The Design and Construction of a 'Backlash-Free' Rotary Table

This will be a many-part series covering my efforts at building a CNC Rotary Table (RT). Please forgive any obvious mistakes. Yes, of course you are going to get my biased opinions. Yes, there will be photos and drawings. No, there won't be detailed plans: they are not really the point of this exercise. Updates will happen every few days.

The background to all this is the rather long thread here at CNCZone about a Backlash-Free Rotary Table, started by Zoidberg
from Sweden in Jan-2009. It went for 80 pages, which is not bad. There were some passionate arguments about technology, especially on just how you can get (make) 'backlash-free' reduction in a home shop or hobby context. Shall we summarise by saying 'opinions differed'? But we all had fun.

Attachment 270536

Let's remove any doubts right up front: this is NOT about hanging a stepper motor on the crankshaft of a manual rotary table. Such tables do not have the bearings for continuous CNC operation, plus they usually rely on manual brakes to stop them from moving under load. As a manual indexer they may be fine, but they are NOT a full CNC Rotary Table.

For the most part, the discussion was about the means for gearing the motor rotation down to to the table shaft. Arguments pitted ordinary gears against worm drives against toothed belts. There were some esoteric variants, but generally they were just
'variants'. Some felt that all the options have backlash, while others (myself included) argued that modern toothed belt profiles
do not. A few felt that gears could be made backlash-free, while others argued for special worm drives. Various rather expensive
commercial variants were brought into the discussion, but it was generally conceded that they are unrealistic for a hobby
construction. The issue was never really settled to everyone's satisfaction.

Attachment 270538

The really big problem here is the rather large reduction ratio needed between any drive motor and the table itself in order that
fine positioning can be had at the rim. Of course, if you don't want fine positioning at the rim then life becomes far simpler, as
shown here. OK, maybe that is a shade too simple? Mind you, for pottery it may be just fine.

Attachment 270540

If you have really simple requirements you can even dispense with any reduction, and just go with a chuck on a stepper. That might be adequate for woodworking for instance. In this context, the motor options we normally have for a hobby construction are either a 1.8 degree stepper motor or a DC servo motor with a 500 line encoder. There are others of course, but these would be the most common.

To give a concrete example, let's consider a 10 micron accuracy at a 50 mm radius. This 10 micron accuracy 'sort of' matches the linear accuracy one might expect from the XYZ axes of a good (hobby) mill. To get that sort of resolution (not accuracy)
requires an angular resolution of approximately 0.0115 degrees (0.69 arc-minutes) at the table. A 1.8 degree stepper would
require a 157:1 reduction to get there (ignoring microstepping for now). That is a lot of reduction. It also places some very tight
limits on allowable backlash, and that is where the trouble lies.

The problems with ordinary gears are fairly well known. First, they have to have some clearance in order to turn, even with the
very best design and manufacture, and this clearance makes for backlash. Second, practical size constraints within a rotary table mean that gear ratios of more than (say) 5:1 are unlikely. To get the required reduction with gears would mean a rather long gear train, and the concensus was that the backlash at the end of such a gear train would be ... bad. To be sure, not everyone agreed.

This does not even consider the problem of getting a perfectly concentric pitch circle on each gear. To be sure one can get very
close, but even 'very close' leaves room for backlash.

A worm gear reduction of about that much is far more possible, but even so there are real problems with backlash. You still need some clearance so the worm can turn against the wheel, although the clearance could be very small. What might be more difficult is (again) the machining error found in the pitch radius of the wheel. Any slight variation in effective pitch radius would mean that the table could be rather tight in some positions and rattling around in others.

It was suggested that you could have a spring-loaded worm driving the wheel. That is true, but there are now two design problems to deal with. First, you have to have the worm really tightly constrained along its axle. That means good miniature bearings.
Second, you have to allow the worm and its support bearings to rotate slightly around some pivot point (or axle) without the
slightest trace of 'rattle' in the rotation bearings. That puts a lot of precision bearings in a very small volume. It's possible, and I did spend some time on such a design, but it is difficult. Well, I thought it was difficult, anyhow.

Of course, since you have a sliding movement between the worm and wheel, there are also problems of surface wear and lubrication. To get lubrication between the worm and wheel you need clearance, even if it is only a few microns. That means some backlash. No lubrication and you get much faster wear. Not easy. Yes, I know most small manual rotary tables use a worm and wheel, but they don't get spun as much as a CNC rotary table would (so there's less wear), and they all come with quite powerful position clamps at the rims, which have to be used when machining anything on them. In effect, they are really indexing tables, not full dynamic rotary tables.

