WM14 Mill spindle issue/adjustment

[picclock]

Hi
I've had this Warco WM14 mill for some time and it has had this issue from new. The spindle axial rotation is not in alignment with the spindle feed axis, and I really would like it to be 

As a better explanation, the spindle is accurately aligned to the bed of the mill, such that if one attaches a dial gauge to the spindle and rotates it such that the gauge reads the table, the spindle is pretty accurate. This remains true regardless of the feed position. However, the centre of rotation changes in the Y plane depending on the vertical spindle feed setting. If a right angle plate is placed on the table and a dial gauge attached to the spindle an offset of 6 thou Y per inch of vertical movement occurs. The error is about 0.35 degrees and equates to approx 0.25mm over the total 50mm of spindle feed. Picture attached to hopefully clarify my poorly worded description.

I am seeking advice on how to disassemble the head if anyone on the forum has one of these mills, and how best to effect adjustments to cure the issue.

Best Regards

picclock

[Arnak]

Hi Picclock,

If the offset increases evenly when you lower the spindle I would assume that that the column is out of alignment in the y direction.

Have you tried packing the base of the column in an attempt to tilt the y direction back to the vertical?

If the misalignment is uneven could the bearings be worn?

You could also contact Warco as in the past I have found their technical support to be very helpful.

Martin

[chipenter]

One way to test is to remove the down feed pinion and key , and rotate the spindle sleeve and test at 90 180 and 270 .

[RussellT]

I have the Chester version of the WM14.

This sounds very peculiar.  If I understand correctly you are talking about lowering the quill using the side handle and finding that it's motion is not parallel to the axis of rotation.

Does locking the quill feed make any difference to the measurements?  Is there any detectable play between quill and the main casting?  Is there any detectable play between the spindle and the quill?  Is the motion smooth throughout the travel?

I have disassembled mine but I can't remember exactly how it comes apart.  If I recall correctly the only fiddly bit was getting the bearings out.

Russell

[picclock]

@ Arnak
The column is true. Checked by simply raising and lowering the head with a dti and ground precision right angle block. Using both sides of the block I detected an error ~0.5 thou - better than I expected. I could change the column to to correct for this but then the spindle rotation would no longer be at perpendicular to the bed.

@RusselT
>> If I understand correctly you are talking about lowering the quill using the side handle and finding that it's motion is not parallel to the axis of rotation.

That's a much better way of describing it. The mechanics are good, smooth operation, no noticeable play, smooth movement. Locking the quill makes no difference but does introduce a side offset of a thou or two max.
In the past I've bodged it by packing the mill vice at an angle when I needed truly vertical holes, which to be fair is not that often. But its one of those things that keeps on niggling at me over time and I think I need to sort it as its become really irritating. Its a good little mill otherwise, perhaps a bit underpowered, and could always do with more travel.

If no other info is forthcoming on the madmodder forum I will get in touch with Warco, though it is well out of date for warranty purposes ~ 6years old.

Many Thanks

Best Regards

picclock

[Pete.]

Something is screwy here. If I am understanding correctly, the spindle trams correct to the table and winding the head up and down the column dovetail keeps tram but feeding the quill down shifts the spindle axis sideways. The only way you can achieve that error is if the OD of the quill is not turned concentric to the bore.

[chipenter]

On my WM15 the bottom of the quill unscrews useing a pin spanner , the top has two left hand locking rings behind the warco badge , then tap on the draw bar and the spindle dropes out .

[Pete.]

Something is screwy here. If I am understanding correctly, the spindle trams correct to the table and winding the head up and down the column dovetail keeps tram but feeding the quill down shifts the spindle axis sideways. The only way you can achieve that error is if the OD of the quill is not turned concentric to the bore.

Sorry for quoting myself but I've been thinking about this some more and it can't be a problem with the quill concentricity. For the spindle to remain in tram when the quill ID is non-concentric to the OD, the head casting bore would have to be machined the same amount non-parallel to the mount as the quill bore is non-concentric to the OD. What are the odds of that being true?

[RussellT]

I'm with Pete here.  There's something odd.  The odds of two cancelling errors seem small - but perhaps someone has adjusted one out somehow.

Can you attach a dial indicator to the quill and clock it against the column.  That should check whether the quill is moving correctly.  If it's not then loose Y axis gibs or a loose head swivel bolt or muck between the head and head mounting might be a cause.

Russell

[picclock]

@Pete
The way I figure it, and it has taken some time for me to work it out. The spindle bearings are good, but the whole spindle assembly is moved vertically by the handle on the side. From the measurements the spindle is not concentric with the part that holds it. I have checked it with a dti against the column before and I think it shows the same error, but I will double check it again today.

Its like the bearing holders of the spindle are not concentric with the part that houses it.

After all this time it will be interesting to see what the problem actually is.

Best Regards

picclock

Edit: added pictures which show ~13thou difference over 50mm quill travel against column.

[RussellT]

13 thou is more or less the same as 0.25mm as mentioned in your first post.

Are the z axis gibs tight and both swivel bolts tight.  Is there anything in between the swivel mounting faces?.

The surprising thing about this is that the axis of rotation is right.

Russell

[picclock]

@RusselT
I've known about this issue for a number of years, and have set up the spindle axis of rotation to be true to the table. I have put a 5 thou ( I think - was a long time ago) feeler gauge shim at the top of the head to correct a small y error which was present without it. However, this corrects a 2thou error across the table width and has no bearing on the gross error or quill issue. At the end of the day if I need to get get the spindle/quill concentricity issue sorted. I will post pictures of the spindle accuracy to the bed for any 'unbelievers ??' when I have the machine in a state to do so. Even if the gibs were lose, which they are not, I always lock the Z axis when working and that jams the gibs into the column. Swivel bolts would only affect the x axis not the Y, and they are tight. I used to tighten them so much I stripped the lower locking bolt thread and had to make a new one - I think I modified a set screw for the purpose. Either way all very tight.

Best Regards

picclock

[Arnak]

Hi Picclock,

I would talk to Warco as they are very helpful and may have come across this problem before.

Martin

[modeldozer]

hi Picclock,

Had the same problem with my Wiess equevelant of the ML18 and the problem was with the back face of the head where it mounts to the slide, where it moves to tilt the head.  Had to take off the head and scrape the surface true to the spinle axis. Not a very easy opperation as the head had to come off various times to scrape, refit and test and repeat.

Cheers.
Abraham

[RussellT]

Hi Picclock

I was asking about the swivel bolts and gibs because if they were slack then the head could droop and then lowering the quill would tend to move it toward the column.

I assume your shim is in the gap between the swivel faces.  That does explain this a bit.  I couldn't understand how the axis of rotation could be right.

If it were mine, I would take the shim out and see what the quill did then.  I would first want to get the quill moving vertically and then tackle the axis of rotation which I think will need the spindle stripping.

Russell

[picclock]

Hi RusselT
You can shim the head so that the quill vertical movement is true, but it requires about 1.5mm of packing at the top of the head and obviously the cutter spindle is far from vertical. With the spindle true shimming of a couple of thou is all that's required.

Have been in touch with Warco re head dis assembly. Their advice is to remove head from column (been there before), remove motor and gear plate, the drive the spindle out with a hammer. Not sure how this is going to work out but I will get to it asap. Main problem is that I want to use it on a fairly regular basis so I'm bound to need it when its in a pile of bits. 

Best Regards

picclock

[RussellT]

I expect when you have it in pieces you'll find it's an easy solution - if you had a milling machine.

Good luck.

Russell

[Joules]

Any members on Google Maps nearby that could offer you mill support whilst yours is undergoing work ?