PDA

View Full Version : Mill table w/in .003 overall...can I improve that at home?



Bolster
01-06-2010, 08:11 PM
I purchased an X3 mill from Grizzly, and the table is within .003 end-to-end (actually .0027, best as I can tell). Spec from Griz is .002, so they are saying "send it back." What a huge hassle, and loss of cleanup time that would be.

Unfortunately there is a diagonal .001 dip or arc across the table, about 1" wide, as if it's where a polishing machine spun down or something. (If it weren't for that, I'd be at under .002 all the way across.)

Is there any way I can improve flatness of the table, without the heroics of packing up the machine? Here are some options I'm considering:

I have a 9 x 12 grade B stone, flat at .0001 across the surface, can I mount sand paper to it and move it across the table, spending more time on the high areas, and stopping to measure progress every once in awhile?

I have heard of lapping on float glass. I have aluminum oxide powder of various grits from another project. Is that a possiblity?

What does one do in this situation?

Bruce Griffing
01-06-2010, 08:34 PM
Why not just return the table? Easier to pack and ship, etc.

doctor demo
01-06-2010, 08:43 PM
Bolster, don't take this the wrong way but.....
Why bother, it would seem that it would be rarely that anything You are working is going to fall into that ""polishing mark'' and cause trouble. Especialy if You have a vise on the table like I do 99.9% of the time.

Steve

Bolster
01-06-2010, 08:47 PM
Why not just return the table? Easier to pack and ship, etc.

That was my idea, too. They said "No. The table is matched to the machine." (Mumbling 'bll-crp' under my breath.) Doctor Demo, you may be right. It'll be interesting to see what the consensus is.

All the same, the question still stands:

"Can you improve the flatness of a mill table without sending it out for surface grinding?"

doctor demo
01-06-2010, 09:08 PM
"Can you improve the flatness of a mill table without sending it out for surface grinding?"
You could scrape it flat, but I wouldn't want to except for the learning experiance and then maybe not.

Steve

PS. See if Griz will make it right with some other comp.

oldtiffie
01-06-2010, 09:21 PM
Bolster, don't take this the wrong way but.....
Why bother, it would seem that it would be rarely that anything You are working is going to fall into that ""polishing mark'' and cause trouble. Especialy if You have a vise on the table like I do 99.9% of the time.

Steve

I'd agree with that Steve.

Most mills only use the middle third most of the time.

Most jobs will span that "polished defect".

Many vises may be "out" that far as well - even without "moving-jaw ""lift"".

Most work has a realistic tolerance of more than a couple of thou anyway.

If I was setting up for the inferred level of accuracy, I'd be checking, packing and shimming as required.

That "error" could be due in part at least to insufficiently adjusted "X" gib clearances as the table will tend to "drop" at the end that is standing out furthest from the saddle. There are other potential causes.

I am quite appreciative of the disappointment of having that error on a new machine, but sometimes I at least have to "wear and eat" and where necessary make allowances and find and use "work-arounds".

RB211
01-06-2010, 09:26 PM
So the table from end to end is off by less than the width of a human hair?
Tell me what your going to machine where that matters? I have this same mill, and I can tell you that the table should be the last thing on your mind. Why don't you measure how much deflection you get on the head while side milling some steel with a 1/2" 4 flute endmill with a depth of cut greater than .010?
I have a DRO on mine, and I can easily make parts to .001 no problem. 99% of the time I don't need to...
Lets say my table is off by .004 from end to end? Oh damn... I better put all my projects on hold, nothing will ever come out right...

darryl
01-06-2010, 11:27 PM
Hmm. That's not much of an error, but why shouldn't you make an attempt to flatten it a bit more- you're willing to give it a go anyway-

What would I do in that situation? I'd be leery of sending it back for either correction or replacement. It might not get any better- it might even get worse, who knows. I'd pull the table off and sand it just like you were thinking of doing. A few careful passes across the sandpaper taped to the surface plate, and you'll see the picture easily. I would NOT leave the table on and muscle the surface plate over it, if that's what you were thinking of. You'll round it over for sure.

