Spindle and rotary axis additions

Add-ons & Modifications 🧰 CNC Setup
21

@gwilki @thisoldlight
Yes I want to agree, but I am not giving up, yet.
I have read the entire Scienci website in the last 2 days, and I have watched Andy’s and Chris’ videos and many more… I have been looking for a low-end solution for a long time.
I nearly purchased the fat round bar machine competitor of Sienci, last week.

And this is an 11th hour turn of events for me… months and months of snooping/learning, gathering a slop of incongruent bits and pieces. I have watched 100s of videos… but I am also very busy elsewhere, so until I revisit repeatedly my understanding flies out the mind.
The Sienci website gave me more answers than many other sources… perhaps I am also more versed.

I had been acquainted to the MK1(only new it as the Longmill then) quite a while back, but had not given it the attention it deserves. I was fortunate enough to revisit at the last minute, to find out I should rethink my approach.

From my extensive noob knowledge I have full confidence the MK2 will handle my requirement. I am not doing anything too high-end. …also, not a production scenario.
Though I posted the link to the Wire EDM setup and I also look at high-end level installations, that does not place me in that realm. Too poor for that.

I have now a better feel for what matches my scenario, I know the router can be a weak link and why I have already opted for a spindle.

The only thing that bothers me is that the Sienci new-kid-in-town next to be offered, is near …and I may have to buy twice. I have been gearing my search to make sure to avoid that… though in their videos they stress that the machine itself was built to accept the upgrades coming.
I am also aware of other superior hardware platforms that are a step up from ARDUINO.
I may have to source that separately.

I am grateful to have gotten so many answers so quickly on my post/reply which is not even in its proper thread.

Your comments are very appreciated.

22

@Andy1
I should apologize to Andy… he brings up something to really investigate.
Do I smell a Sienci secret?

Thank Andy, cool stuff.

23

This may be wrong. I was reading the Vectric docs and it seems for rotary toolpaths that either the X or Y axis is mapped to A. I guess it makes sense because in grbl there is no A axis. In the post processor folder there are post processors that end in XtoA or YtoA. The docs say that it might be necessary to create a custom post processor using the existing ones as a guide.

Maybe there is still hope for rotary tool paths just not for a true 4th axis without replacing the controller.

@NIQ I guess you should take my info with a grain of salt because as you have probably realized I am still figuring some of this rotary stuff out.

24

@_Michael
yes Michael, the various replies above actually spell that out, I am fully aware now of the xtoA, YtoA manual swapping of the axis.

And now that I am better versed with this issue, though a true 4th axis would be the ultimate solution, I am ok with just the rotary capability.
Also, many videos on the Internet show people have successfully implemented a rotary setup on many different machines.

I am now trying to redact a full reply that will include all the pieces, so I can put it in front of Sienci’s eyes so they can confirm its viability.

I am also sending that same document to third party outfits that have circuitry that is GRBL SEP32 level, fully compatible with basic GRBL. …and here we run into the CNC CONTROL bit, gSender part of that equation… not to forget the post processor, as you mentioned.

…on an outside note: I suspect that the near future improvements Sienci Andy & Chris allure to, is just that, a move towards the next level of GRBL.

*One can never know, it could be a multiple articulated arms bot that blows all basic axis systems out of the water… (ex: single arm Dobot)

…like a lathe/mill combination bot that can dynamically hold and machine a stock in all spherical directions needed. huhh! why not a machine that can roam the Earth and locate the proper material ore, load it into its hopper and spit out an EDM level part at the production spout! LOL

Progress is being made., Thank you.

25

One note: I typed the word ‘gear’ or ‘rotary’ (can’t remember)’ in the forum’s search box, and 1 thread, partially on this subject, came up.

26

OK, progress…

I place this reply in this thread, because it is a continuation of the subject, and so it can help others, and some of you may chime-in to clarify my beginning meager understanding and answer the questions below.

I also think a moderator should move this part of the thread which started at my original post/reply out of this ‘Laser can do…’ thread… Laser mentioned here-below as well.
Title could be: SPINDLE and ROTARY options for MK1/MK2.

…and I know, Andy clearly states they do not support the spindle setup, one must get support from the spindle provider.
I also wish Andy and Chris would kick in on this and provide all the answers to the below questions in a write up like on the SPINDLE page on the Sienci website.

…when combined, one not familiar and presentations and discussions peppered with I N I T I A L I S Ms , it takes forever to get things across rapidly in one doc, done.
I found the Sienci page on SPINDLES and finally decided I had to check the VFD initialism.

…and that obviously affects the entire discussion, as the rotary add-on is in the middle of this fire.

Now that I know more, I see the SPINDLE setup implies a VFD.
So, the VFD drives the SPINDLE and I am assuming it also runs the axis stepper motors!?,
Am I correct? please confirm.