Attachment 270542

Finally, we have toothed belts. The old (or original' designs of 'timing belts' such as L & H and variants thereof had all manner of gaps between the pulley teeth and the lugs on the belt. Yes, that meant you could have backlash, and variable backlash at that.
However, we have come a long way from the early L and H profiles. Modern profiles such as the HTD, AT and GT series are far better, and they have been designed expressly to zero out the backlash. It's all in the belt profile and how it meshes with the pulley teeth you see. Googling will get you lots of vendor explanations, so I won't expand any further.

Attachment 270544

Belt stretch has similarly been improved with steel cores (the string inside the rubber), fibreglass cores, Kevlar cores and carbon fibre cores. The belts are meant to operate under some tension (to seat the belt teeth into the pulleys) and they can deliver a surpisingly large number of kWatts are very high speeds. This means you could string a couple of toothed belts together and
still have negligable backlash on a rotary table. However, you would still have the problem of that 157:1 reduction ratio. A
realistic limit for a single stage of reduction in the confines of a rotary table might be 6:1. Three high-ratio stages would be needed, and you would have to tension every one of them. That's not so easy either.

Despite all these problems, I decided I really needed a rotary table, and it had to be better than a converted manual
indexing table with lots of backlash. OK, make that 'I really WANTED to make one'.

Before closing this chapter, I had better draw a distinction between resolution, accuracy, linearity and repeatability. I will use
angular rotation here as it is most relevant.
* Resolution is the minimum step the hardware is capable of doing. You might for instance have a resolution of 0.001 degrees.
* Accuracy is effectively the difference between what you ask for and what you get. In this context it is a bit complex, so we will
pass on for the moment. Suffice to say that resolution is often (usually) finer than accuracy.
* Linearity is part of accuracy: does a command for a 1 degree rotation give you exactly the same movement when the start angle is 0 degrees compared to 180 degrees for instance? Usually the answer is 'close but not exactly'. An off-centre drive pulley can for instance cause cyclic non-linearity.
* Repeatability is another part of accuracy, but it pulls in the concept of hysteresis. If you tell the system to go to +90 degrees, then have it wander off for a while doing things, then again tell the system to go to 90 degrees, repeatability asks whether it come back to exactly the same position as before. What the target angle is does not matter here; it's whether you get there or not. For instance, if you approach the 90 degree position from two different directions (CW and CCW), hysteresis usually means you do not get to exactly the same position. So repeatability is complex as well.

In the next few Chapters I will discuss my options for bearings, housings, spindles and so on. After that comes a provisional
design, a few major changes (which altered the entire project), further engineering details like motors, and so on. As part of the
series I will go into measured performance specifications.

Cheers

Looking forward to seeing the progress on this!

cheers, Ian

This looks promising. What testing equipment will you use to make sure that you have reached your goal? Your test setup will have to be at least 10 times more accurate than the goal to validate the construction. If you don't have a solid way of testing your design, then how do you know when to stop improving on it?

Quote:

---

What testing equipment will you use to make sure that you have reached your goal?

---

Good question, of course. All will be revealed in the coming chapters.
I have a PhD in physics and I have spent some 40 years doing research into measurement technologies for 'difficult' areas.

Cheers
Roger

good sum up of the old thread on my cnc router it has GT2 belts and pulleys there is almost no backlash it`s about 0.001mm but that might be the stepper

Bearings

This chapter is mainly theory. If a thing goes around for any distance, it has to have a bearing. We have lots of options for bearings, starting with a round bit of hardwood axle stuck through a greased hole in some more hardwood under a cart, and steadily improving from there. We will skip metal and plastic bushes and go on to some common rolling element bearings. Any number of good textbooks or Application Notes might say all this better. The web site for any bearing manufacturer or dealer will have lots of photos and drawings of bearings.

Deep Groove Bearings
These have round balls travelling between an inner and outer ring or race in a deep groove, invented (well, patented) by Philip Vaughan in Wales in 1794. There are a lot of companies now claiming to be the world's best producer of DGBs. I don't know what the world production of DGBs is, but I did read that one German company claims to produce about one hundred million DGBs a year. (Can you imagine what happens when 'they' give each of these bearings an internet connection to report on their reliability?)

The first advantage of DBGs is that they can be cheap. The second advantage is that the balls contact the races over a moderate area (compared to a shallow groove). That improves the load rating a bit. But there is always some clearance and so some rattle (less rattle = more $$). To get tight position tolerance you can put two DGBs up against each other so the balls press on the sides of the track rather than the bottom. This iis called pre-loading, and it works, but it reduces the load rating.