Maybe it would be overkill to do all this- I dunno. I think a .003 thou deviation is quite a bit- if you're trying to square something up, make the sides parallel, etc, 3 thou can mess you up. Why bother with precision ground 1-2-3 blocks, or precision ground parallels, or toolmakers vise, when your reference surface is not flat.

gwilson
01-06-2010, 11:58 PM
I sent a Valumaster mill back to MSC. The table angled out of the horizontal by something like 2 3/4 thousanths from the front to back of the table. It could have been more,I can't recall now. It was at work,but I had fork lifts,and large rolls of packing plastic wrap to re-coccone it. Still a PITA.

I won't be buying another of those!! the table is supposed to be a little higher in the front to allow for cutting pressure,but this one was too far out to suit me.

Bolster
01-07-2010, 12:09 AM
So the table from end to end is off by less than the width of a human hair?

You must have fat hair...mine's thinner than that.

It's the 1" wide dip that gets me most...it's sharply defined, and often where I want to set a square to measure tram. And it also messes with my tram readings when I'm swinging an indicator in a circle, and hit that point; finding a .001 deviation in the table when the diameter of swing on the indicator is 5" makes you think you're out of tram. I find myself cranking the table around to the flatter spots to get better readings. So naturally, I wondered if I could flatten it myself. As you can tell by my original post, I don't think it's a big enough deal to return. OTOH, it is out of spec, according to Grizzly.

Gwilson, are you saying the tables are purposely not flat, in some cases? My table gradually slopes up on either end. You think that's on purpose?

RB211
01-07-2010, 12:46 AM
You must have fat hair...mine's thinner than that.

It's the 1" wide dip that gets me most...it's sharply defined, and often where I want to set a square to measure tram. And it also messes with my tram readings when I'm swinging an indicator in a circle, and hit that point; finding a .001 deviation in the table when the diameter of swing on the indicator is 5" makes you think you're out of tram. I find myself cranking the table around to the flatter spots to get better readings. So naturally, I wondered if I could flatten it myself. As you can tell by my original post, I don't think it's a big enough deal to return. OTOH, it is out of spec, according to Grizzly.

Gwilson, are you saying the tables are purposely not flat, in some cases? My table gradually slopes up on either end. You think that's on purpose?
Do you have the regular X3 or Super X3? Regular X3, tramming is not something your going to do a lot, if at all beyond the initial machine setup.

Paul Alciatore
01-07-2010, 01:18 AM
Some points.

1. If it were mine, I would return it. It is out of spec. Once you start to try to correct it, it will certainly void the warranty.

2. Lapping it by hand will work, but will be a lot of work. A LOT! A 9X12 lap sounds a lot too large: it will be very hard to prevent rounding the edges down. I would use one that is about 1/2 to 1/3 the depth of the table and perhaps 1/5 or less of it's length. Keep it completely on the table while lapping: DO NOT allow it to go off the edges.

3. It will be difficult to control the flatness while lapping. Frequent checks will be needed. See my finish lapping technique below if you insist on doing it this way. But more work will be needed with coarser grades of abrasive than suggested below. A lot more.

4. If you are determined to try to correct it, give some consideration to a large fly cutter or other large diameter cutter. It will need to be balanced, so a two tool design would be best. A piece of bar stock, perhaps 6 or 8 inches long, with a center hole to mount a spindle adapter (R8 or whatever fits) and a slot on each end to hold a square tool bit. Tram the mill as best as possible FITST. This is an important step. Run it as slow as you can. Use very light cuts, half a thousanth or less. You can probably do the whole top in two passes. After perhaps 0.001" or 0.0015" of depth of cut, check the surface and RE-TRAM the mill, again, as best as possible. Again, this is important.

4 (continued). After you get it basically flat, then use a small stone or glass with several grades of fine sandpaper (150 - 400 grit) or abrasive to get a nicer finish. Use oil for a nicer finish and to keep the abrasive dust down and only about 25-50 strokes with each grade on each area of the table. Keep moving the area being lapped around. Use different patterns to go around the table: CW, CCW, figure 8, double 8, etc. So, you will make 10 to 20 circuits of the table to do the 25 strokes. Stroke side to side, not front to back as it will again be easy to round the edges if you do. Check it again after this limited number of strokes. Do not repeat the coarser grades of abrasive. It would be OK to use the finer grades for a second time or even third to get a better finish. I would think you could get well under +/-0.001" with this procedure and it will be a lot less work than lapping alone.

