5 Axis MM Conversion Discussion

[Tom Ayres]

After you asked your question, Darren, I've been thinking of how a 5 axis machine can be made. Whew! The conception of a 3 axis is easy comparatively speaking. I guess the start point would be to find software that can handle 5 axis then figure out how it handles the additional 2. For another day...

[Fox]

5 axis software is super expensive. So you'd better find the affordable full 5-axis software and then build the machine. The machine is the 'easy' part. CNC zone has some cool 5-axis examples. http://www.cnczone.com/forums/diy-cnc-router-table-machines/98554-5axis-photos-new-build.html

[darren salyer]

Zero aspirations for a 5 axis here.

[pblackburn]

If you have $10,000 USD you can play with the software that is needed for it to work effectively. Rhinocam has Visualmill with 5 axis continuous machining.

[Tom Ayres]

Didn't say it would be cheap, or easy, but the concepts are cool. There must be a hundred or more ways to make one workable. Pete, surely they have a free demo or something...

[pblackburn]

That is just the software. To make 5 axis work effectively, continuous machining is the way. Everything works in harmony. You can go indexed machining but not worth it in the long run. Rhinocam is well spoken of, however I never used it.

[domino11]

Tom you should get one of these then. :)

http://www.doughtydrive.com/

[darren salyer]

Im always afraid when you have to call or email for prices. Pretty cool though..
How does SB do their 5 axis machine?

[ger21]

I think they're around $2000/axis.

[darren salyer]

Someone else can be the pioneer on the 5 axis setup.
I nominate Tom Ayres.....
Do I hear a second???

[racedirector]

Seconded! :)

[Tom Ayres]

Vetoed! :D I'm still going to give it a hard look though.

[darren salyer]

Too late, Tom, its already been seconded.
All in favor say Aye......:)

[pblackburn]

Aye

[racedirector]

Aye! Sorry Tom, waiting with baited breath on you getting one of these drives and making it work :)

[darren salyer]

Aye

[Tom Ayres]

I'll be happy to...if I can receive financial support from each and everyone of my MM Brethren ;)

[darren salyer]

You have my moral support, sir!!

[Tom Ayres]

I already have some morals...

[KenC]

Some moral isn't enough, I'm give even more moral support :D

[Tom Ayres]

Ok, Ok, I'll 'look' in to it...After watching the video of the DoughtyDrive B/C axis assembly, way cool by-the-way, there would have to be a major y-car change and the height would have to be at least 3 times taller than what we have as a standard to be effective. The spindle would have to be air cooled (I don't think a water line would survive all that flex). And then there's the issue of Z flex. Crazy changes to make work. It does give me ideas for modifying their concept though. I'd have to start a thread on cnczone, not going to happen any time soon...

Any one of you machinist guys interested in fabricating a few parts if I can design something reasonable?

[darren salyer]

Now there's the Can-Do attitude we've come to expect from you, Tom. :D

[Tom Ayres]

More the 'Can-Try' attitude.

[domino11]

One little link started a war I think.. :D

[Tom Ayres]

How does one zero 5 axis? "How many licks does it take to get to the center of a tootsie roll tootsie pop?":rolleyes:

[domino11]

I have seen videos where they have machined an aluminum box and they touched off with a probe to zero to all five sides of the box first then started the machining process.

[Tom Ayres]

I'll have to look for some vids...

[darren salyer]

No war, Heath.
I'm intrigued but will admit it is way past my skill level.

[Tom Ayres]

Well there will be no 5 axis using the b/c drive, cost is $2950 + Freight plus another $945 + freight for a 2hp (1.5kw chinese) spindle. Ha Ha, LOL. the drive uses 300 in/oz steppers that can't be changed, claim zero backlash.

[darren salyer]

That's not terrible pricing if you needed to start a 5 axis machine.

