[dugpits]

Ok, so my G0704 mill table is not flat. I don't think it is a leveling or tramming issue.

Brief description: During tramming, I noticed the middle lands are higher than the narrower edge lands. The middle lands are from .002" to about .0045" higher. The middle lands are shaped like a "U", more or less. They are higher at the edges of the t-slots, then lower at the bottom of the U. But even the bottom of the U is about .002" higher than the edge lands.

Even the edge lands have the U shape, but not as drastic - probably because they are not as large. I can measure between .0005 and .001 difference between the middle and the edge.

ALSO, the table is shaped like a banana along the X axis. At the extremes of travel the table is much higher. This problem I am uncertain if adjusting my gibs would solve (not sure if the table overhang weight is making the table rise up on the opposite end).

SO - I thought about skimming the table with a fly cutter.

But I've heard "Don't do that, it isn't a good idea." and also "Sure do it"

I'd like to get some more opinions and the why's associated with the opinions.

If I need to skim it'd be less than .010". I went ahead and made a flycutter that uses a heavy duty lathe turning tool with an 80 degree carbide insert. Cuts aluminum nicely and isn't too out of balance. Was running around 800 RPM on the aluminum. Leaves very little in terms of tracer marks.

Thoughts?

Note - I can't get the ENTIRE table (well, I could but the fly cutter would be more out of balance). I can get the entire middle land length along the full width of the table if that makes sense.

[Robin Hewitt]

I skimmed my mill bed. I started with a fly cutter which went fine until it hit a hard spot in the cast iron. Went back with a fat carbide endmill which complained a bit when it hit the hard spot then went on to find another one. I have a round column mill so I was able to offset the head and cut around the wells. Finished the last 1/16" across the front with a file. I wrote a XY only G code that cut with the quill locked.

It worked for me, accuracy is much improved. Don't know if I would recommend it though, it's a bit risky, you have to be brave

[ninefinger]

Ok, so my G0704 mill table is not flat. I don't think it is a leveling or tramming issue.

Checking for a twist with a level would be pretty cheap to do before I went ahead and milled my table. Pretty big gamble for "I don't think"....:stickpoke

My 2 cents...

Mike

[dugpits]

Checking for a twist with a level would be pretty cheap to do before I went ahead and milled my table. Pretty big gamble for "I don't think"....:stickpoke

My 2 cents...

Mike

Pretty cheap...assuming you have a machinist's level already.

I will be close to the Grizzly plant in Springfield, MO this weekend and might pick up the 8" x .0005" level while I'm there just to have. about $50.

However, I still "Don't Think" it is a twist issue. I can see light under a straight edge on my table. The table is higher in the middle (Y direction) than the edges. Along any given y, the x runs pretty true until the last 1 - 1.5" of travel.

If it was twisted I would expect a diagonal high/low pattern. Not a pattern within the lands. If it was bowed I would expect a low to high to low or vice-versa...but not a mountain zero-up/down/up-up/down/up-zero that I have.

I can't think of anything that would cause what I'm seeing except a non flat table.

I won't be able to do anything until at least the 28th since I'll be in Missouri with family until then...so if I get the level I will check it out.

Thanks for the feedback

-Doug

[cyclestart]

Thoughts?

Tooling plate ?
I'm half kidding but they do have advantages over a T-slot surface.

A quality straight edge would be another way to explore the flatness of the table. Lacking that, possibly piano wire ? (trying to think of a common alternative)

[dugpits]

Tooling plate ?
I'm half kidding but they do have advantages over a T-slot surface.

A quality straight edge would be another way to explore the flatness of the table. Lacking that, possibly piano wire ? (trying to think of a common alternative)

I used a parallel and I can get a .0015" feeler under it.

How does the piano wire work? Never heard of that one.

I ABSOLUTELY want a tooling plate - just not ready to produce it yet and don't have the raw stock and have to design just what I want. We use these a lot at work...allow for easy pinning of parts for setups. I have a modular vise (the movable jaw at least) that is low profile and will work well with a tooling plate since it has holes for alignment.


Now to figure out this piano wire stuff...

