Leveling lathe on uneven floor before a CNC retrofit. Suggestions?

[Uguessedit]

Greetings everyone I’m new to post here and have a question for anyone who has experience or prior experience with the issue I am contending with. I’ve had this 13 x 37 lathe for around 6 months now and it’s been a good machine. I’ve done a few pre CNC upgrades but now I have all the components here so in preparing for my retrofit I want to get this leveled out. I’ve noticed and verified it has settled enough that the parts I have been turning had developed a significant taper (.026). The initial mistake i made was using hard rubber pads under the adjustment feet which is where the settling came from. This was in effort to keep it from moving about the floor, reduce potential vibration, and give me some room to manipulate leveling. Initially this worked well. Low and behold it was just a bad idea. That aside I removed the rubber feet and I am having a heck of a time leveling the stand. Stand consists of two lower cabinets and a center plate connecting them with coolant drip plate at the top.., typical of course. So I lowered all 8 adjuster feet and at the right side as far as it can go (7/8”). The left side points toward the garage door and in fact only a few inches clearance so you can imagine the floor taper I am dealing with. With approx a 1” taper toward outside in the 5ft span of the stand. So I made a couple large 6061 blocks for the front to get rid of the front adjuster feet and have a set level face to ease setup. I tapered blocks to match the floor and checked with a machinist level so I know that the front two blocks left to right are level to .005 per 10” span. This had no bearing on the actual bed being level yet. I will deal with that later. So now I want to level front to back. It seems the 12” span that the stand has a 1/8” drop to the rear as the floor appears to not only taper toward the garage door but towards the center of the garage dividing line. This lathe is approx 18” from this center line if you can imagine it now sitting on a double taper. As I stated I have the front level. As soon as I try and crank the leveling feet up in the rear to tilt the lathe forward “one foot adjuster only” it lifts the entire front lathe at both ends right up off the blocks I had it resting on. So the stand and lathe are apparently rigid. I did loosen the base fastners from lathe to stand to free assembly up. I had all but one foot loose and could go around and move them all by hand (rattling around) and the entire lathe had its weight balanced on the one adjuster foot. This is obviously the heavy corner where the CNC control cabinet is attached, 3ph motor, vfd, etc. I thought well I will just lift up on the attached cabinet adjacent to it further off the back of the lathe. Unfortunately the cabinet is not as strong and it simply flexed doing nothing to help tilt the stand forward. The only thing I can think of is to drill holes in the floor and attach the stand so I can pull down on the corners removing the apparent twist and forcing against the settled misalignment. Problem with this is that it isn’t my home to modify. Mentioning the issue to the family member I live with I had a negative response about drilling into the floor so I’m looking for an alternative method to getting this stand assembly to straighten out and remain level. At which point I will then shim the lathe atop the stand to compensate for any differences between the top side of the cabinets and at each end. Keep in mind I run coolant and we don’t want that running off the end tray either. To clarify I have been using a machinist level to .005 per 10 inch which i am sure is suitable for leveling this enough to rid the twist in the bed and misalignment from the chuck to tailstock. A few photos attached to clarify what I’m staying and hopefully I made some sense here. To add; Moving the lathe isn’t an option I haven’t anywhere to put it and it bows the roof trusses winching it up enough you can watch the outside rooftop dip so I don’t want to create further issues and so it needs to stay put until I move it out of here sometime in the future. Please share what you’ve done and if you dealt with a similar issue that would be greatly appreciated or if you’d just drill holes in the floor and patch them up later not saying anything? Something I’m considering I don’t want this sitting like this anymore and becoming impossible to reverse any settling in the future.

 

[10K Pete]

You want to move those aluminum bricks so the almost touch the bolts. I say that because it looks like the floor is coming apart! Other than shims there's machinery grout, a non-shrinking grout you pack in under the feet once the unit is leveled. Requires a nut on top and bottom of foot and a bolt that's embedded in the floor...

That's all I know. I've set some big 'uns..