I base my comments about lapping on experience with both telescope mirrors and lapping metal flats in the shop. Rounded edges are very easy to make. In fact, telescope mirror makers RELY on this fact to create a spherical mirror surface. With equal sized mirror and lapping tool, the top one becomes more concave and the bottom one more convex. The degree of curvature is controlled by swapping them top and bottom as needed. This edge rounding is so easy to do that it is completely automatic and great efforts, great control would be needed to keep the surface flat. This is why THREE flats are lapped together in all combinations, not just two.

oldtiffie
01-07-2010, 04:44 AM
Why is all the attention directed only to the top of the mill table.

The top is supposed to be parallel to the underside of the table as that surface slides on the supposedly flat upper surface on the saddle.

It is quite possible that the mill table was accurately machined parallel and flat and with the same dimension over the top of the table and its underside and distorted after manufacture.

If the thickness between the table upper and lower faces is the same and the upper face is not straight/flat then the bottom face is just as out of flat as the top face.

If it were me and I wanted to be sure of where the fault was and what it is I'd have it on similar sized 1-2-3 or similar blocks on a good surface plate and go right over that table with a good surface guage or height guage and a quite good (0.0005"/0.01mm) test dial indicator.

I'd be looking for bending and twisting and out of parallel etc.

"Doing something" for the sake of doing it may well create more problems than it solves - if it solves any at all.

A bit of creative analysis and job planning is required here.

Rushing and haste can lead to repenting at leisure.

Barrington
01-07-2010, 06:25 AM
My table gradually slopes up on either end.I'm guessing you're measuring the apparent flatness using an indicator fixed to the spindle/head ?

Table overhang will tend to lift the area under the spindle at the extremes of travel if the gibs are slack. It might be worth checking whether the measurement changes significantly when you lock them up, or if you press up and down on the overhanging table.

Cheers

.

Black_Moons
01-07-2010, 07:50 AM
I can say I would'nt put up with a 1" wide 'dip' in the table as thats clearly a manufacturing error.
Slight warps across the length of the table is meh, but when you have a small error like that you know it will eventualy bite you, and being a dip you'd have to lap the ENTIRE table down before you leveled it out to that dip.

Laping is very hard too from what I understand if just for one simple fact:
You want to cover everything with even strokes and pressure...
Unfortualy this pertty much means that without having the tool extend off the sides, its gonna unevenly wear more in the middle of the table as you also need to overlap your strokes.
Unfortualy when moving a tool that extends over the sides, the pressure is no longer even anymore, it incresses due to less support and is slanted due to uneven support, rounding off the corners insted of making it flat.

Meaning you need proper refrences to see what you are doing (See 'blue' contact dyes like prussion blue (spelling?)) and a decent amount of skill if you want to improve a surface insted of just lap away at it till the end of time.

Laping is not 'sanding some wood to smooth it for a topcoat'
Laping is 'microscopicly removing hard metal to a achive a surface flatness of better then 0.001" (In this case) with very fine grit sandpaper as you'd never be able to polish out corser sandpaper scratchs without removing several more mils. Not a quick or easy process!

Its likey gonna take many hours to lap that thing and get it back to where it should be, and thats assumeing you allready have all the supplys and skills. how long would it take you to send it back?


Also likey any attempt at laping the table will result in whoever sold you the mill not taking it back.

APEowner
01-07-2010, 09:33 AM
If you can't live with the defect (and I wouldn't) I think you need to send it back. If it were a high spot it might be a different matter but you're talking about taking material of the whole rest of the table to bring it down to the hight of the dip. Even if it's just 0.001" that's alot of metal to remove. Also, as other's have pointed out, Grizzly is willing to stand behind it now. After you try to correct it yourself they're not going to.

vpt
01-07-2010, 11:05 AM
I concur, either live with it for the next however many years you will wn it or send it back.

If you don't 'need' it right now I would take the time to send it back.

Your Old Dog
01-07-2010, 11:15 AM
Most of my milling projects are 4" or less long. If they were 15" long on a 30" table your.003 error is now halved. I wouldn't worry about it unless you are machineing motor heads.

lakeside53
01-07-2010, 12:47 PM
Why is all the attention directed only to the top of the mill table.

The top is supposed to be parallel to the underside of the table as that surface slides on the supposedly flat upper surface on the saddle.

