6-axis Horizontal Machining Center for Education

[ishi]

In the real World, everything is built down to a price...…..if you want to get more profit from the build you go offshore to the cheapest supplier and cross your fingers for the quality and longevity of the cheaper inventory you are now offering.....ask me how I know that to be true..

Usually, and more often than not, anyone who wants better performance strips out the cheap stuff and loads it up with top of the class goodies...….custom cars are in this category.

You would not find anyone buying a from the ground up custom car for a performance agenda unless it was a strictly race track performer and made to order.

As this endeavour is to specifically train people in the 6 axis mode, quality of the output is not a real issue as the end products are not a production item that had to be assembled to give you a marketable product...…..that is, the test pieces get binned when the class goes home or used as trophies.

I would think that once you had mastered the HMC 6 axis mode of working...….(I like this description of this type of machine better than the omnio thing)…. doing it for real on a better high class machine would be something you got paid for and did in another environment that had high class machinery for their production, something that in a class room is not really needed as part of the deal.

This would dramatically cut the cost of the outlay down …..these training machines could/would experience many crashes and so would not have a very high resale value.

I suppose if you wanted to sell such a machine on you could patch up an EG built machine if it showed wear and tear, so high class components are not really a benefit for the initial purchaser.....a 5 year life span would be a good expectation as technology races along.
Ian.

Ian,

I tend to agree with you, but I would like to point out a few caveats.

In my opinion, a training machine should have the same controller and the same precision that a production machine has. What a training machine does not need is a very large envelope, very high rapids, very high feeds, and chip evacuation. To understand this, let me make an analogy: when I learned flying, I did so on a Cessna 172SP. Compared to an airliner, it has a small capacity (4 passengers), it climbs slow, it flies slow, and unloading luggage isn't particularly easy. In this analogy, your envelope is my passenger count, your rapids are my climb speed, your feeds are my cruising speed, and your chip evacuation is my luggage unloading.

But this little Cessna had two things that very important to me as a student pilot: it had the exact same avionics as a jet that is 10 times more expensive (Garmin G1000 at the time), and it had a GPS that is every bit as precise. In fact, in many instances, student planes are loaded with modern avionics (including autopilots) that are much more advanced and sophisticated than what you will find on many airliners still flying today in the Western hemisphere.

Why is that important? Regarding the instrument panel, because you want student pilots to get as many hours in front of this complex system as they possibly can, and the sooner they start, the better. Familiarity with an instrument panel might save your life one day. Regarding the precision of GPS positioning, I think we can all agree that having students landing on the wrong runway because their GPS location is off by a few hundred feet would not be such a great idea.

Well, the same translates to controller and precision. I am an ardent advocate for the fact that a training machine should have the exact same controller that a production machine is using. From a marginal cost of production standpoint, controllers are cheap (they're just PCs with a fancy enclosure) and software is free. Therefore, there is positively no reason not to use the best that modern technology has to offer.

As far as precision is concerned, I respectfully reject the idea that student parts should not be made to the same tolerances as production parts. In fact, during my initial training as a machinist, we cut a few parts without caring much about tolerances, but it's only when we did our last project for which tolerances were important that everything came together. And when we did that project, I made sure to buy my own carbide endmills and lathe cutting tools with carbide inserts in order to produce as good a surface finish as I could get, on machines that were pretty accurate to start with, but were spoiled by poor cutting tools. And as you know, you cannot have an accurate piece if you do not have an accurate machine. You might be able to compensate for a bit of backlash here and there, but this takes a lot of practice, and I do not think that this is the kind of thing that we should teach future machinists, because this is not the kind of skill that they will be hired for in the workplace. Instead, their mastery of 5-axis machining (and 6-axis in the future hopefully) is what will give them job security.

Envelope, rapids, and feeds can be scaled down, but controller and accuracy should not.

Now, once again, I am convinced that we can have comparable accuracy with parts that are on average 30% cheaper than the ones we are planning to use for the initial build. But there are many good reasons to go for top-of-the-line parts initially, most of them I have already outlined in prior posts.

One last comment: you keep referring to offshore or "Chinese" as a source of low cost and low quality parts and machines. I tend to view this opinion as fairly shortsighted, and I would like to invite you to reconsider it. For that, all you have to do is look at history. After World War II, Japan started churning out parts and tools with low cost but low quality. Eventually though, they managed to increase quality tremendously, while keeping prices low. Today, I don't think you view Japan as a source of low cost/low quality goods. In fact, as far as CNC machines are concerned, I view Japan as being the country setting the benchmark alongside Germany.

