Is this normal?

[A_Camera]

I bought a 4th rotary axis, which I am not sure if I should be happy about on a "you get what you pay for" basis or should have received something better. Are my expectations too high from these 300$ things? This is what I bought (Option 2):

https://www.aliexpress.com/item/1005002123527118.html?spm=a2g0s.9042311.0.0.47104c 4dI8LQr7

Everything looks nice, but...


https://youtu.be/EzWOo5JMguk

So the question is, if you think this is "normal" or not. I did complain to the seller, he asked to make a video to show the problems, so I made one today. I am still waiting for his reaction, but basically what I asked for is:

1. A new large gear or a chuck or both, because I thing that's the biggest issue.
2. A new MT2 live center., or at least a new pin for my MT2 taper.

Unfortunately, I don't know how to replace the pin or the large gear, so I also asked for instructions in that case. I don't know if the center pin is glues or not, but it sits so tight in the bearing that I can't remove it. I removed the three screws at the back of the large gear, but then there is nothing else to remove, and the gear still is firmly attached to the axis, so even there I need instructions because I don't know what to do. The chuck is more straight forward, I din'd try to remove it, but probably held only by three screws.

What do you think? I have zero experience with rotational axis, but dry running seems fun and works well with UCCNC.

I am also making a new 48V 7A supply just for this purpose, and also made a "Lathe mode" box, which basically is giving pulses and direction signal so that I will be able to run as a "lathe". I know, I know, don't laugh... it's not a lathe, but I will play with it and it may work for soft material. Anyway, if not good enough but I like the idea then I'll change the 1.8Nm motor to something considerably stronger, or a servo. Anyway, this device is based on an Arduino Uno and it will give pulses matching a set rpm, which can be set (as it is designed) between 0-350 chuck rpm. Of course, it has a display and it will accelerate/decelerate as well.

As usual, if there is an interest for this box then I will share the Arduino code, but will first make a box and everything before I will show it. It's a simple thing, but will be fun to try.

[m_c]

Woohoo, I can finally reply.

I wouldn't be too worried about any of that.
The pulley runout isn't ideal, but as long as the belt isn't going tight at any point, I wouldn't worry.

Runout on three jaw chucks is perfectly normal, as chuck scrolls are not a precision item.
If you need better tolerances, then you either need to use custom bored soft jaws, a collet chuck, or a 4 jaw chuck.

[cropwell]

Now that replies are possible :-

m_c The chuck as far as I can tell is a 4 jaw and hopefully independant jaws.(edit K12-100 4 jaw is self-centreing - meh).



My comment (sent as a PM) is that you should remove the chuck and measure run-out on the face and periphery of the back plate. If this is within limits then you should be able to set up the workpiece in a 4jaw chuck using a dti (plenty of vids on Youtube).

Run-out on the driving pulley unless massive is inconsequential.

Hope this helps,

Rob

[CharlieRam]

As per the above comments, runout of the pulley shouldn't be an issue, the 0,1 runouts of the chuck could be down to the chuck centering but also shouldn't be much of an issue on a router. I think the most important thing will be the backlash in the chuck axis (C axis?)
I tried to reply to this on Friday but it wouldn't let me!

Sent from my M2003J15SC using Tapatalk

[dazp1976]

I'd say that for both ends to move a bit. The alignment (balance) of the internals - bearings, shaft, etc, to the body. Are just slightly off from straight.
Machined parts will likely come out looking fine.

[JAZZCNC]

You have to remember these are rotary axis, not lathe's so the runout will hardly be noticeable for the majority of work and if you require higher precision then you either use 4 jaw chuck or use soft jaws and bore them as MC suggested. The pulley runout is of no concern whatsoever in this application.

[A_Camera]

Now I am back. I was a bit busy so I could not answer. Anyway, I don't know what happened, why replying was not possible earlier.