So how many steel balls are made each year? Again, dunno, but Wikipedia claims 'In 2008, the United States produced 5.778 billion bearing balls'. The accuracy of that claim, better than 0.02%, seems debatable to me, but it is still a very big number. I do wonder where they all go.

Angular Contact Bearings
These are similar to DGBs but the surfaces on the races are tilted at an angle to the centre line. This means that putting two of them in opposition actually improves the contact area. They are meant to be used like this with a pre-load. Naturally, this means they are more specialised, there is lower production, and they are a little more expensive. Double ACBs are superior to DGBs for spindles.

Thrust Bearings
These are quite specialised for forces along the axle, and not really suitable for an MYOG rotary table by themselves.

Crossed Roller Bearings
These are a very exotic sort of bearing using rollers rather than balls and usually having quite a large diameter. They can be adjusted for rattle and can give superb performance. Naturally, they are also quite expensive: say 100x the price of a cheap slightly smaller DGB.

Spindle Length

Attachment 270782

A single bearing is great for constraining a spindle at a point, but you need two of them (one at each end) to keep a spindle pointing in the same direction all the time. Since each bearing has a tolerance, the farther apart the two bearings are, the less the spindle can rattle around - up to a point.

Both DGBs and ACBs can be used like this, with one at each end of a spindle. A little design cunning and you can squeeze the bearings together to remove the clearance in the two bearings. A large nut at one end is often used. However, it is not quite that simple: you have to make sure that the bearings are kept square onto the spindle. Some bearing suppliers sell specially machined nuts for this.

A secondary consideration is that the spindle should not flex between the bearings. If you have a 6 mm shaft running through the armature of a motor, the armature had better be very well balanced or the shaft will bend in the middle. Thin shafts are not good. Fortunately, one normally wants a large bore through the middle of a rotary spindle, so a large diameter tube is used. Provided the wall thickness is reasonable, the spindle is 'stiff' and all is well.

That's enough theory for now. The next two chapters will cover a design I put together for a fairly conventional 'spindle with two bearings in a box'. But don't jump to any conclusions just yet.

Glad you are doing this.
Glad you are using real numbers.
Very glad you are making clear the distinctions between resolution, repeatability and accuracy.

someone who use`s the correct words is always a cleaver person but someone who makes it so a nob can understand what is being writing about when they have a PhD is a practical person.

looking forward to the design.

Quote:

---

Originally Posted by RCaffin

Good question, of course. All will be revealed in the coming chapters.
I have a PhD in physics and I have spent some 40 years doing research into measurement technologies for 'difficult' areas.

Cheers
Roger

---

Then I will just sit back and enjoy the ride! Have you heard about the infinite monkey cage?

Quote:

---

the infinite monkey cage?

---

Er ... apart from the BBC show?
Sorry - I don't think my brain has grasped the connection yet. But I don't watch TV, so ?

Cheers
Roger

It's a radio show, so no need to watch TV. Seeing as the show is about popular view on science and physics I thought you might enjoy it. I certainly find it amusing. Helps me get through long production runs and makes me wish I had chosen a career in theoretical physics.

Ah, I see. Not sure the show is available in Australia though. I haven't looked. Could maybe get it via the web. But I am not going to run Internet connections AND Mach3 on the same machine.

A career in theoretical physics? A bit chancy these days. If you think the business world is tough, you haven't seen an academic cat fight. A whole new world of hurt.

I worked in a Gov't research lab with external funding. It was a good place, until the management screwed up so badly that the funding body said stuff it and cancelled the whole deal. So the lab was closed. Mind you, the last few years there had not been pleasant anyhow. Those who can do research do research; those who can't try to get into research management, even though they know nothing about it and usually just make the researchers miserable. When the pressure comes on, they start back-stabbing. After that I went freelance as a research consultant, and earned twice as much per year with better holidays. My wife did ask why I had not done that 5 years earlier - not for the money, but for the better work environment.

Cheers
Roger

Housing and Spindle

This chapter outlines the basic concepts behind a spindle for a rotary table. Any resmblance it might have to the design of a lathe headstock is purely coincidental of course. Actually, I believe some have simply bought a small replacement headstock complete with shaft and bearings and used that.

The basic concept is very simple. You have a stiff tube with a bearing at each end. Keeping in mind the previous discussion about pre-loading bearings, the design will need to be able to (nominally) move one of the bearings along the shaft a tiny bit to get this preload.