Interestingly, the same keeps happening throughout history. Today, we look at Switzerland as the place that makes chocolate and fine timekeeping pieces. But a few hundred years ago, the best pocket watches would come from Britain. One day, Swiss farmers started producing low cost/low quality knockoffs, and they took over. This evolutionary process keeps repeating itself over and over again, all the time.

Well, if that is true, we should expect the same to happen for Chinese CNC machines. Sooner than we all think, most of the critical innovation in the space will come from China, and their machines will match or exceed what you can buy from Germany or Japan. In fact, it is quite likely that many of the top-of-the-line manufacturers that we know today will have gone out of business or will have become irrelevant by that time.

What does that tell us? Well, first, you cannot fight against the tide of history, there really is no point trying. Second, instead of trying to make it a zero-sum game, try to understand how we can work together. In that respect, the partnership between DMG and MORI is inspiring. Good ideas can come from any place. There is talent everywhere, we just have to keep open eyes and an open mind to find it.

Cheers!

Ismael

[ishi]

Now that I have a broad outline of the CAD design, I am doing it all over again from scratch, but this time in a fully componentized and parametric fashion. I could have done it that way from the get go, but my CAD skills still need quite a bit of work, and it's not always easy to properly constrain a CAD design with the right parameters upfront. It's only when you've gone through the whole design once (or a few times) that the core constraints and parameters start to emerge (for me at least). In fact, I really like working in three steps:

1. Pencil on paper
2. Quick and dirty CAD
3. Proper CAD

In doing this new design, I will also make sure to properly group components in such a way that kinematics can be simulated with the CAD tool (Fusion 360).

[RCaffin]

Rule of thumb from my research days: we don't bother with a field test until version 6. Same story really.

Cheers
Roger

[ishi]

Rule of thumb from my research days: we don't bother with a field test until version 6. Same story really.

Cheers
Roger

I'd be ecstatic if I could go to the field as early as version 6!

(^_^)

[ishi]

Another long day of modeling, trying to understand bodies, components, and joints in Fusion 360. Things are starting to make sense, and I spent most of the afternoon playing with the X-axis carriage, bringing its weight from over 250kg down to a more manageable 160kg. I also made sure to include ground pads for the roller blocks underneath. And I managed to reduce the height of the carriage by 25mm by adding some mounting pads for the motor brackets. I'm actually quite happy with the whole thing, and I'm pretty sure that we could bring weight below 125kg by relieving some material with round holes here and there. But I'll let this iteration sit for a while before making any further changes to it.

In order to understand the overall design, just keep in mind that this box is tall as it is because if needs to provide clearance for the bottom of the rotary table's torque mounted that will ride "within" it, being actually mounted onto the Z-Axis carriage. As a result, I can't really add spars and ribs within the box to make it more rigid. I could add some underneath though...

[RCaffin]

I suspect we had spent over $0.5M by that stage (V6).
But it did work and did solve the problem. Great joy (or was it relief?) all around.

Cheers

[ishi]

I suspect we had spent over $0.5M by that stage (V6).
But it did work and did solve the problem. Great joy (or was it relief?) all around.

Cheers

No kidding...

[RCaffin]

No kidding.
It was solving a T&M problem in a $4B market. The cost was considered minor.

Cheers

[ishi]

No kidding.
It was solving a T&M problem in a $4B market. The cost was considered minor.

Cheers

Well, in that case it sounds like a solid investment.

[handlewanker]

Hi, when I was at school we had a maths master who delighted in creating a large equation on the board and then inviting us to simplify it to it's lowest denomination...…..usually something like a+b/c=0......whatever.

If you hark back to the beginning of your proposed design it now is almost at the end of it's simplification.

The final offering will be an email to the customer to down load a list of parts and a supplier(s) who will ship directly to wherever you are......the World has come to your door but it's an answering machine that takes the orders......where does the $150K value come into this plan.

I, as I have said before, am an inventor per se, that is, I have dreamed up designs and actually made parts that sold for real money......my latest design is for a unique type of cylinder head for a 4 cycle engine and the asking price is 50 million dollars for the patent......ludicrous?....when it comes to ideas nothing is ludicrous...…...anyone can make a Rolls Royce but it was the sheer genius of Alec Issigonis that created the conceptual design for the Mini and put the UK on the car makers' map at a time when industry was going down.

