perfect parts via. solid models....

I cut chips for a living, well I do more than cut the chips but hey that's small business for ya...anyway one of the things I do everyday is solid model parts from a customers drawings for CAM purposes. That is nice work....Solidworks? Pro/E? or some high-end stuff? Is the rifling the correct twist rate?

I agree that +/- 0. is pure fiction, it has never been done, not by the highest speed, lowest drag, super secretest weapons grade unobtainium process in the world. At some level even different measurement methods and instruments have a tolerance.

It is precisely these engineering aids (CAD) that are leading to the detachment of engineers from the reality of actually making the parts. Honestly, in many cases and in firearms specifically super tight tolerances are a double edged sword. Yes, some areas require them but many areas do not and in fact tight tolerances can hinder reliability.

I can't tell you how many times I have received drawings to quote from where everything had a +.0005 -0. tolerance. I quote that job and my price blows the customers head off. Come to find out the engineer hit "auto dimension" and never checked the tolerance settings. The newer CAD/CAM software is amazing, but speaking as a guy that deals with both sides, I think everybody that pushes lines should spend time using those drawings to cut chips.

Ok....rant over...

Nice work!

It sounds like you run a successful shop, which is not as easy as it sounds.

This is kind of nerdy of me but I still get a kick out of thinking up a design, modeling it, and then machining it all in one day. It's just neat to start the day with an idea, and hold it in your hand by the end of the day.

I've never done any design/machining work for the firearms industry, but we quoted some work from a major manufacturer recently. It's always more fun to make widgets when you know where they go.

Solidworks is a steep learning curve especially if you don't do it full time. Don't be embarrassed about how long it took, if they are correct that's all that matters, they are invaluable to modern manufacturing. Faster is always better but as they say around here......you can never miss fast enough.

...in fact tight tolerances can hinder reliability

Click to expand...

I thought tolerances had more to do with parts interchangeability than how well they fit together...
At least, that's what I gathered from rooming with engineering students for 3 years.

Tight tolerances can only aid in reliablity unless the design itself is flawed.

Good solid models. I have been using Solidworks for 9 yrs now. I like the engineering functions where you can test your assemblies by applying force/stress and make sure your tolerancing and clearances are acceptable. We designed a sound system speaker rack to support suspended speakes and we could change the structural material to determine if it had adequete strength. It will even bend the assemble if it can't support it.

I mainly work on nuclear sub components for Northrop Grumman and EB. Alot of sub-safe hatches and steering gear. We don't design anything for them unless it is jigs and fixturing to complete machining/welding processes.

I have wanted to take the time to draw a gun complete but haven't invested the time in it. Most of the solid model files you find onling are not assemblies but one block that has been extruded or cut to look like a popular firearm. A CMM machine would be ideal to get part shapes but we don't have one here. CMM's can usually output in .SAT, IGES etcc and then you just import into Solidworks to complete.

Parrot Cannon we machined




Submarine Hatch

Wow, I used to work for a pump company that made pumps for subs. Seemed like a giant PITA with all the documentation and testing. I'm sure it was required but it was very involved. Then again being 1200 feet underwater in a steel tube with tons of explosives at one end and a nuclear reactor at the other just doesn't seem right. I guess that kind of thing makes you cross your t's.

I used to be a safety engineer, and growing up my best friends father was a former nuc engineer. I heard lots of stories about ADM. Rickover and his obsession with sub safety.

As far as the tolerances go, and to make this firearms related......think of the two primary assault rifles of our time. The AK-47 and the M-16. The AK is renowned for running reliably....not being very accurate but never stopping. The M-16 is renowned for being very accurate but requiring regular maintenance.(cleaning) The M-16's early history was plagued with failure because the troops didn't understand or weren't told that it needed to be relatively clean to run. The difference? Tolerances.

Machining things to perfection is fine and dandy but it still is up to the designer/engineer to strike the right balance and use the tolerances correctly.

Thru the decades I witnessed the change over from drafters to CAD operators. My thoughts are that drafters thought about their work and understood the basics of design better than CAD operators. As CAD took over and the operators became more experienced things became better especially with the solid modeling programs. Experience is the best teacher. Tolerances are what they are it’s a mixture of art / science based on experience with a understanding of cost and functionality.

My guess is that once the missile keys are turned in control there is just some blinky red button that actually launches, so someone needs to make sure that the "button" doesn't get pushed unless it's supposed to......

yeah that was my first thought but honestly that seemed too......hollywood....? I guess. Yes, very sobering. I wonder if other strategic assets have similar protocol.

Every so often I get into disagreements at work over the topic of tolerances, in which I make the point that +/- 0.0 is a concept that only exists between your ears and is not achievable on planet earth....

...So I guess I've realized one add'l place where +/- 0.0 can exist. In cyber space.... which is kind of like an extension of your mind in some ways.....

Click to expand...