Attachment 271130

You have two main options here for the bearings. The first is a single bearing at each end, as shown before. The second is a double bearing (usually) at the front end and a single bearing at the rear end. The former is simple, and if the loads on the bearing are not severe, usually enough. The latter gives you a bigger load rating at the front end, but it is unlikely you would ever need that increased rating in a hobby shop.

In this context, please note that if you want any sort of bore through the spindle (as in a lathe), then the bearings you use here are not going to be small. Skate bearings need not apply. So you will almost automatically get a decent load rating.

Now you come to a rather important question: how do you hold the front end bearing(s)? You could use a standard bearing block or pillow block with a bearing: that has been done. However, safely pre-loading the bearing can be difficult as these are not designed for that. Alternately, you can embed the front bearings in the front plate of the housing somehow.

The two most likely methods are to embed the bearing(s) in the thick front plate, usually by pressing them into a machined cavity (see above drawing), and capturing the bearing(s) in what can be called a 'split block'. The merits of each method depend a bit on the quality of machining which can be achieved.

For instance, if you machine a cavity as shown above and press the bearing into it, you can be pretty sure that the bearing will be square. This is good. However, if the machining is not sufficiently accurate, the fit of the bearing may be a bit loose. This is not so good. You can try to clamp it in place, but clamping is never perfect.

If you make the plate thick enough you can embed two bearings in the hole and pre-load them together. This works for both DBGs and ACBs, and makes for a really rugged front end. It's shown in the diagram above.

On the other hand, if you split the thick front plate cleanly and reassemble it (with bolts) with a thin shim between the two halves, you can then machine a slightly oversize (SLIGHTLY, damn it!) cavity. Then you can remove the shim and bolt the two halves of the plate together around the bearing, and capture the bearing tightly. However, without lots of more complex engineering, keeping the two half-plates perfectly aligned when bolting them around the bearing is difficult. This can upset the alignment of the bearing.

You can also put two bearings into the bore in a thick split block and preload them, but the engineering gets a bit more tricky. The upper and lower halves are really two separate bits of metal.

OK, given that we have the front end looked after, what about the other end of the spindle? That is much simpler: all you have there is a less-critical single bearing. Clamp the bearing in the rear plate and allow for slight movement along the shaft. Well, almost. If you have a preloaded pair at the front end, that statement pretty much holds. However, if you want to use the rear bearing for half of the pre-load, the design does get a bit more complex. You may have to have a very large nut on the spindle to pre-load the arrangement, or even have some sort of mechanism for pushing the rear bearing through the rear plate - which gets rather complex when you want lots of rigidity. On the whole, the nut is usually simpler. Some bearings manufacturers even sell such nuts, with very square faces to them. They are not cheap.

Finally, before rushing to a design, one must decide on the bore through the spindle and the wall thickness for the spindle tube. Once you do this, the OD starts to get rather large rather quickly. In the diagram above there is no bore at all: clearly very unsatisfactory. We will look at this in the next chapter.

Some Practical Details of Construction

In following chapters you will see photos of some machining on my CNC where it looks as though I am using wood as a sacrificial base. Indeed, I am using wood, but it's old and very dry Australian hardwood. This needs some explanation for those used to softwoods, like the rest of the world - but Australian readers familiar with old hardwood can skip the explanation if they want.

The Australian gum tree (eucalyptus species) is widespread across Australia, and up until recently was routinely used for ... well, most everything. The trees were essentially free for the loggers from State Forests. Some wattles (acacia species) were also used for anything. But these were all being taken from original old-growth forests, which were being decimated at a terrible rate, so Australia has largely switched to plantation-grown radiata pine. Huge moans about the cost from the loggers - ignore. But while it is very easy to work green gum and wattle, once they harden it is another matter. There are legendary stories about people trying (futilely) to recycle 50 year old hardwood beams - and most of them are true.

Attachment 271330

In a nutshell, old Australian hardwoods, taken from old sheds, old buildings and old fences, have to be worked with sharp carbide tooling, and some care. We use carbide-tipped saws and even carbide-tipped drills - sharpened of course. Ordinary HSS tends to glow red and blunten. So when I machine up a slab of old hardwood for use as a spoilboard or sacrificial base, it can be almost as hard as mild steel. I resurface it every now and then of course, taking off maybe 0.1 mm. I have access to a fair bit of really old hardwood from my farm. Some of it has a beautiful grain and colour too.