Of this I am certain, the concept of a DIY kit machine for a teaching environment is not a plan that will attract anyone to invest large amounts of money in.

You are talking about off the shelf items that must integrate precisely, assembled by the facility who will use it...…….unless the facility has a trained work force that know how to assemble such a collection of parts the end result will be less than satisfactory......my opinion.

I cannot see a teacher and a group of wannabe pupils suddenly acquiring skills that are diverse from the teaching person's training background, unless A goes with B and the whole lot just slots together with precise alignment like an Ikea furniture kit...….and Ikea do source many if not all of their products from China now.

I'll sum up and say that unless you actually source your components from the suppliers yourself without any rework etc, this project has no hope of flying with fare paying passengers this time this century.

Products from suppliers, that are catalogued, may look and have the same specs, but as they too rationalise their production and sourcing, you would not be aware of a simple design change that could affect your build kits at the assembly end.

Unless you actually supply the complete kits yourself I think you will fall down big time on this concept.
Ian.,

[ishi]

Hi, when I was at school we had a maths master who delighted in creating a large equation on the board and then inviting us to simplify it to it's lowest denomination...…..usually something like a+b/c=0......whatever.

If you hark back to the beginning of your proposed design it now is almost at the end of it's simplification.

The final offering will be an email to the customer to down load a list of parts and a supplier(s) who will ship directly to wherever you are......the World has come to your door but it's an answering machine that takes the orders......where does the $150K value come into this plan.

I, as I have said before, am an inventor per se, that is, I have dreamed up designs and actually made parts that sold for real money......my latest design is for a unique type of cylinder head for a 4 cycle engine and the asking price is 50 million dollars for the patent......ludicrous?....when it comes to ideas nothing is ludicrous...…...anyone can make a Rolls Royce but it was the sheer genius of Alec Issigonis that created the conceptual design for the Mini and put the UK on the car makers' map at a time when industry was going down.

Of this I am certain, the concept of a DIY kit machine for a teaching environment is not a plan that will attract anyone to invest large amounts of money in.

You are talking about off the shelf items that must integrate precisely, assembled by the facility who will use it...…….unless the facility has a trained work force that know how to assemble such a collection of parts the end result will be less than satisfactory......my opinion.

I cannot see a teacher and a group of wannabe pupils suddenly acquiring skills that are diverse from the teaching person's training background, unless A goes with B and the whole lot just slots together with precise alignment like an Ikea furniture kit...….and Ikea do source many if not all of their products from China now.

I'll sum up and say that unless you actually source your components from the suppliers yourself without any rework etc, this project has no hope of flying with fare paying passengers this time this century.

Products from suppliers, that are catalogued, may look and have the same specs, but as they too rationalise their production and sourcing, you would not be aware of a simple design change that could affect your build kits at the assembly end.

Unless you actually supply the complete kits yourself I think you will fall down big time on this concept.
Ian.,

Ian,

At this point, if you don't mind, I'd rather focus on the machine's design than its business model.

If anything, I'll be the first and only customer for it. And that alone will give me immense satisfaction.

Cheers!

[handlewanker]

No problem, it will be an exciting moment when the wheels go round....hope you're going to make a video of it performing.
Ian.

[ishi]

No problem, it will be an exciting moment when the wheels go round....hope you're going to make a video of it performing.
Ian.

You'll be among the first to see the chips.

[ishi]

I could not resist adding ribs to the underside of the X-Axis carriage. Weight got increased by 8kg, but I'm willing to bet that rigidity got a huge boost...

[ishi]

It took me about a year, but I'm finally starting to understand how alignments and joints are supposed to work in Fusion 360. Once you work with the software instead of trying to fight it, it's actually quite enjoyable... And I love the ease with which chamfers and fillets can be added. User parameters also seem to work really well. Now, I just wish I had spent more time reading the manuals (I mean, watching the tutorials) when I started a year ago... Anyway, here is a first version of the carriage assembly with rails and motor brackets, all perfectly aligned and jointed. Tomorrow, I'll add the ball screws, including the shaft couplers.

[ishi]

I find beauty in some of these pieces...