I teach chemistry on the high school and college level, and one of the first things I tell my students is that no "real world" measurement is perfect. Ever. Perfect measurements only exist in mathematical models, which your solid models are, of course, an example of. So you are 100% correct, and your buddies at work are wrong. All "real-life" measurements carry a certain amount of uncertainty, which is why many scientists (and I would imagine engineers as well) use significant figures. To convey the degree of certainty present in a measurement.

Jason

It's easy to confuse tolerance on individual parts, and design clearances between two parts in an assembly.

The reality is that you can't put a 1" peg in a 1" hole. You need some clearance.

Too little clearance and a even a very fine particle could bind up a mechanism.

But since you are going to have a Tolerance on each part, you're design clearance also has to accommodate the largest in spec peg and the smallest in spec hole.

So clearances and tolerances are not the same thing, though they are related.

Holding to a tight tolerance can allow you to decrease the design clearance, however, holding a tight tolerance can require special efforts and extra processes, which cost money. Putting a tighter tolerance on parts can quickly make a project not cost effective.

The trick is to understand the mechanism well enough to know where you really need a tight tolerance, and where you don't.

My experience is that EVERY design decision is ALWAYS a compromise between several considerations.

At least that's how I understand it. I'm always open to learn something.

Click to expand...

Meh... that's a load of hooey. Just machine it tight & let the end user fight through 500 rounds of breakin till it runs right.

SSN Vet said:

It's easy to confuse tolerance on individual parts, and design clearances between two parts in an assembly.

The reality is that you can't put a 1" peg in a 1" hole. You need some clearance.

Too little clearance and a even a very fine particle could bind up a mechanism.

But since you are going to have a Tolerance on each part, you're design clearance also has to accommodate the largest in spec peg and the smallest in spec hole.

So clearances and tolerances are not the same thing, though they are related.

Holding to a tight tolerance can allow you to decrease the design clearance, however, holding a tight tolerance can require special efforts and extra processes, which cost money. Putting a tighter tolerance on parts can quickly make a project not cost effective.

The trick is to understand the mechanism well enough to know where you really need a tight tolerance, and where you don't.

My experience is that EVERY design decision is ALWAYS a compromise between several considerations.

At least that's how I understand it. I'm always open to learn something.

Click to expand...

This is where Geometrical tolerancing comes into play.

http://en.wikipedia.org/wiki/Geometric_dimensioning_and_tolerancing

http://www.qualitycoach.net/product...-tolerancing-pocket-guide-asme-y14-5-1994.asp

There are two main classes of tolerances, dimensional and geometric. Dimensional tolerances are addressed in “Recommended Practices for Dimensions and Dimensional Tolerances” written by Markus Hauser, Mike Strub and Tom Hendrix, dated April 18, 2000.

Geometric tolerances are the more complex of these two types. Geometric tolerances provide more flexible means for controlling shape than do dimensional tolerances. They achieve this by enabling tolerances to be defined independently of explicit dimensions. This enables tolerances to be specified that are more closely related to the functional requirements of the design, such as strength and fit. These tolerances are the subject of this recommended practices guide.

Once you understand it GD&T it makes designing and machining of parts easier to achieve the balance you are looking for. You can usually see the relationship that one parts has to the assembly and it takes that into account. It is very much different than straight dimensioning a part although that still has a place in machining.

I was a CNC Machinist then moved into CNC Programming in the mid 80's and did Dies and Molds for many years before moving into Project Managment. I have programmed lasers, mills, lathes, plasma, punch, and mills with 4th axis's. My main speciality evolved into 3-D machining making turbine components for GE & Westinghouse. The tools have evolved over the years as well as the software and it has made prototyping and designing better as well as faster. You still need skilled craftsman to ensure things are done correctly though. Too many places have button pushers and very little skilled labor is being trained for the marketplace right now- but that is a different rant and off topic.

So I guess I've realized one add'l place where +/- 0.0 can exist. In cyber space....

Click to expand...

Even there it's a problem. Due to the way computers store decimals (floating point numbers) any numerical value you input in our 10-based fractional system becomes stored in the computer as the closest approximation it can get in a 2-based fractional system. The same thing is done for whole numbers (ie - base-10 to binary), but with whole numbers any number you can represent in one you can represent in another. The same isn't true for fractional numbers.

You can match SOME numbers exactly, and you can get most plenty close enough for anyone that would care, but if you take the number out enough decimal places it will differ just a bit.

When I was majoring in Mechanical Engineering 'way back when dinosaurs still roamed the earth I had a professor who enjoyed telling a joke to illustrate the difference between theory and real world. He would say, "If you take a pretty girl and put her in that corner of this room (pointing to the far back corner) and put an Engineering major and a Mathematics major in the doorway, then tell them both to move half the distance to the girl, then half of the remaining distance, then half of THAT distance and so forth the Mathematics major will tell you that you'll never reach the girl. The Engineering major will say, "Yes, but you can get close enough for all practical purposes..." "