…and questions:
What pieces of the MK2 are replaced/stay the same?
…what dominos fall here?

  • Will the VFD drive the Longmill Laser
  • CAM gCode generation, post processor?
  • will gSender/toolpath be affected?
  • will the Longboard remain in the show?
  • wIll the Longboard have to be run separately yet concurrently?
  • Wiring?, etc. (surely the wiring will change)

That took me to the EMAinc website.
It will take me some time clear the waters, obviously.

Thanks everyone for your interest and contributions.

27

@NIQ I don’ know who you are asking to confirm, but I’ll provide what I know.

You are not correct in assuming the the VFD runs the stepper motors. They are powered by drivers in the Long Mill controller. There are four of them.

None of the Mk2 pieces are replaced by the VFD. The spindle replaces the router. The VFD powers the spindle.

No, the VFD will not drive the laser.

I don’t know what you mean by your second question.

When you ask if gSender.toolpath will be affected, affected by what? The VFD?

Yes the Longboard will remain in the show. It controls the motor drivers, and the arduino in it processes the gcode.

I don’t know what you mean by your 5th question. When you say “run separately” again “separately” from what? If you are referring again to the VFD, I respectfully submit that you do not understand the purpose of the VFD. It does nothing but provide power to the spindle. It has no function with respect to the stepper motors.

Again, your sixth question is too vague.

I have been waiting to split this conversation off to a new topic since it started moving so far off the original topic. I will do that now.

28

@_Michael Michael: I believe that you will find in the Vectric docs that the way in which their software handles a rotary tool path is to “flatten” X. Think of unrolling toilet paper off the roll and lying it flat. That’s what Vectric does. Then, it’s a simple matter to create the tool paths.

Strictly speaking, this not 4 axis milling. It will work, but it has its limitations.

29

@gwilki Thank you very much for splitting the thread.
You emphatically confirm that it is not easy to ask the correct questions as I expressed in my previous reply, since it gave you trouble following what I ask.
But you gave me enough clues, so that I can investigate further using an improved approach.

Note: I read anew more attentively into the ‘Adding a spindle to your Longmill’ a gorgeous write-up, by the way.
…and I see the answer about the gSender/post processing I had listed.

  • under the topic ‘Firmware and gSender settings’

Thank you.

1 Like
30

Hi @NIQ . I am new to the world of CNC (just 3 years) and I had been led to believe that a 4th Axis was generally not an easy to provide solution with 8bit GRBL and Arduino controllers. Since then, there have been many other controllers including 32 bit ones which I think will manage the task more easily. Looking in the wayback machine at SourceForge has revealed these potential solutions:

I have a friend who has recently installed a 32 bit Duet controller on his new CNC machine (Rat Rig) and I believe that true 4th axis is possible with this controller. He is @NewAtThis and he may be able to assist you in some way or another.

Finally, hop on over to LightBurn laser software forum. You may have to dig a little but there are folk who have added a rotary axis to their machines and that information should be easily available. There are some super helpful people over on that forum.

31

@jepho
O! thank you thank you… now I have a lot to chew on.
3 years, sounds like a vet to me… I am ground zero noob.
The stuff you bring up is gold. The leads you provided will speed up the process immensely.

I started to investigate the GRBL SEP32 from Bart Dring.
…and as I mentioned earlier, I now understand better that just a rotary axis is not a true 4th axis, and that all I may need is the rotary… but I also know the rotary is limited, compared to the true 4th axis…

So, I am leaning to adjust my plan to go for the simpler rotary at first and cut my teeth on that, then go for the 4th…
But I will conduct my research including the 4th axis info, to perhaps jump into that right off the bat. …certainly the learning experience will not be a waste.

I so thank you.

32

:grin:
No worries. You are welcome. We all start at ground zero too.

Yes, his stuff looks good. I am not sure what limitations are imposed by GRBL in 32 bit. A lot of CAM software should be able to use it without any difficulty. I do not know much about Vectric software but they can map a toolpath onto a 4th axis. I suspect you will need to a fair bit of research to match controller to software. I think my ideal machine would probably be a Stepcraft for simple hobby level working. Production environments would need something else.

You could look at a Masso/Gecko controller for true 4D but not cheap. The CAM software would be quite high end, I suspect but UCCNC would work because Stepcraft configure this to work with their machines. I think I am correct in stating that their bottom of the range machines (D series 840) can use the 4th axis and an automatic tool changer which they supply. As work envelope increases you would need to step up to their M series. Stepcraft appear to offer the most complete CNC solutions to a hobbyist.

The largest D series (840) with a 4th axis.