[ishi]

By the way, if you do a lot of CAD and have never tried a 3Dconnexion SpaceMouse, you might want to give it a shot. Once you figure out that the knob is your piece, it's really effective. And it gives you 6 degrees of freedom, in exactly the same way as the machine we're designing offers them. I can't wait to get this device connected to the machine directly...

https://www.3dconnexion.com/products/spacemouse.html

[ishi]

Finally, here is a complete assembly for the X-Axis carriage. I still need to add the nut bracket and linear encoder, but I'll work on these once I have the X-Axis drivetrain and the Z-Axis carriage properly modeled. The nut bracket should be easy thanks to the underside ribs that were added yesterday, but I expect the linear encoder to be a bit of a challenge because of space constraints. Most likely, I will mount it close to one of the ball screws, using the flanges that were added to the carriage in order to install the motor bracket and ball screw support mount.

What did I learn in modeling this assembly? That iron castings are awesome, because you can add all kinds of mounting options for peripheral components, with a minimum amount of machining.

Next: X-Axis drivetrain so that we can put the X-Axis carriage into the new machine assembly.

[Goemon]

Goemon,

I totally agree with you. Granite was just simpler for me to design for. Simple shapes, no internal frame, etc. This is why we started with this material initially. Now that our project has evolved the way it has though, we are seriously considering switching to epoxy granite. It would be a lot more design work and the selection of the supplier would be more challenging, but the benefits are clear, exactly as you outlined them: much better vibration damping, higher robustness, easier mounting of rails, etc.

In order to facilitate our investigations, can you point me to some literature on the subject? Also, can you recommend any supplier?

Many thanks in advance for your help.

Ismael

I don't know of any vendors that sell a pre-mixed epoxy granite kit. Most of us here would make our own. It's very easy.

There are plenty of sites where people have documented their own epoxy granite builds and provided details of the mix they used, From my own trial and error, I couldn't really detect a huge difference from varying the ratios of sand and rocks. Just keep it simple. The biggest increase in strength came from adding carbon fiber and steel to my design.

For the E.G., I ended up using pre-mixed sand and granite from Homedepot. They sell it in 50lb bags for around $5. It's the same sort of mix contractors use to mix with concrete. For the resin, I mainly used 820 from soller composites with their slow hardener. It's the best room temp cure resin I know of. For the smaller sections that's fit in my curing oven, I used a stronger heat cure resin (the same kind I use for my carbon fiber parts).

For a commercial operation, I would recommend using all heat cure resin. As well as being stronger with higher temp resistance, it would allow you to make multiple bases per day instead of it taking 48hours to cure one.

I also did all of my own carbon fiber work on my gantry. I know of companies that make (and sell) carbon fiber gantries and there are a number of companies who make epoxy granite machine bases but I don't know of any that do both. If you don't want to make your own, maybe you could just buy ready made epoxy granite bases.

If it was me, I would keep the epoxy granite base manufacturing in-house. Finished epoxy granite machine bases weigh as much as cast iron ones (or more). They aren't the easiest (or cheapest) things to move around.

[ishi]

I don't know of any vendors that sell a pre-mixed epoxy granite kit. Most of us here would make our own. It's very easy.

There are plenty of sites where people have documented their own epoxy granite builds and provided details of the mix they used, From my own trial and error, I couldn't really detect a huge difference from varying the ratios of sand and rocks. Just keep it simple. The biggest increase in strength came from adding carbon fiber and steel to my design.

For the E.G., I ended up using pre-mixed sand and granite from Homedepot. They sell it in 50lb bags for around $5. It's the same sort of mix contractors use to mix with concrete. For the resin, I mainly used 820 from soller composites with their slow hardener. It's the best room temp cure resin I know of. For the smaller sections that's fit in my curing oven, I used a stronger heat cure resin (the same kind I use for my carbon fiber parts).

For a commercial operation, I would recommend using all heat cure resin. As well as being stronger with higher temp resistance, it would allow you to make multiple bases per day instead of it taking 48hours to cure one.

I also did all of my own carbon fiber work on my gantry. I know of companies that make (and sell) carbon fiber gantries and there are a number of companies who make epoxy granite machine bases but I don't know of any that do both. If you don't want to make your own, maybe you could just buy ready made epoxy granite bases.

If it was me, I would keep the epoxy granite base manufacturing in-house. Finished epoxy granite machine bases weigh as much as cast iron ones (or more). They aren't the easiest (or cheapest) things to move around.

Goemon,

Thanks a lot for all these details. Unfortunately, we're not (yet) setup to do any manufacturing in house. We'll have to outsource.

I have yet to get any reply from these guys:

Home Page | BaseTek

We'll keep looking for a suitable supplier...