Seig mill table not flat

[Dave ShedmanParticipant @daveshedman]

I have a Seig SX 2.7L and the table is not flat across its width. I put a DTI on the milling head dovetail and traversed the Y front to back (160mm); the max deviation I saw was 0.12mm high, the very front and rear of the table both equalled zero.

It would appear therefore that the table has a 'hump' in it and this is consistent along the length (700mm).

I was thinking of removing the table and getting it skimmed flat at an automotive shop. Is this a good idea or could it relieve stress making the hump even worse?

I have had the table off the machine before and the hump was still evident after several weeks at rest so the hump cannot be attributed to the machine set-up.

Any advice much appreciated as always.

[Dave ShedmanParticipant @daveshedman]

[Anonymous]

It is tricky to make measurements on a machine tool that measure one thing, and one thing only. The measurement as described doesn't prove that the table has a hump. If the table surface does have a hump as described then it ought to be easily detectable with a steel rule.

Andrew

[Dave ShedmanParticipant @daveshedman]

Hi Andrew,

That's actually how I first noticed the hump – I found I could rock a steel rule across the table width. Appreciate your caution regarding measurements; I used the most 'solid' method I could to mount the stylus indicator and used this to measure the table top 'undulations'.

Like I said, when I removed the table for some time I could still detect the hump with the rule.

[not done it yetParticipant @notdoneityet]

Posted by Dave Shedman on 17/04/2023 12:45:14:

Hi Andrew,

That's actually how I first noticed the hump – I found I could rock a steel rule across the table width. Appreciate your caution regarding measurements; I used the most 'solid' method I could to mount the stylus indicator and used this to measure the table top 'undulations'.

Like I said, when I removed the table for some time I could still detect the hump with the rule.

Agree. A straight edge edge to centre should show a relatively huge gap, between straight edge and table, at the other edge of the table

[Dave ShedmanParticipant @daveshedman]

That's what I've found. The main question is; can this be suitably corrected with a 'skim' taken off at an auto garage as is done to cylinder heads?

[old martParticipant @oldmart]

I would look for a small machine shop in your area and give them the dimensions of the bed. They would be able to assess the problem and even a milled surface would be plenty good enough for a home shop machine. They will have the means to sit it on a mill bed. I milled the bed of the Tom Senior mill on the drill mill at the museum to about 1.5mm to remove 90% of the surface damage. I drilled through the bed in about 4 places at the bottom of the centre tee slot for hold down screws and it sat on four 124 blocks as the bottom wasn't flat. The milling was done in about 3 passes on the width and 2 on the length as the biggest cutter I had was 80mm. Afterwards the top was rubbed down on the surface table which had wet and dry 120 paper on it, taking several weeks altogether and it is quite good now.

Try your rule on the bearing surfaces on the underside of the bed also. If the rule has two edges, check with both sides, if the rule is good, it is unlikely to be out of straight. When I was working, I got the inspection to check one of my new Rabone Chesterman 12" steel rules and it was not good enough to be classlfied as a straight edge. The rule part of my Mitotoyo combination set was classified as a straight edge and along with the attachments was the only combi set in the calibration system as it was used for checking certain aircraft parts.

Edited By old mart on 17/04/2023 14:53:27

[John HaineParticipant @johnhaine32865]

Is it under warranty?

[Dave ShedmanParticipant @daveshedman]

I doubt if the machine is still under warranty, I have had it for about 3years now (from new) and am only just getting to use it!

[Dave HalfordParticipant @davehalford22513]

Posted by old mart on 17/04/2023 14:43:47:

I would look for a small machine shop in your area and give them the dimensions of the bed. They would be able to assess the problem and even a milled surface would be plenty good enough for a home shop machine. They will have the means to sit it on a mill bed.

As above or some who does engine reboring which ever is nearer.

We don't have many or even any garages that could carry that out in the UK.

[Henry BrownParticipant @henrybrown95529]

I'd be asking how the machine shop of choice was going to mount and clamp the table, this will be critical, you don't want to make it worse than it is by using someone who isn't familiar with this type of work.

[John HaineParticipant @johnhaine32865]

How wide is the "hump"? If it's the central "lands" between the tee slots, could you mill most of it on the machine itself with a face cutter?

[Anonymous]

It's not as simple as just skimming the surface. Not only does the top of the table need to be flat in two directions, it also needs to be parallel to the ways in two directions. So I wouldn't go to an automotive shop. I'd look for a machine reconditioner who understands machine tools.

Andrew

[Neil WyattModerator @neilwyatt]

It would best be done on a surface grinder.

Neil

[old martParticipant @oldmart]

For that relatively small error, a surface grinder would be best, if anyone has one big enough to run over 700mm.

[John MCParticipant @johnmc39344]

I've helped a friend sort out exactly the same problem on a Seig X3. An error of ~0.15mm across the table (Y direction).

At first we thought the table had distorted but checking the ways, which were flat, suggested otherwise, possibly a machining error?

A local machine shop confirmed the hump was on work side of the table. The machining to correct was done on a surface grinder. Two steel bars were placed on the magnetic chuck and ground flat. The ways were then placed on the ground bars, there was enough "grip" from the magnets to hold the table for grinding.

[Dave ShedmanParticipant @daveshedman]

Thankyou for all the advice, a local machine shop looks like the way forward.

Would anyone be able to recommend one in the Nottingham/Derby area?

Many thanks

[Dave ShedmanParticipant @daveshedman]

I spoke to a machine shop engineer who asked me to check the gib adjustment.

I found to my surprise that the largest error at the top of the 'hump' (across the y axis) increased by about 0.02mm when the gib/X axis lock lever was applied.