Attachment 271332

Another thing I should mention is the use of steel dowels between the wood base and the metal parts I am machining. Using these, correctly sized of course, allows me to reposition parts to within 0.01 mm. Yes, I have measured that. It helps to use 4 or 6 pins rather than just 2 though, so you get some 'averaging', if you want that 0.1 mm. In addition, by using a suitable symmetrical layout of pins, I can flip parts over and get that same 0.01 mm registration with no further checking. Hard steel pins and deep holes in hard wood (or metal) are the trick. And the front edge of the wood base is always machined smooth, so I can line it up on the mill table with a good large square off one edge.

How well does all this work? Well, for a start, I have to work to get 0.1 mm accuracy with a manual machine, and I have to take especial care to get 0.05 mm. I am not an expert machinist. OK, and my lathe is a small Chinese one I bought years ago (230 mm x 500 mm), and my mill is a basic RF30 from Taiwan. (At least it has an R-8 fitting and a 3-phase motor, not a Morse taper and a single phase motor!) However, I find that my CNC machining centre, an Australian Adept, can be programmed to machine to 0.01 mm quite readily. For this to work I have to use sharp tooling on aluminium alloys, but I can (and do) correct sizes by 5 or 10 microns successfully. 50 microns is a large step for the CNC.

For the rotary table my plan was to construct it out of slabs of Fortal aluminium. These parts have to be held for machining on a sacrificial base, so first I have to make the base. A suitable blank for this sacrificial base is bolted down on the mill table and then the entire top surface is skimmed flat with a fly-cutter - the M6 cap-heads holding it down on the table are recessed to permit this. Then I drill a 6.0 mm hole in the very centre over the central T-slot, and drill some pin holes and holes for self tappers across the base. That prepares the sacrificial base, and the central 6 mm hole is the XY origin.

Once the base has been prepared I can always re-zero the machine after a power-down by the reverse procedure. That is, I align the table to put the mill-head zero in the middle of the table where I want it (over the middle T-slot), then I locate the sacrificial base relative to the mill-head. This is done by using a 6.0 mm pin (usually a 6.00 mm carbide rod with a tapered end) in the spindle collet and sticking it into the 6.0 mm hole in the base while the base is loose on the table. Wood always drills slightly under-size, so the fit is always tight. I then square the base up relative to the mill table by rotating it around the 6 mm pin (with a good square) and bolt it down. The 6 mm hole is then defined as the XY zero. Yes, this works very well - I have done it many times.

Then I place a prepared aluminium blank on the base roughly centred over the XY origin, clamp it down, and drill the same pin holes and self-tapper holes in the metal. If necessary I can also drill a central M6 hole which will let me drop an M6 bolt through to the T-slot underneath. Once that is done I can remove and replace the part using the pins without a worry.

Attachment 271334

You can see the heads of the STs in the photo here, with the pins being in the holes next to the STS. Yes, it sounds a bit complex, and yes, many metal parts do end up with a few surplus pin holes in them, but so what? If I was really worried about appearances I could fill the holes with aluminium Devcon or JB Weld or something similar. In this case i was not concerned.

What's with the self-tappers (STs)? Well, I recess the ST holes in the part so the head of the ST is below the surface. That way I can clamp the part down securely and machine the entire surface flat in one go. By the time I get 8 good-size STs into a hardwood base (already bolted to the mill table), the metal part is not going anywhere. Even 4 STs is enough (with 4 dowel pins) if I machine gently. As I am not doing production work, I go gently.

So the plan was to use my Adept CNC to machine up the parts for the rotary table. The Fortal aluminiun is lovely stuff, and offcuts of quite adequate size were reasonably priced from www.fortal.biz in the USA. A lot cheaper than buying the stuff in Australia, even after postage, not that anyone stocks it here in Australia anyhow. I can recommend the web site. My first design is in the next chapter.

your right about the aus hardwood same as some of the wood over here bang it with a piece of steel and it almost sounds like banging two piece of steel together.

My Original Design with Shaft and Bearings

There had been lots of discussion about backlashfree rotary tables in the thread with that name, so I had some idea of what other people were thinking.

To be sure, many people were happy with a stepper motor on a manual table, but I wasn't. The two big problems with the manual table approach (to my mind) were that the manuals generally relied on clamps or brakes to get stability, and they all have severe backlash. If what you want is an Indexer they might be fine, but not if you want to machine in full 4D.

Attachment 271412

The next step up seemed to be a solid shaft with a bearing at each end, a motor at the back end driving the shaft via a toothed belt, and a chuck at the front. You have probably seen plenty of these in the past, and cheap ones are readily available on eBay etc.

Attachment 271414

Incidentally, for those with more money, there are always a few old commercial rotary tables on eBay - but usually the problem with them is not so much the cost but the weight. If I put this one on my CNC I think the CNC would probably collapse. I love the little OSHA-style warning plaque on the front about not lifting it with a bent back! Frankly, I would be more concerned with dropping it on a finger or a toe!

Attachment 271416

For some of them I am sure the cost of transport would be greater than the eBay price, although to be fair this table includes its own 'tram tracks' for positioning as well. I am not sure it would fit in my workshop though, nor that I could affford the paint job it needed.

Attachment 271418

So in very general terms, my plan was for a shaft with a hole through the middle, supported by two large bearings - probably deep groove bearings. Well, not particularly expensive you see. Just a lot of work.

Attachment 271420

Fair enough, so I bought a box of Fortal offcuts from www.Fortal.biz, including a large bar of Fortal (at the front left) and some slabs of Fortal, some large deep groove bearings, and started machining. The first thing to do was to turn that fat bar into a spindle.

Attachment 271422

The bar was turned into a 60 mm OD rod fairly easily (with an awful lot of swarf). Hey - this stuff machines really nicely! Then I started to wonder how I was going to put a bore through the middle of this, on my small lathe. One thing for sure: I can't chuck a 60 mm rod in the middle. Not quite so easy! This was going to take some thought.

Attachment 271424

The best idea I had was to make up a holder for the bearings I had bought and to attach that holder with one bearing in it to a lathe steady to support the 'free' end. The other end could fit in my 3-jaw chuck. Feasible, but not easy. My lathe would be long enough, but only just after you allow for the tool holder and clearance and so on. Don't even think about the part flying off the lathe by accident.

Attachment 271426

At the same time I was thinking about the reduction needed for this. I would need at least two toothed belts to get at best about 36:1 reduction (2 x 6:1), looking something like this. Mind you, the ratio shown here between small and large is only just bigger than 5:1, so 6:1 would be pushing it. Small and large pulleys are commercially available of course, but really large pulleys mated to really small pulleys are just a shade more expensive. I was a bit alarmed at the quotes I had received for this. What's more, 36:1 was not really enough for what I wanted anyhow.

How to attach the large toothed pulley to the 60 mm spindle was also occupying my mind. Easy enough to do if you are not seriously concerned about concentricity - but I was concerned. I was also concerned about making sure the pulley could not slip on the spindle.

Attachment 271428

The 'obvious' solution would be a Taper-Loc connection, or similar. Good Taper-Loc fittings are available, but any with a 60 mm bore turns out to be rather large - not to mention expensive. The size made them a bit like trying to put a quart into a pint pot in fact. The problem is they are designed to transmit many kilowatts of power in a big industrial motor drive, which is simply not needed here.

All in all, not easy. Maybe not easy enough. Yeah, I'm lazy.

As I mentioned earlier, I had also played with the idea of a large worm and wheel, but getting rid of the inevitable backlash (no matter how small) was going to be very hard. After some time in front of the screen playing with various designs with lots and lots of precision bearings and some complex machining, I gave that one up. Maybe I could do it, but maybe not well enough. Sigh.

Perhaps you get the idea that I was just not really convinced that I had the right design yet. Well, sort of. I have learnt to listen to those little doubts, so I sat back and let things stew for a while. Sometimes the subconscious should be allowed to work without interference. Thinking can just get in the way.

After a while I started to get a vague but interesting idea, partly from reading about various other peoples' efforts here on CNCZone, and so I went for a wander through everyman's tool bazaar (otherwise known as eBay). That's for the next chapter.

Harmonic Drives

As explained towards the end of the last chapter, I was not entirely happy with the simple and obvious design for a Rotary Table based on a spindle and two bearings. So far I had been trying to minimise the cost of the bearings, but then I began to wonder what might come from removing this restriction? Could any increased cost for the bearings be recouped elsewhere? It would not be the first time that has happened to me.

Attachment 271698

Well, Deep Groove Bearings (DBGs) are the cheap end of the market, while Crossed Roller Bearings (CRBs) are undoubtedly the expensive end of the market, at least for my application. So I collected some catalogues for CRBs. (Brief note: Crossed Roller Bearings should not be confused with Cylindrical Roller Bearings. The latter are different. Yeah - I had never heard of the latter either.)

Attachment 271700

Can one use CRBs for a Rotary Table? Clearly the answer is not only yes, but it would seem that CRBs are ideally suited to making Rotary Tables. What's more, they apparently have a very high load-handling capacity, due to all those rollers with their large contact area. Interesting, but how do you mount them? (Nice pictures of CRBs come from a Schaeffler KG catalog on CRBs. Pity they got the force arrows in the left hand pic back to front though.)

Attachment 271702

So I read on through the tech manual. That's where it became clear that handling CRBs was a shade more complex than DBGs. Clamping forces figured high in the discussions about these, along with the need for rigid support. You see, while CRBs are large in diameter, the races are not especially heavy in themselves. They often rely on the surrounding metal for real support. This could be complex - but based on recent experiences, I thought maybe my CNC could handle these tolerances. Somewhat more problematic was whether my wallet could handle them - and anyhow, while very nice, they don't solve the reduction gears problem.

Somewhere along the way I had come across a comment that Harmonic Drives use CRBs - and have a large reduction ratio, and do not have much backlash. There are even a few threads here on CNCZone about using a Harmonic Drive in a Rotary Table. Mind you, the first impression I got was that they also had serious wallet problems, but looking is cheap, and the write-ups were encouraging. So off to the web for more information.

Attachment 271704

How do these things really work? Well, my first impression was ... confused. I am sure their engineering drawing is useful and accurate, but understanding it was another matter. That required some sort of slow interaction.

Attachment 271706

So I sat down and modelled the same unit in a simple 2D drawing package (AutoSketch) from the data in the previous drawing. This was not hard - AutoSketch is a dream to use. (No, AutoDesk did NOT create it: they bought it from some Russians.) Anyhow, the Harmonic Drive gear reducton spline (the core of the unit) is the curved red thing with a pink interior, while the rollers for the CRB are the olive green squares. Obviously the races are part of the structure. The black lines with pink interior are the input shaft, and the green bits are the normal output stage. The blue bit seemed to be the 'base' for the whole thing, although there is a gotcha here which I will cover later. Yes, I could have used a 3D modelling package for this - provided I understood what it was I was modelling. First walk, then fly.

However, Harmonic Drives are expensive things. I wondered: do the Chinese make a clone? Off to eBay and AliBaba to search. The answer was no, but there are one or two Asian dealers who have a supply of ssecond-hand Harmonic Drives stripped out of something or other - decommissioned Japanese robots maybe? Being secondhand they are of course much cheaper, and since they are coming from Asia you can always haggle a bit. To my surprise, I found I could get an SHG-25-50-2UH for just a bit over US$300, including FedEx freight.

The 'SHF' means it is a 'silk hat type' (what the h*ll is that? Do they mean 'top hat'?), the unit is 'size 25', whatever that means (the dimensions are in the book), with a 50:1 reduction. The 2UH bit says it has a hole through the middle and it has integrated output bearings. Anyhow, it has the dimensions as shown in the first drawing. And that size was so close to the 100 mm chuck I had in mind too. How incredibly tempting.

Yes, that's a whole lot more expensive than the bar of Fortal and the two bearings I showed before, and you could buy an Asian rotary table complete with stepper motor for that much money - but two things motivated me. First, the very suitable gearing and the excellent bearings were ALL included in the HD, and second, the precision I could get from an HD made the Asian rotaries look very inferior. If I am going to do this, I thought, let's do it properly.

So, I bought one. It turned up quickly, it was as described, and it was in good condition. The gearing was very smooth. All of a sudden, I could see this project becoming a whole lot simpler. The consequent redesign is in the next chapter.

Quote:

---

Originally Posted by RCaffin

Harmonic Drives

As explained towards the end of the last chapter, I was not entirely happy with the simple and obvious design for a Rotary Table based on a spindle and two bearings. So far I had been trying to minimise the cost of the bearings, but then I began to wonder what might come from removing this restriction? Could any increased cost for the bearings be recouped elsewhere? It would not be the first time that has happened to me.

Attachment 271698

Well, Deep Groove Bearings (DBGs) are the cheap end of the market, while Crossed Roller Bearings (CRBs) are undoubtedly the expensive end of the market, at least for my application. So I collected some catalogues for CRBs. (Brief note: Crossed Roller Bearings should not be confused with Cylindrical Roller Bearings. The latter are different. Yeah - I had never heard of the latter either.)

Attachment 271700

Can one use CRBs for a Rotary Table? Clearly the answer is not only yes, but it would seem that CRBs are ideally suited to making Rotary Tables. What's more, they apparently have a very high load-handling capacity, due to all those rollers with their large contact area. Interesting, but how do you mount them? (Nice pictures of CRBs come from a Schaeffler KG catalog on CRBs. Pity they got the force arrows in the left hand pic back to front though.)

Attachment 271702

So I read on through the tech manual. That's where it became clear that handling CRBs was a shade more complex than DBGs. Clamping forces figured high in the discussions about these, along with the need for rigid support. You see, while CRBs are large in diameter, the races are not especially heavy in themselves. They often rely on the surrounding metal for real support. This could be complex - but based on recent experiences, I thought maybe my CNC could handle these tolerances. Somewhat more problematic was whether my wallet could handle them - and anyhow, while very nice, they don't solve the reduction gears problem.

Somewhere along the way I had come across a comment that Harmonic Drives use CRBs - and have a large reduction ratio, and do not have much backlash. There are even a few threads here on CNCZone about using a Harmonic Drive in a Rotary Table. Mind you, the first impression I got was that they also had serious wallet problems, but looking is cheap, and the write-ups were encouraging. So off to the web for more information.

Attachment 271704

How do these things really work? Well, my first impression was ... confused. I am sure their engineering drawing is useful and accurate, but understanding it was another matter. That required some sort of slow interaction.

Attachment 271706

So I sat down and modelled the same unit in a simple 2D drawing package (AutoSketch) from the data in the previous drawing. This was not hard - AutoSketch is a dream to use. (No, AutoDesk did NOT create it: they bought it from some Russians.) Anyhow, the Harmonic Drive gear reducton spline (the core of the unit) is the curved red thing with a pink interior, while the rollers for the CRB are the olive green squares. Obviously the races are part of the structure. The black lines with pink interior are the input shaft, and the green bits are the normal output stage. The blue bit seemed to be the 'base' for the whole thing, although there is a gotcha here which I will cover later. Yes, I could have used a 3D modelling package for this - provided I understood what it was I was modelling. First walk, then fly.

However, Harmonic Drives are expensive things. I wondered: do the Chinese make a clone? Off to eBay and AliBaba to search. The answer was no, but there are one or two Asian dealers who have a supply of ssecond-hand Harmonic Drives stripped out of something or other - decommissioned Japanese robots maybe? Being secondhand they are of course much cheaper, and since they are coming from Asia you can always haggle a bit. To my surprise, I found I could get an SHG-25-50-2UH for just a bit over US$300, including FedEx freight.

The 'SHF' means it is a 'silk hat type' (what the h*ll is that? Do they mean 'top hat'?), the unit is 'size 25', whatever that means (the dimensions are in the book), with a 50:1 reduction. The 2UH bit says it has a hole through the middle and it has integrated output bearings. Anyhow, it has the dimensions as shown in the first drawing. And that size was so close to the 100 mm chuck I had in mind too. How incredibly tempting.

Yes, that's a whole lot more expensive than the bar of Fortal and the two bearings I showed before, and you could buy an Asian rotary table complete with stepper motor for that much money - but two things motivated me. First, the very suitable gearing and the excellent bearings were ALL included in the HD, and second, the precision I could get from an HD made the Asian rotaries look very inferior. If I am going to do this, I thought, let's do it properly.

So, I bought one. It turned up quickly, it was as described, and it was in good condition. The gearing was very smooth. All of a sudden, I could see this project becoming a whole lot simpler. The consequent redesign is in the next chapter.

---

Roger...

I also am using a Harmonic Drive gearset in my design. The integrated bearings are pretty heavy duty and designed to directly carry a load (they are used in robotic arms for example.) I have a 100:1 unit that I plan to pair with a 200W servomotor I have laying around. It really is amazing that the flex spline has only two less gear teeth than the outer ring gear. I also have a hollow shaft version. Looking to see how you implement yours in your design.

Hi Louie

What size drive? Just curious.

Yes, design coming.

Just a thought: I suspect the 200 W motor will be at least 10x more power than will ever be needed. With 100:1 gearing, the torque from the motor will be multiplied by ~100x, and that ought to be enough to serve as a planer!

Cheers
Roger

the Asian rotaries are not very good at all the one I have I needs to fix it some how but not to sure yet how I will do it.

it has 2 bearings one at each end and 2 thrust bearings as well the bearings are held in by having the nut on the end of the shaft very tight the thrust bearing just sit in there between chuck and bearing and the other end between nut and bearing to pull it down clean and rebuild it is very hard to do as the thrust bearing are bigger than the shaft id and smaller than the od of the bearing.

so finger tight it tap the thrust bearings around to they are in place tighten a bit more check to see if they have moved.

this thread is giving me some good idea`s of how to fix it