It is quite possible that the mill table was accurately machined parallel and flat and with the same dimension over the top of the table and its underside and distorted after manufacture.

If the thickness between the table upper and lower faces is the same and the upper face is not straight/flat then the bottom face is just as out of flat as the top face.

If it were me and I wanted to be sure of where the fault was and what it is I'd have it on similar sized 1-2-3 or similar blocks on a good surface plate and go right over that table with a good surface guage or height guage and a quite good (0.0005"/0.01mm) test dial indicator.

I'd be looking for bending and twisting and out of parallel etc.

"Doing something" for the sake of doing it may well create more problems than it solves - if it solves any at all.

A bit of creative analysis and job planning is required here.

Rushing and haste can lead to repenting at leisure.


What he said....

Bolster
01-08-2010, 01:28 PM
I'm beginning to lean toward the 'send it back' side of the equation. I haven't even told you how out of tram it was when it arrived (column leaning right), or how much trouble it's been to get it straight, and how it won't hold tram once I get it there. Right now I've got a .007 shim under the right side column bolts, which was OK 2 weeks ago, and is now showing not enough again. Maybe I just got a crooked mill.

The other explanation is that I'm too inept to tram the mill correctly. But all the previous measuring and tramming was done under the supervision of not one, but THREE machinist professors at the local JC, all of whom are experienced.

darryl
01-08-2010, 01:43 PM
Sounds like the whole machine should go back. It shouldn't be going out of tram as easily as it sounds like it is. If there is a reason for that, you'd know it- too heavy of a cut, heavy chattering, a crash- but what if you can't find the reason for it - something would not be right about the machine. Maybe it was built on a friday, with parts that were also made on a friday :)

The only other reason I could see for it going out of tram for no reason is if there was cosmoline trapped in the ways and it's still squishing out. But you said you spent considerable time cleaning it, so I'm guessing you would have taken care of all that.

Bolster
01-08-2010, 04:54 PM
Alas, the one thing I didn't do was pull the column COMPLETELY and clean the mating surfaces. I just lifted it enough to get the shims in. If it's not too much effort I may try doing that. Thanks for all the helpful advice from this forum so far.

form_change
01-10-2010, 04:23 AM
I buy second hand machines and live with inaccuracies or fix them - a new machine should arrive to spec. or you are basically paying a new price for a "not new" machine.
I would recommend Connell's book on machine tool reconditioning as a worthwhile read for anyone contemplating doing restoration work on a machine as it explains what you are checking for and how it can be improved.
If you want a restoration project, then keep the machine but as another respondant said, as soon as you've touched it there goes the warranty. I don't know your particular circumstances but there would be any number of machines that would be better to spend your time on. For example, say you spent 100 hours getting the machine up to the mark. All you will have done is brought it up to a condition that someone else can buy from the manufacturer for the same price as you originally paid. On the other hand, a secondhand machine of pedigree is worth spending the time on as at the end of the day someone else will probably pay what you originally paid plus reward you for the effort you have made to improve things (provided you did it right). I replaced a Tiawanese Drill/Mill for precisely that reason - no matter what I did to it, I would not be able to get more than the price of a new machine. My Tom Senior on the other hand I would hope to at least recover my costs. Paradoxically, it's not for sale because after doing the necessary work it suits just fine...

Michael

Black_Moons
01-10-2010, 04:34 AM
Wait what...
'I replaced a Tiawanese Drill/Mill for precisely that reason - no matter what I did to it, I would not be able to get more than the price of a new machine.'

You replaced (sold) it because no matter what you did you would be.. losing money by selling it?

form_change
01-10-2010, 05:14 AM
No, I sold it for what I bought it for without doing anything to it - I was thinking about poly V conversions for lower speed, providing Z axis travel repeatability (round column machine), adaptor plates for mounting things like RT's and so on. At the end of the thought process I realised I could spend a lot of time and money 'fixing it' just to get it to a stage where I really wanted it in the first place. I also figured that even if I did all this there was a chance that I would not be happy with it later anyway, so I was better off getting starting a fresh and find something with what I should have gotten if I'd thought about things properly.
(A friend in real estate once said that if someone talks about putting on extensions when they buy a house from him, he makes a note to call back in around 18 months because they will probably want to sell to buy what they really want - same sort of thing)