I didn't want to pull it apart before I was through the discussions with the seller, but now we made an agreement so I am good to go on. Basically, the seller admitted that the MT2 live center is not normal and we agreed on my proposal refund which basically covers the purchase of a new live center. He could not ship one from EU, and I did not want him to ship from China, which is why we agreed on refund so I can buy from someone else. For the chuck play, he basically said that that's normal, "you get what you pay for", which I understand very well also, even though I was hoping for something better and it is actually outside his own specifications.Anyway, with the refund I paid 250 USD for the 4-jaw 100mm 4-th axis and the MT2 tailstock with the damaged center pin, which I think is still usable, unless I really need the very sharp pinpoint edge. I think that's a good price.

What I did so far to improve the situation is that I removed the chuck and the pulley and cleaned up everything. I reassembled the pulley and also moved the stepper pulley out a bit. After this it's a bit better, there is not as much play in the pulley as before, but of course, the chuck is the important part.

Before reassembling the chuck, I made some measurements on the chuck plate, and that shows no real problem as far as I can judge, so that's good. The thing I noticed was that it was very hard to remove the chuck, and the reason for that is because the chuck plate inner flange (I don't know if that's the proper name of it) is I think a bit too large in diameter, so the chuck is VERY tightly fitted on that flange, I think it is too tight because there is ZERO possibility to adjust the chuck. I will see if I will try to do something about that later, but for now I don't have time, so I just reassembled the chuck after cleaning the surfaces and made some new measurements. To my surprise, the play is now half of what it was before, so it is definitely within the specifications, which means that I am satisfied for now.

So for now, I am OK with this as it is, I will test it ot for some real work also before I try to fix anything more. Perhaps one thing I will do is that I remove the chuck once again and try to clean that flange with some very fine sand paper, maybe some grit or something else will be removed by doing this which might improve it even more. I feel that before staring to grind anything I will need to learn some more about using the 4-th axis. I don't want to destroy it or make it worse, after all, a slow start with a less then perfect chuck is better than a destroyed and useless one.

[A_Camera]

If you need better tolerances, then you either need to use custom bored soft jaws, a collet chuck, or a 4 jaw chuck.
You are right, the pulley play does not matter that much, unless the play is transferred to the chuck.

This is a 4-jaw chuck, it is a self centring type, I don't know if that's good or bad, I think in some situations it is good, in some it is bad.

[A_Camera]

My comment (sent as a PM) is that you should remove the chuck and measure run-out on the face and periphery of the back plate. If this is within limits then you should be able to set up the workpiece in a 4jaw chuck using a dti (plenty of vids on Youtube).


Thank you for the comment and the PM. I was too busy so I had no time to answer earlier. Don't know what happened to this post or the forum, why replying was not working before. Anyway, yes, the is a self centring 4-jaw 100mm chuck. Maybe it's a bit too large for my tiny CNC, but I thought it's better to buy a bigger one than a too small one, and also got it what I regarded a good price, 280USD including EU taxes and three days shipping from Spain. Now with some refund the price I paid was only 250USD, so that's even better.

Anyway, I did what you suggested (after I agreed with the seller) and the chuck plate seems fime apart from what I mentioned before about the flange, which I think is a bit too large. I will see if I will continue improving it, though grinding is not my favourite first choice, since I don't have anything which is good and reliable enough for this sort of precision grinding. The Dremel seems very rude and inaccurate, but maybe that doesn't matter. Anyway, I will see later on if I will continue or not, and in which order.

One thing I discovered was that the jaws for the ID clamping (not the ones which are in the video) are definitely something wrong with because one of the jaw is not closing completely, but maybe that's normal, maybe you not supposed to clamp in it like if it was OD jaw. I will need to check how well it grips on a pipe before I conclude that there is something wrong with it, but in any case those jaws are not as important as these OD ones for me.

[A_Camera]

I think the most important thing will be the backlash in the chuck axis (C axis?)

Backlash is good, as far as I can measure, but will make some real tests later on through milling a piece of material.

[A_Camera]

You have to remember these are rotary axis, not lathe's so the runout will hardly be noticeable for the majority of work and if you require higher precision then you either use 4 jaw chuck or use soft jaws and bore them as MC suggested. The pulley runout is of no concern whatsoever in this application.