[Robin Hewitt]

Piano wire could work, presumably you stretch it like a guitar string over drill rod at either end then measure the height to bed along it's length. But why am I reminded of a famous piece of accidental verse, as in "I'm a poet, but I don't know it"...

And so no force, however great
Can stretch a cord, however fine
In to a horizontal line
That shall be absolutely straight

[LongRat]

I'd probably have a crack at it if I were you, but I would definitely extend the fly cutter to reach the whole table width in one pass. I am just surprised at how far out yours is, I don't remember seeing any reports of out of flatness for these mills even close to the numbers you are measuring. But then I haven't followed every thread.

[ninefinger]

Doug,

A diagonal pattern of high to low (measured at the spindle center line)would be expected if the twist was in the surface of the table itself but usually the twist is in the base. So as you move the table in the x-axis you don't see any movement relative to your spindle cause the table is moving in the cross slide and is constrained by it. If you move in the y direction you move through the trough (or hump) created by the twist in the base. I wish I had some visuals to help explain....

I hear you on the cost of the level - I borrowed one (actually 2) to level my mill.

I know its hard to visualize that you can twist that lump of cast iron but its true and in my case worked wonders for me in terms of tramming the spindle to the table - that is why I keep bringing it up.

Mike

[cyclestart]

How does the piano wire work? Never heard of that one.

Chalk it up to a late night and imagination in overdrive ?
The idea is borrowed from a router blog :
Machine frame - MadVac CNC
it's near the bottom of the page by the Moglice link
Whether the builder put too much faith in the lack of sag I couldn't say

We cross posted, it looks like you have some good reference tools.

[photomankc]

I think the "don't do that's" come along the lines of the fact that you can have issues involved that may not be the surface of the table alone. Your rise at the edges can come from the table levering up due to all that overhanging weight. We have long tables and kinda narrow saddles unless you know that is not the case then when you cut the table what you did was take a flat surface that was tilting a bit due to force and cut a nice convex surface into it. What happens later when you tighten up the gibs and it doesn't tilt as much now. I know this happens on mine because the weight of the vice obviously makes it more pronounced (0.0025 vs 0.001). Mine also leans front to back because the dovetails on the saddle are not exactly the same height. None of this can be corrected by cutting the table.

OTOH if the table surface is really boned that badly and you know it then I guess there isn't much you can do to make it worse right, but I would confirm this off the machine on a surface plate before I gave any thought to having the machine cut that surface in situ.

[TroyO]

How about lapping it instead? Get ~320-500 grit wet/dry sandpaper (Or several sheets if needed) and glue it with spray adhesive to a flat surface (Surface plate or even a piece of thick glass or a granite countertop cutoff). Wet it down then flip the table top-side down and run it in a figure 8 pattern, then reverse directoin for the same number of counts.

You can mark the top surface with ink or magic marker so you can see your progress. Removing the table to do it also eliminates the rest of the machine as the root cause.

I did this to my Mini-Mill table and it improved things greatly. I got scared and only did it until the center ~4 inches was still colored by the marker.... but I plan to go back and do the rest at some point.... it was the right move and made it way better I just chickened out before the job was all the way finished, LOL.

If you find a home resale store that sells used tubs, cabinets, etc they usually sell granite remnants and sink cut outs for pennies on the dollar. I got two 12x20ish hunks for $7.50 last time. It may not be certified but it's still "purty dern near" flat.

[dugpits]

I think the "don't do that's" come along the lines of the fact that you can have issues involved that may not be the surface of the table alone. Your rise at the edges can come from the table levering up due to all that overhanging weight. We have long tables and kinda narrow saddles unless you know that is not the case then when you cut the table what you did was take a flat surface that was tilting a bit due to force and cut a nice convex surface into it. What happens later when you tighten up the gibs and it doesn't tilt as much now

Totally agree - I was trying to describe this in my first post but didnt do it very well. I am concerned with skimming at the far ends of travel because of the table weight. Not sure how to alleviate the problem besides tightening the gibs.

Table is definitely jacked unless I'm doing something extremely moronic (even by my standards).

I will make a video when I get back from Joplin and post before I cut this thing...or do whatever to it.

