +/- tolerances on dimensions.

I second P.Brewster P.Brewster 's recommendation of gdandtbasics.com They are clear and thorough. They have a lot of videos on Youtube too. There are LOTS of other on-line resources now.

I am curious what you want to have made that will need GD&T?

Chris
 
C_Claycomb said:
CNC Cookbook has a really good page on Surface Finish. Form and finish probably should not be looked at as the same thing. You can have a dead flat surface (form) with a 60grit finish, just as you can have a wavy surface with a 1200grit finish.

Just a thought, if you want to know what finish you want, get some Norax or Trizact belts where you can determine the grit in microns, run some flat steel lengthways past a contact wheel and you will have a good idea of what a surface ground finish of that micron level will look like.

Seems to me GD&T is only important if you are making batched mid-techs where you will be assembling scales to tangs, or in folder batches were you want parts to screw together and have their profiles match up. In these cases you want to understand Datums, Position, Flatness, Parallelism, Profile (maybe Surface Profile too) and maybe Perpendicularity. If you are having folder parts made you could add nearly all the others. Then you use Theoretically Exact Dimensions (boxed dimensions) in conjunction with the geometric tolerances, rather than using +/- tolerances on dimensions.

I second P.Brewster P.Brewster 's recommendation of gdandtbasics.com They are clear and thorough. They have a lot of videos on Youtube too. There are LOTS of other on-line resources now.

I am curious what you want to have made that will need GD&T?

Chris
Click to expand...

Well I'm mainly wanting to be able to specify how the finish comes out when it comes to milling... For example here are some parts that I had milled and they are pretty terrible:

i-ScPtRNR-X2.jpg
i-bbDmwWL-X2.jpg
i-C2dRvCr-X2.jpg


I want to be able to specify that I want a nice smooth finish that doesn't have stutter marks (like around the rim) and that the stepover isn't too large to create bumps. Basically, I want a nice smooth finish that is ready for blasting or tumbling for the final finish.
 
REK Knives said:
Well I'm mainly wanting to be able to specify how the finish comes out when it comes to milling... For example here are some parts that I had milled and they are pretty terrible:

i-ScPtRNR-X2.jpg
i-bbDmwWL-X2.jpg
i-C2dRvCr-X2.jpg


I want to be able to specify that I want a nice smooth finish that doesn't have stutter marks (like around the rim) and that the stepover isn't too large to create bumps. Basically, I want a nice smooth finish that is ready for blasting or tumbling for the final finish.
Click to expand...
Why not just speak directly to your machine shop? Assuming their Cad/cam programmer is in house, they can change the step over amounts and they’ll probably give you a new price because there will be a lot more time
 
REK Knives said:
Well I'm mainly wanting to be able to specify how the finish comes out when it comes to milling... For example here are some parts that I had milled and they are pretty terrible:

I want to be able to specify that I want a nice smooth finish that doesn't have stutter marks (like around the rim) and that the stepover isn't too large to create bumps. Basically, I want a nice smooth finish that is ready for blasting or tumbling for the final finish.
Click to expand...
OK I had a feeling this is where this was going. I say this from the experience of having dozens of outsourced knife jobs go south - specifying requirements that you can't measure yourself is a wasted effort. If push comes to shove, you'll need to demonstrate that the parts are out of spec, and your demonstration needs to be bulletproof if you want the shop to accept the rejection and redo the work. The best bet is to find a good vendor by way of conversation about your needs, unless (a) you're willing to buy a CMM machine, or (b) you're paying your shop to do a full "first article inspection".
 
3D sculpted clips are slow and time consuming with potential work holding challenges. Not cheap work.
 
P.Brewster said:
OK I had a feeling this is where this was going. I say this from the experience of having dozens of outsourced knife jobs go south - specifying requirements that you can't measure yourself is a wasted effort. If push comes to shove, you'll need to demonstrate that the parts are out of spec, and your demonstration needs to be bulletproof if you want the shop to accept the rejection and redo the work. The best bet is to find a good vendor by way of conversation about your needs, unless (a) you're willing to buy a CMM machine, or (b) you're paying your shop to do a full "first article inspection".
Click to expand...
Yep I totally agree. I was just wondering if there is a way to specify this on the CAD drawing. I do have a shop that has the equipment to verify for me if needed.
 