My intuition (may well be wrong) is to try for a properly capable controller and get used to it and the software requirements for CAM. Then you only have to fit the axis when you feel like you are wanting to try. Of course, it is great to know that our machines can do it. I am at the stage of wanting to upgrade my 8bit controller but it remains to be seen as to whether I want to engage with the higher level of knowledge required to cut 4D.

33

Within the parameters I set myself, with a RatRig 1075 CNC machine, I decided on the Duet3 MB6HC, 32-bit controller, which can support 6 stepper motors natively, and easily extended to support many more.

I’ve not really had an answer on the compatibility of the Duet3 with gSender, so haven’t really had much to do with it.

Given my new found knowledge, however, StepCraft would be the way to go, although I am really happy with the RatRig.

Good luck!

34

Exactly Grant,

Here are a couple of resources

Definitions:
What's the Difference Between 3-axis, 4-axis & 5-axis Milling? - CloudNC.

Vectric Pro 10 How to Design a Rotary Project - 4th Axis Rotary - Vectric Design - YouTube

Scorchworks https://www.scorchworks.com/
a free g-code wrapper

Andy

35

@Andy1 Andy you are a gem!

I am so glad I posted here, this is culminating to the understanding I have sought thru 100s of videos that left me empty, asking the same questions in too many places to even track.

The links you provide are dead on the money… saving me massive time from the chaos.

The rotary, I now see, IS a 4th axis(as I had assumed earlier) not a separate realm but only a variation of the available 4th axis options. Thus now I can have more informed exchanges.

I also had wondered how the interaction between the bit and the stock was maximized to acquire the spherical power, as a robotic arm can do. …and that is depicted quite clearly in the 1st link you delivered… the more axis, the more can be done, without resetting the stock(loss of precision)! …and the 4th axis level satisfies my requirement.

@jepho
Stepcraft is beyond my means, but can partially be part of my solution.
I am heavy on their site now, to plug the gaps.
They too are Vectric centered… so, all good so far, ASPIRE remains in the equation.

Thank you immensely to both of you!
…and everyone else for chiming in.

1 Like
36

Not a problem. It is difficult to avoid all of the holes in the beginning. I fell down so many of them, I like to try and prevent it where it can be avoided for other people new to the hobby.

:grin:

37

In terms of everything that changed between MK1 and MK2, I wrote a lot about this in our blog: every thing you need to know longmill | Sienci Labs

4th axis support has been a very commonly requested feature.

There are a lot of different systems beyond just the rotary axis hardware that needs to be updated. The software and control systems also need to be redone as well, and of course, built in a way that’s affordable. I suspect a lot of people say they want us to make a 4th axis since we have a passion to make CNC technology more affordable and bring it to a price point that is accessible for the general hobbyist. It is also worth noting that the CAM for 4th and 5th axis machining is substantially more complicated, which needs further development to bring it to the hobby level as well. Otherwise, I assume everyone would have spent the 100s to 10s of thousands of dollars a 4th axis setup can cost by now. :wink:

At the end of the day, we are working on a lot of different things that will make this tech accessible for the hobbyist. Making one prototype is trivial. Scaling it up so that thousands of hobbyists with no experience is 99% of the challenge. This includes producing in high numbers for economies of scale so that we can make enough money to support our business, as well as doing documentation, testing, and resource development so that users can have a good experience with any of our products.

There are a lot of cheaper options out there in the range of hobbyists. However, I think there are a lot of compromises, such as accuracy, capacity, full 4-axis independent controllability, support, etc… that we’d want to address with our own product. These compromises I suspect is a main reason they are not so ubiquitous.

I think in the next year or so, we’ll have more of a tech stack that’ll make 4th axis what people want it to be: affordable, simple, and capable, but it’ll take a while.

My suggestion would be if you need the 4th axis, find a machine that can do it. If you decide to build something to support it, please share it.

As it stands, the LongMill is fairly primitive electronically. You should be able to swap out controllers and plug in the stepper motors directly into a multi-axis board. We have enough documentation and open source design files that someone should be able to set something up. We can’t provide one on one direction or support, but the info is there. As other people have mentioned, for a 2-axis + rotary set up, you could unplug the Y motors and use a rotary axis, then change the steps per mm.

Based on my research, it’ll probably be a little bit expensive to do this. The proper capacity rotary axis mechanism is not cheap, and the additional electronics capable of 4 or 5 axis is not cheap either. There are cheap options but the support is very limited and requires a lot of technical and engineering know-how to use.

In any case, no precise roadmap, not that we have one to share yet. But rest assured we are a company that is actively in development with a few things that LongMill users will appreciate.

5 Likes
38

Thanks Andy, I will definitely be interested in 4 axis capability when you develop it!

39

Thank you for this information.

40

Hi Andy,
I just wanted to say how much your transparency and openness with all of us is appreciated. It is very refreshing to see!

Thanks,

Greg

2 Likes