I then remove the X axis gib and the error immediately decreased from 0.15 down to 0.05. (The gib is a tapered type with a lateral adjusting screw at each end).

What I fail to understand is how the gib seems to be causing the table to bow upward. I assume that if excessive pressure was applied by the gib and/or the lock lever, this would have the opposite effect; in that the force applied would in effect try to 'spread apart' the tables own dovetails and hence force the table surface to distort in a concave manner?

[SillyOldDufferModerator @sillyoldduffer]

Posted by Dave Shedman on 20/04/2023 09:07:48:

I spoke to a machine shop engineer who asked me to check the gib adjustment.

I found to my surprise that the largest error at the top of the 'hump' (across the y axis) increased by about 0.02mm when the gib/X axis lock lever was applied.

I then remove the X axis gib and the error immediately decreased from 0.15 down to 0.05. (The gib is a tapered type with a lateral adjusting screw at each end).

What I fail to understand is how the gib seems to be causing the table to bow upward. I assume that if excessive pressure was applied by the gib and/or the lock lever, this would have the opposite effect; in that the force applied would in effect try to 'spread apart' the tables own dovetails and hence force the table surface to distort in a concave manner?

Umm, I can't explain it either! More likely the table is lifting when the gib is tightened rather than bending. If so, Andrew's comment might explain all: ' It is tricky to make measurements on a machine tool that measure one thing, and one thing only. The measurement as described doesn't prove that the table has a hump. '

High precision measurements are seriously difficult to do correctly. Part of the problem is they look straightforward, but aren't. Below about 0.05mm tiny errors in the setup start dominating the result, making it essential to get the technique right. Experienced metrologists get it wrong, so always approach machine measurements with caution. Having the same fault reappear after paying to have the table re-ground would be depressing.

My advice is to use measurement to help diagnose real faults, not to start with tricky-to-get-right measurements. They come with a high risk of incorrectly concluding that the machine is wonky. In contrast, there's nothing like cutting metal to identify genuine faults, and although 3 years old, this mill hasn't done much cutting yet. Might be a wrong 'un, but humped beds are almost unknown. Is there any evidence it doesn't cut properly?

My preferred approach is to machine a test part and then measure the part, not the machine. Much easier to measure small workpieces accurately than big machines. If the workpiece is wrong, then investigate the machine, bearing in mind that what's wrong with the workpiece provides strong clues.

Don't ignore the other main causes of trouble! In my workshop there are three: the tools, the materials, and the operator. After nearly 10 years self-taught metal-bashing, the main problem in my workshop is still me…

Dave

[Clive Brown 1Participant @clivebrown1]

I can't help with an explanation but I'd echo Dave's advice. If a component as massive as a milling machine table is really being distorted by adjustment of the gibs then you've got a major problem, virtually impossible IMHO.. My advice is not to do anything rash, such as skimming / grinding. Perhaps take the table off the machine and see if you can get the use of a decent sized surface plate to make some careful measurements

[Dave ShedmanParticipant @daveshedman]

I will not be looking to carry out any grinding/machining of the table surface now!

I have done a few sketches and am wondering if the gib is not in full contact when viewed in section.

If only the top part edges of the gib were in contact this could have the potential to force apart the dovetail in a direction that may and cause the convex bow.

I also note that the table casting does seem rather thin in section when viewed from underneath.

I will get it on my surface plate and check it against the ways. I will also look at the contact surfaces of the gib with blue.

I am considering bolting down a slab of steel and flycutting this to provide a level playing field to clamp work onto.

I also wonder if fitting a parallel gib and drilling in some tapped holes for several adjustment screws may give a better gib solution than the tapered type?

[Nicholas FarrParticipant @nicholasfarr14254]

Hi Dave Shedman, If the gib is tapered across its width and only contacting the top of the dovetails, then I suppose that could deform your table they way you say. As far as replacing it with a parallel gib, that might not work very well as at least one one the dovetails will also be tapered along its length with respect to the others. It would be best to check that all the dovetails have been cut at the correct angle, as that may cause the same effect if one or more are wrong.

Regards Nick.

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Dave Shedman on 20/04/2023 09:07:48:

I spoke to a machine shop engineer who asked me to check the gib adjustment.

I found to my surprise that the largest error at the top of the 'hump' (across the y axis) increased by about 0.02mm when the gib/X axis lock lever was applied.

I then remove the X axis gib and the error immediately decreased from 0.15 down to 0.05. (The gib is a tapered type with a lateral adjusting screw at each end).

What I fail to understand is […]

.

Sounds like you spoke to a good man, Dave

I think you are already on the way to doing this, but my advice would be to remove the table and check it on a surface plate … then either rectify the errors, or just map them for now.

Then reassemble and find out exactly what is happening with the gib adjustment and locking.

It certainly sounds more like a case of ‘bad fitting’ than a serious machining error

The devil in making machine tools is in that fitting work, and that’s where the pennies tend to get saved in tools machines destined for the hobbyist.

MichaelG.

Edited By Michael Gilligan on 20/04/2023 11:27:23

[JasonBModerator @jasonb]

Posted by Dave Shedman on 20/04/2023 09:07:48:

 

What I fail to understand is how the gib seems to be causing the table to bow upward. I assume that if excessive pressure was applied by the gib and/or the lock lever, this would have the opposite effect; in that the force applied would in effect try to 'spread apart' the tables own dovetails and hence force the table surface to distort in a concave manner?

 

An overtight gib on the x-axis will squeeze the two dovetails of the table together on the underside and therefore cause the top surface to bow

Edited By JasonB on 20/04/2023 13:23:45

[bernard towersParticipant @bernardtowers37738]

Sounds like a surface grinder job to me as the casting has aged and moved after it has been machined. Too much machining and in too much of a hurry!

[Author]

Posts