This is a 4-jaw chuck... which is why I expected higher precision, but sure, I understand that for most things this error will not make a difference.

[magicniner]

4 Jaw INDEPENDENT chucks give better precision than 3 jaw chucks because each jaw is adjusted INDEPENDENTLY.
What you have is a 4 jaw scroll chuck with nothing to give you better precision than a 3 jaw scroll chuck.
You can get scroll chucks to meet your expectations but expect to spend a lot of money, more than you expect!

[A_Camera]

Yes, I understand the independent chuck jaws can give good precision because each jaw can be adjusted individually, that's obvious. But it also means a lot of extra time before the work piece starts spinning. Anyway, it looks like that I could get MUCH bigger and better precision by just cleaning the surfaces.

[mekanik]

I seem to recall you can get 3 jaw chucks that have adjustment in the backplate, you put a short mandrel in the jaws and dial out the runout then lock it.

Regards
Mike

[m_c]

I seem to recall you can get 3 jaw chucks that have adjustment in the backplate, you put a short mandrel in the jaws and dial out the runout then lock it.

Regards
Mike

I don't think I've ever seen a scroll chuck with that kind of adjustment, but I have seen collet chucks with that kind of setup.

I should have said independent 4 jaw in my original post, but Magicniner has already clarified that.

You can improve accuracy by grinding/boring the jaws (remember you need to clamp them onto something near the correct size first, to preload them, and standard jaws are generally hardened to some extent), but as I mentioned, that accuracy is likely to vary over the scroll. Scrolls are rarely a precision component, and accuracy varies over the scroll.

[Kitwn]

This link is to a video by Dan Gelbert who started from nothing and built up a whole business based on high precision machining. In this video he demonstrates a scratch-built lathe accurate to one micron, including details of the chuck. This is only one of many videos by Dan. Fascinating stuff for geeks like us.

Kit

https://www.youtube.com/watch?v=sFrVdoOhu1Q&t=40s

[A_Camera]

This link is to a video by Dan Gelbert who started from nothing and built up a whole business based on high precision machining. In this video he demonstrates a scratch-built lathe accurate to one micron, including details of the chuck. This is only one of many videos by Dan. Fascinating stuff for geeks like us.

Kit

https://www.youtube.com/watch?v=sFrVdoOhu1Q&t=40s

Thank you for the link. It's really interesting to watch, too bad it ended so suddenly. I wish there was a part two and three also...

He said it's "very easy" to make a machine like that, though I think that's overestimating most of the humanity, because even though I regard myself as very capable of doing many things, I doubt I could make one which is as good as his machine. Anyway, it was a joy to watch that. I love the idea of air bearings also and his demonstration of it. I also love how quiet that machine runs, but I suppose that's largely one benefit of the air bearings as well.

Back to the topic, yes his chuck is special, like the rest of the machine. The idea of moving pins in the chuck jaw is good, but again, requires very high precision, so in the end, I think traditional chucks are better for the "average Joe" most of us are.

[cropwell]

A 4 jaw chuck allows you to hold square stock. An independant 4 jaw allows you to do offset milling, but is a PITA to set up every time if you don't need it.

[A_Camera]

A 4 jaw chuck allows you to hold square stock. An independant 4 jaw allows you to do offset milling, but is a PITA to set up every time if you don't need it.

Yes. That's why I was only interested in the self centering type. Individually adjustable chucks are very good when really needed, but I figured that an independent 4 jaw is a real PITA most of the time because every time you swap a work piece you must recenter it, so in the end you might not get better accuracy in general due to lack of patience, even if on occasions you can get 100% accuracy after spending a lot of effort on centering. I don't think I will buy a 4-jaw chuck with independent jaws, but will however buy a 3-jaw chuck later on since that is more suitable for some work.

[Muzzer]

Another option is multisize collet chucks. https://www.ebay.co.uk/b/burnerd-collet/bn_7024823313

Still quite a few around and a set will cover a wide range of sizes with very good runout. There are several types but mine (EC?) go up to 38mm with no gaps. You might get a chuck for £200 and a set of collets for a similar price if you keep your eyes open.