[pblackburn]

Add that to the low cost of the software to generate your gcode. 3k + 5k = 8k. Or you could build another Mechmate when your business need justifies it. ;)

[Tom Ayres]

You are right, but if it can make it for less...now we're talkin'. I don't really see the need for anything I do though. The videos of a 5 axis Thermwood on youtube shows more in line of how I'd want mine designed. Looks more solid. I just don't understand the cabling set-up, you'd think it would twist right off but maybe the software is aware of that.

So true Pete, but you'd still have a 3 or 4 axis machine.

[pblackburn]

It can be done
https://www.youtube.com/watch?v=nITLI_WcnuM

[sailfl]

Very nice machine. I think it is designed to cut only foam. Light like a electric foam cutter.

[racedirector]

Heres another one on kickstarter.....

https://www.kickstarter.com/projects/443862761/regal-robotics-500-series-large-scale-5-axis-cnc-r?ref=category

[Tom Ayres]

Yep, but I don't do foam. There's no way that kickstarter machine could work on wood, It'd be all over the place from flex.

Sorry Darren, we've somehow taken over your thread...Next subject please...post something you've done recently...thanks

[pblackburn]

Sorry Darren, we will move this to Tom's thread

[darren salyer]

Its all good.
Been under the weather the past couple weeks, so no new pics to post.

[pblackburn]

Tom, I believe the the DoughtyDrive is using a planetary gearhead with positive drive belts. If it was a harmonic drive, I can see the Zero backlash statement but the site says 30:1 with + - 20 arc seconds. This would lead me to think it is a planetary reduction. The NEMA 24 stepper would be fine with that much reduction.

[pblackburn]

Maybe the easiest avenue that would be the most useful for smaller detailed parts would be a trunnion. You could make detailed roses and other items that I think would be more useful to the woodworker. Anyone here ever venture that route?

[Quadro]

5 axis is achievable, the plasma one i designed (http://youtu.be/bZ2xyM2Gm9I) it to be able to take a 1.5kw square air cooled motor as well. The bearings I used were ball bearing, if it were to be used with a router I would up grade it to tapered bearings.

The CAM software is the one of the main hurdles.

[Tom Ayres]

Pete its obvious that I need to learn a lot more. But what makes the planetary drive better than the harmonic drive? I haven't seen a "positive belt" yet. Do you have any visual references of the differences?

[Tom Ayres]

Oh, a trunnion bed was my first thought but is very limited to size. I have seen them being used for milling engine blocks and porting heads. Not really the direction I'd want to go.

[pblackburn]

A harmonic wave drive is better as it is truly Zero Backlash. Planetary is always in the arc seconds or minutes depending on the quality. Harmonic drive is the way to go but finding a reduction from 10:1 to 50:1 is hard and expensive. That is why most opt for planetary gearheads. Too much reduction and your speed to torque ratio becomes a factor.

The trunnion would be suited for most people unless you are building a large scale model in the feet range.

[ger21]

I think the best (cheapest) approach is to study all the products available from a manufacturer like Harmonic Drive, and design around those components.
http://harmonicdrive.net/

Then watch Ebay constantly and hope what you need turns up. Expect to pay $250-$500 each minimum. Also expect some rather complex engineering, and some custom fabricated parts.

Realistically, though, a 5 axis router is really good at certain things, but below average at most other router tasks due to a lack of rigidity. Unless you have a steady flow of work that requires a 5 axis machine, the only reason to build one is for the cool factor.

[pblackburn]

I agree with Ger. We have a Haas VF6 with a trunnion at work. Great machine that is underutilized. I see the most useful items that a 5 axis can make are small items. The super large stuff is rare and in a specialized field. The small items are added to your existing product to give the wow effect.

[Tom Ayres]

Ger Thanks for the link, lots to chew on.

What are the common sizes for these trunnion beds? Or should I ask, what is the most useful bed size in your opinion? Are we talking a 4" x 4", 8" x 8", 12" x 12", bigger?

[Tom Ayres]

Well, had a little time to look into the harmonic drives that Pete had mentioned. Found that there is another world out there that needs to be explored. I'm intrigued and very interested in doing a 5 axis but will report back when I have a better grasp on some ideas.