I can see no great purpose to using GD&T on a pocket clip. Specifying finish is something else.

I imagine you must be able to get something like the SPI Composite Pocket Set No. 30-695-1 in the US. Also called a roughness comparator plate. Handy for seeing what finishes look like.

I agree with Contender Machine Contender Machine on the inch Ra numbers. I think 63( 1.6 micron Ra / N7) would be pretty good, certainly for the sides of the clip, but maybe 32 (0.8micron Ra / N6) for the outer face of the clip if what you show isn't enough. That said, for the extra time involved in improving the finish of the outside of the clip, I might be tempted to have a concave form made to which I would attach abrasive paper and give it a quick hand polish. I think it would be done faster than asking a mill to half the step over.
This might help with how to put the info on a drawing.

The Basics of Surface Finish | GD&T Basics

Surface finish can be a confusing subject, especially when it comes to units or the difference between Ra and Rz. In this article, we will provide a broad overview and answer some of the most commonly asked questions... What is Surface Finish? Surface finish is the term used to describe the...
www.gdandtbasics.com www.gdandtbasics.com
www.cnccookbook.com

Surface Finish Chart, Symbols & Roughness Conversion Tables

Complete Guide to Surface Finish Symbols, Roughness Charts, RA, RZ, Measurements, and Callouts. From basics to advanced, symbols to inspection, from CNCCookbook, the leading CNC Blog on the Internet.
www.cnccookbook.com www.cnccookbook.com

It might also be worth having a conversation with the machinist because some of what is shown looks like they attempted to have finer step over where it was going to show, and less fine where it wasn't (so much), and that the profiling around the top near the holes suffered from chatter. Maybe different work holding would help. I wonder if they are using a tab to screw the part down and that rough area is done in a different operation to the rest of the machining, just to remove their work holding point? If a part is machined in different operations, it can help to design in a surface break rather than expect the machinist to perfectly blend two surfaces on the mill. Example would be a typical shield emblem where the concave upper edge could be done in a different op to the convex sides and the corners would just need de-burring. Continuous tangent radii cut with a ball nose mill really all need to be done in one operation, and ball mills generate relatively high loads on the part, so if there is scope for it to move, that is what is likely to do it.

Chris
 
Hello. I have worked at a 5-axis CNC shop for the past 10 years doing QC, QA, and now process management. As others have said, smoother surface finish means more $$$’s. Also as others have said, if you’re going to try to get these clips remade, you’ll need a strong argument that something is out of spec. In my opinion, your best bet is a smaller mom and pop shop that you can work closely with. The shop I work at isn’t very big (15-20 employees), but we have things like ISO9001 and AS9100 certification, as well as ITAR registration. All of that kind of stuff brings up costs too.
 
Oh, and a surface roughness sample comparison is gonna be easier than learning GD&T for surface roughness.
 
dl351 said:
Hello. I have worked at a 5-axis CNC shop for the past 10 years doing QC, QA, and now process management. As others have said, smoother surface finish means more $$$’s. Also as others have said, if you’re going to try to get these clips remade, you’ll need a strong argument that something is out of spec. In my opinion, your best bet is a smaller mom and pop shop that you can work closely with. The shop I work at isn’t very big (15-20 employees), but we have things like ISO9001 and AS9100 certification, as well as ITAR registration. All of that kind of stuff brings up costs too.
Click to expand...
I completely agree here about trying to work directly with the shop. If they are willing and motivated enough they should be able to figure out what you desire for final quality. But as stated the smoother the finish the longer it takes so therefore the more it costs. Those clips are definitely a tricky part to machine but done right they should come out of the machine requiring very little hands on polishing. The shop may also have very minor suggestions for design changes that can make a world of difference in the time it takes to machine them. 3D surfacing can be challenging at times but it can produce amazing results both visually and dimensionally.
 
You must log in or register to reply here.
Back
Top