[dugpits]

How about lapping it instead? Get ~320-500 grit wet/dry sandpaper (Or several sheets if needed) and glue it with spray adhesive to a flat surface (Surface plate or even a piece of thick glass or a granite countertop cutoff). Wet it down then flip the table top-side down and run it in a figure 8 pattern, then reverse directoin for the same number of counts.
.

That's a good idea....how long would it take do you think to remove about .005" ??

I would think that would take some time but I might look into it...

[k_yarina]

Is it still under warranty? Have you tried contacting Grizzley?

[dugpits]

Is it still under warranty? Have you tried contacting Grizzley?

I received in late February 2011. So technically it is within a year - but I had to cut clearance on the bottom of the table for the ballnut. So now it is "modified" and therefore outside terms of warranty...bummer

[TroyO]

That's a good idea....how long would it take do you think to remove about .005" ??

I would think that would take some time but I might look into it...

I will take some time... invest in the best sandpaper you can get (I like Norton) and it will go pretty fast. A decent evening endeavor I'd think. I'd go for the lower side of the grit (Maybe even 240 if you can find it in wet/dry). and get lots of paper so you can swap it out when it gets dull.

Better paper=fewer changes and less time.

[bjones]

grinding the underneath of the table as you've said can explain why the table is warped. it doesn't take much to cause cast iron to warp. it's very tempermental stuff. send it out to a shop and have it reground properly.

[SCzEngrgGroup]

Personally, I think trying to fix this problem by any means other than proper surface grinding is virtually guaranteed to fail. ANY slop or flex anywhere in the machine will lead to uneven-ness in the final surface. If the table is twisted, then I think a new table will be the only economical option. If not, then having the top surface ground by a commercial should not be very expensive, and will be by FAR the best approach. Fly-cutting, on the machine itself, I believe would be a complete disaster. Lapping, while it might work, would take weeks to remove 0.005".

Regards,
Ray L.

[wizard]

Do not try to fly cut the table, do not try to lap it, do not, do anything until the problem is fully diagnosed.

First off lapping will not leave you with a flat surface. Fly cutting is useless if you can't cover the entire table. Further it doesn't sound like you have even attempted to tune the machine or really do other diagnostic work. The banana shape you see on the X axis may very well be a loose or poorly fitted gib.

Rather you should consider taking the table off and inspecting and if required scrap the table in or have it ground. Be careful about whom you have grind the table as many machinist simply aren't setup to do the job properly. Either way don't do anything to the table until you have verified all the defects you want to correct.

Now I have no doubt that the table could potentially be warped, we just don't want to assume so. The castings aren't stressed relieved in most cases with this low end hardware. The good thing for you is winter is here. You could throw the table outside in the cold for a few weeks and then into an oven and back outside again to try to relieve any left over stress in the casting.

Ok, so my G0704 mill table is not flat. I don't think it is a leveling or tramming issue.

Brief description: During tramming, I noticed the middle lands are higher than the narrower edge lands. The middle lands are from .002" to about .0045" higher. The middle lands are shaped like a "U", more or less. They are higher at the edges of the t-slots, then lower at the bottom of the U. But even the bottom of the U is about .002" higher than the edge lands.

Even the edge lands have the U shape, but not as drastic - probably because they are not as large. I can measure between .0005 and .001 difference between the middle and the edge.

ALSO, the table is shaped like a banana along the X axis. At the extremes of travel the table is much higher. This problem I am uncertain if adjusting my gibs would solve (not sure if the table overhang weight is making the table rise up on the opposite end).

SO - I thought about skimming the table with a fly cutter.

But I've heard "Don't do that, it isn't a good idea." and also "Sure do it"

I'd like to get some more opinions and the why's associated with the opinions.

If I need to skim it'd be less than .010". I went ahead and made a flycutter that uses a heavy duty lathe turning tool with an 80 degree carbide insert. Cuts aluminum nicely and isn't too out of balance. Was running around 800 RPM on the aluminum. Leaves very little in terms of tracer marks.

Thoughts?

Note - I can't get the ENTIRE table (well, I could but the fly cutter would be more out of balance). I can get the entire middle land length along the full width of the table if that makes sense.