Pete

 

[petertha]

Just an opinion but I'm not sure the rubber/puck style jackscrew levelling pads are the culprit. I see them used on lots of lathes & mills so long as they are appropriately weight rated. My 14x40 gearhead lathe & RF-45 mill both have them because my garage floor has a slope & I have had no issues. Lathe stand is 2x2 welded frame & levelling has been absolutely stable over the years. I can say that my particular ('97 vintage) Taiwain sheetmetal stands on RF-45 mill leaves something to be desired. In hindsight I would have made a tubular frame for that too. Maybe the new ones are better.

I was inquiring about base construction on this particular mini-Bridgeport mill. It might provide you some insight & ideas re floor conditions. Again, despite different ways to support & secure the base, most employed the puck feet connecting to the (stable) floor.
https://www.hobby-machinist.com/threads/935-floor-base.70746/#post-593389

Maybe your floor owners would allow you to cut 4 cores in the floor, deepen into underlying gravel & fill with good quality cement. The end result would be 4 flush 'discs' of slightly different color than the existing floor. Not even as bad as the nearby cracks LOL. I think the core drill is somewhat common Home Depot type rental equipment. I think they do this to cut drains & things in basement floors?

 

[bazmak]

You can level the lathe but you cant take out any twist in bed unless the lathe/stand is bolted to a base
or concrete floor.Althoug rubber in incompressable it will squeeze.The important part is to remove the bed twist
the easiest way is to LEVEL everthing. In theory the lathe can be on an incline and not level The bed twist can still adjusted but it is
difficult to check

 

[petertha]

I should have read your post more carefully. I assumed the taper was already corrected but it was more a matter of adapting the machine/base to an irregular/non-level floor. If you truly have a twisted bed, I agree with Bazmack: you need solid anchoring under each foot to independently pin & jack to un-twist it. And your stand has to be rigid enough to translate these forces to the lathe bed itself without just bending & absorbing the stress.

But let me back up a step. Have you verified that your headstock is not out of alignment with the lathe bed? My 14x40 has 4 vertical headstock retention bolts & perpendicular shim screws to adjust this (headstock rotation when viewed from top). If the headstock spindle axis is not parallel with bed axis, that would explain a taper despite the lathe being 100% level & 100% un-twisted. When you consider the trigonometry, just a whisker of rotational misalignment can translate into significant taper cutting. This of course assumes the lathe type where headstock casting is mounted to the lathe bed. If yours head/bed is cast as one then disregard, that option is not available to you.

 

[Hopper]

There is not much point in chasing your tail trying to get a flexible sheet metal stand level to precision tolerances. Sit it on the floor. Set packing under it as you have done to get it as close as you easily can with no gaps between the bottom mounting surfaces of the cabinet and the floor. Then tighten down the bolts. This means the cabinet is not pulled out of shape when you tighten the bolts down. (Yes those blocks and shims need to be right at the bolts, preferably with a slot in them so they slide in and wrap around the bolt like a U.)
Don't even try to correct the 1/8" front to rear slope on the cabinet base if it causes problems.

Then do your final leveling of the lathe bed. At that point you should be able to pack under the lathe feet on top of the cabinet to get the bed level. Or use Myford-style leveling studs and nuts to make it easier. (See Bazmak's Myford threads for how he made his own nifty adjusters.)

In a nutshell, let the cabinet fall where it will and do your adjustment at the lathe feet.

And don't even think about adjusting headstock alignment until after you have got the bed straight by getting it level all the way along and ensured headstock bearings are good and tight. No point in aligning a headstock to a banana bed and shaky spindle.

 

[Anatol]

I will be watching this thread as I need to do it for my 12x36
With all respect to those more experienced who have already replied -

I'm not too experienced in leveling lathes specifically but know about cracked concrete floors and (flexing) sheet metal. I don't know what area you're in but you seem to have floor subsidence. I would be inclined to spread the load of the feet by sitting each cabinet on a flat and rigid plate - and maybe remove the adjusting screws? