MMD-45 Mill Tight Spindle

[cba_melbourne]

....... I only backed off the bearing adjustment nut slightly from contact. There is limited options with the locking tabs. The next tab was too far away for my liking.........

Machine tool spindles need a very firm preload. If you adjust preload like a car wheel (slightly from contact as you say) you will get chatter and poor surface finish, and you will find it hard or near impossible to hold close tolerances.

Also, tabs are bad news to adjust preload. I assume you are talking a tab washer locking onto a hex nut? That would only let you fix the adjuster nut every 60 degrees? That is unacceptable, the difference between a too loose setting and a too tight setting is more like 20 degrees - sure it depends how fine the thread is, but they usually use a fine pitch like 1mm or 1.25mm or 1.5mm for the preload adjuster nut. It is usually a castellated nut (a nut that is tightened wit a hook-spanner). And it is common practice to have a grub screw radially threaded into this castellated nut to lock the nut at any random position, with a brass or copper plug between grub screw and spindle to protect the spindle thread from being dented. A bit less common is to use two nuts tightened against each other to lock.

Some machine makers give a torque figure in the maintenance manual for how much to tighten the lubricated preload adjuster nut. But this is not common with smaller machine tools and certainly not with hobby lathes or mills. Adjustment at the factories is more an empiric instruction like tighten to contact, then add 1/8 of a turn. Or on a lathe with standard chuck installed "spin as fast as possible by hand, and the chuck should spin for between 3/4 and 1-1/4 revolution". The latter method depends a lot on the grease used, and requires the spindle to be first run-in for a few minutes to distribute the grease first.

There is a more accurate method to adjust the preload. You observe how the spindle warms up. In the case of a mill with taper roller bearings and grease lubrication, the quill should become warm to tough after about 10-15 minutes running at maximum rpm. But themperature should not rise any further, regardless if running at max rpm for several hours. If it does not get about 40-50 degrees after 15 minutes, the preload is insufficient. But if it ever gets hot to touch, preload is excessive. You will find out that after about one or two hours running, you can slightly increase preload. That is because a bearing is invariably overfilled with grease - even if you think you only filled it to 40% as specified. It is in the nature of most people to always err on the generous side with grease. The excess grease causes excessive heating by grease churning. After about an hour running most excess grease is pushed up the quill and/or through the bottom seal.

Is it worth the time and trouble to adjust the preload by temperature rise? It is up to you and how you plan to use your new mill. I think yes it is worth it, even if it is not expensive precision bearings. Others will say I am overthinking it again, and will be working "good enough" if treated like a trailer wheel bearing.

[Oldneweng]

Also, tabs are bad news to adjust preload. I assume you are talking a tab washer locking onto a hex nut? That would only let you fix the adjuster nut every 60 degrees? That is unacceptable, the difference between a too loose setting and a too tight setting is more like 20 degrees

The nut is a hook spanner adjusted nut. Every 60 deg? Maybe if you are lucky. One of the tabs is bent around in the plane of the washer. Looks like manufacturing defect.

It is usually a castellated nut (a nut that is tightened wit a hook-spanner).

No it is not a castellated nut. It is a bearing adjusting nut. A castellated nut has the slots going right thru and is secured with a pin of some type. The "castellated" section is usually of a reduced diameter. The reason I know this is that one of the web pages I was reading, to work out how to fix the mill originally had the same admonition when someone used the term "castellated nut".

I had to put that in. I don't care what its called. I know what you mean.

And it is common practice to have a grub screw radially threaded into this castellated nut to lock the nut at any random position, with a brass or copper plug between grub screw and spindle to protect the spindle thread from being dented.

Funny you should mention this. It was my intention to investigate this avenue.

Regarding the preload, the only mention I have seen up to now regarding adjustment of the bearings came from one of those web pages. This method was tighten to contact and then back off 1/8 of a turn. I realised this was wrong. Maybe they meant tighten 1/8 of a turn?

I will play around with preload later on, but I am a bit nervous at over tightening for obvious reasons.

Thanks for your assistance

Dean

[Michael G]

... it is common practice to have a grub screw radially threaded into this castellated nut to lock the nut at any random position, with a brass or copper plug between grub screw and spindle to protect the spindle thread from being dented.

Never seen that one. This is what I commonly see for locking up preload on bearings -
Lock-Washers-Lock-Nuts.jpg
You have 17x4 locking opportunities, so a little over 5 degree increments.

Michael

[BaronJ]

Hi Dean,
Without a torque wrench setting preloads is very much subjective. Does the bearing manufacturer have any notes in this regard ? Run the machine for a while and try to get a feel for how the spindle behaves. You can always adjust for more if you get any chatter. The spindle on my machine runs slightly warm and since replacing the plastic gears is much easier to turn and the motor gets nowhere near as hot as it used to. The surface finish is much better as well.

[cba_melbourne]

Never seen that one. This is what I commonly see for locking up preload on bearings -
Lock-Washers-Lock-Nuts.jpg
You have 17x4 locking opportunities, so a little over 5 degree increments.

Michael

Hi Michael. I have never seen one of those washers used on any hobby machine. Probably because it requires a slot in the threaded portion of the spindle, which would add to manufacturing cost. Hobbymachines need to be low cost, or nobody can afford to buy. The whole 300kg mill in question here, costs less than the naked spindle bearings of a Bridgeport or about $1700.

Edit: on hobby lathe spindles you do not see this type of washer, because the threaded part has the thinnest wall thickness anywhere along the spindle. An axial slot here would seriously weaken the spindle. Whilst for commercial reasons, every manufacturer seeks to maximize the through-spindle bore diameter. Buyers often base their machine choice on this diameter.

[cba_melbourne]

Hi Dean,
Without a torque wrench setting preloads is very much subjective. Does the bearing manufacturer have any notes in this regard ? Run the machine for a while and try to get a feel for how the spindle behaves. You can always adjust for more if you get any chatter. The spindle on my machine runs slightly warm and since replacing the plastic gears is much easier to turn and the motor gets nowhere near as hot as it used to. The surface finish is much better as well.

I am not aware of any hobby machine, lathe or mill, costing below $10k, that would use a torque wrench for preload setting (I have not seen them all though). I can see two main reasons for this:
- to use a torque wrench, the bearing journal on the spindle would have to be very finely ground to the same finish and the same tolerance band as the bearing cone itself is made to. This means just a few microns diameter tolerance for a tight sliding fit to the cone. To manufacture such an accurate spindle would alone add some $500 to the cost of a machine and price it out of the market. If you look at these Chinese hobbylathe/mills, the journals are often not ground at all, just turned. And if you see them ground, its a very very coarse finish. They make the journals on purpose oversize, because they can not hold the tight tolerance that a sliding fit would need. But without a sliding fit, any torque wrench would simply max out to get the cone moving on the spindle in jerks. No precise torqe adjustment is feasible.
- the same is true for the thread on both the spindle and the adjuster nut, these would have to be of tight tolerance and good finish, otherwise the uncontrollable friction would render any torque specification inaccurate. If you look at these on a Chinese hobbymachine spindle, you understand instantly what I mean.

There must be more reasons why even more upmarket hobbymachines (Emco maier, waabeco, myford...) do not use torque wrench specifications to adjust preload. There are probably just too many variables with such small machines.

[Oldneweng]

Michael's picture is almost right. The lock washer only has 6 tabs on the outside. 3 on each side, one of which is the bent one mentioned. The central tab locates in one of the spline slots so no extra machining is involved except one of these slots is slightly longer, presumably to fit the tab. I found the washer fitted better facing the other way anyway which meant it could fit in any slot. The tab is bent at 90 degrees so it sits along the bottom surface of the slot.

Spindle Bearing Lock Washer.jpg

This one only allows for quite large increments.

This mill appears to be different in some regards to the run of the mill () mill. It weighs in the vicinity of 380 kg and the bearing journals seemed to be a ground finish. Not only that, the spindle shaft between bearing journals was ground.

Spindle Bearings and Quill.jpgSorry about the low res. This picture came from the net, but it almost shows the finish.

The bottom section of the spindle was hardened where the slots are to grip the spindle. I did not investigate how far up this hardening went. The reason I know it was hardened is that the bottom face of the spindle was very very rough and not level. I cleaned it up in the lathe to allow it to sit upright to press the bearing on.

Dean

[Pete F]

Hi Dean,
Without a torque wrench setting preloads is very much subjective. Does the bearing manufacturer have any notes in this regard ? Run the machine for a while and try to get a feel for how the spindle behaves. You can always adjust for more if you get any chatter. The spindle on my machine runs slightly warm and since replacing the plastic gears is much easier to turn and the motor gets nowhere near as hot as it used to. The surface finish is much better as well.

I can't say I've ever seen torque of the pre-load nut quoted, either by an equipment manufacturer or bearing manufacturer. What they will sometimes quote is the torque to begin to move the spindle from rest. An easy way to set up to do this is to wind some string with a weight on it around a pulley etc on the spindle and hang a weight on it. Nice theory at least Never done it that way and never will. Oh, and obviously the spindle needs to be horizontal unless you use a spring scale. All in all, rather impractical, although it is indeed the "correct" ie most accurate, way to set spindle pre-load.

In practice what I do is adjust the pre-load so I can feel a little "stiction" when I try to turn the spindle by hand, but once rotating there isn't any undue resistance. It's one of those things that is hard to describe, but when you feel it you'll know. It's very subtle and the differences between properly adjusted spindle bearings and ones improperly adjusted is only a few degrees on the adjustment nut. I'm surprised some of these Chinese manufacturers are using such agricultural settings for pre-load, the likelihood of falling on a point where the tab was just right to lock it would be about the same as winning a chook lotto at the local RSL (ie it happens but don't count on it!). By default, you'd pretty much guaranteed to have to run the spindle at lower than ideal pre-load. If it were my machine I'd ditch the lock washer and cross drill the locking nut for a grub screw and brass plug.

[Michael G]

Michael's picture is almost right.

I prefer to think that your machine maker almost has it right

Michael

[Oldneweng]

I prefer to think that your machine maker almost has it right

Michael

I will pay that one Michael. It is more accurate?

Dean