[sailfl]

Not to throw water on the fire, but if you build a 5 axis machine, how many times do you think that you would get some one that would want you to cut some thing that requires 5 axis machine and would they be willing to pay what it cost? Yes, it would be fun to own one but how often would you use it?

[Tom Ayres]

It's not about the use, if it's available jobs can be found. It's more about the challenge and discovery to me. We all can try to justify doing something like this for whatever reasons, but how many do it just to learn something beyond our norm, few. I've always sought challenge over practicality (if I could afford to do so), I agree not always wise though. But right now I can at least give it a little attention for curiosity sake.

[smreish]

If one does decide to upgrade the MM platform with a b/c head, think about the extra mass and uplift on the y-car. I would suggested adding a second drive motor and gear rack to the hold down side of the car. The extra mass and extended moment arm is a bit for one motor to handle.

I almost added a second y motor on my extended slide MM #5 for this exact reason.

Not a huge cost, but a consideration while your pondering the future! :)

[Fox]

If that will be the case why not just use a stronger motor !? Saves the extra expense and hassle on rack, gearbox driver etc. .... the y is not that far apart tot worry about tilting IMO.

[smreish]

Fox,
You could use a bigger motor and larger amplifier.
But, if you were upgrading your current machine, the G203 and PMDX allow for "step sharing" on the output of the card might be an affordable solution.

Good points made by all.

[sailfl]

That is why we have a mind, to think, dream, create and enjoy!

[darren salyer]

Related to Walt Disney, Nils?
I like that quote.

[pblackburn]

I truly believe a trunnion addition makes more sense than having the spindle position itself. I can see the market for small intricate items more than the large. It cost of making the trunnion would be small compared to the cost of the software to program continuous machining.

[Tom Ayres]

Well here is a 80:1 Harmonic Drive gear I got from Ebay this week, new old stock, I couldn't really retrieve any specs on it in the short time it had left in the auction but got it for less than $27.00 delivered (couldn't take a pass on it if it would work). All the others run $399 and up, most of the larger ones around $1k I'm not sure I'll be able to use it for additional axis, seems small, only 2.75 inches.

Well if I can't use it for this I'll use it for gear reduction on my bead roller build I will one day complete. (or relist on Ebay for some profit, lol)

[Tom Ayres]

Oops I forgot to post the picture of the drive
https://farm6.staticflickr.com/5476/14006302901_58ab07c7c1.jpg

[pblackburn]

Size is fine, with that much reduction, the torque will be great but your speed will be slow. You will be at the high end of the stepper speed so the stepper torque will drop off. I would not be afraid to try it. You will have a little time in designing a mount for it with a seal. Looks like you made out well.

[Tom Ayres]

Pete, you are right, lotsa torque, I was hoping it was going to be 30:1 but no such luck. Heck it could be an indexer gear if nothing else. Of course, after getting this in the mail today I looked on ebay and found a bunch more with the motors which I didn't find on a previous search, oh well, I still got a deal. Its a pretty cool gear set-up.

[Duds]

You could always pick up a six axis arm style robot for $8-10k fit a spindle to the payload and bolt it to the garage floor. Just remember to duck when it comes swing your way.

[isladelobos]

The old robots can't do this, their control boxes (servo motors and his drivers) have very little capacity for large files or g-code. new robots are very expensive, and the work area is reduced or need a very expensive platform with more than 6 axis.

You can put new drivers to an old robot for control this with a new control box but is the same, more than 10K in equipment.

The Mechmate chassis and design not is a good idea for a 5 axis machine, because use V wheels and vertical gravity forces, need a lot of modifications for maintain long Z axis in lateral forces.

I'm testing now linuxcnc with 5 axis and steppers, seem the latest versions can do 9 axis.

15130

[isladelobos]

For these about software,

This is an interesting project:

https://www.kickstarter.com/projects/2003668803/5axismaker-first-ever-affordable-5axis-multi-fabri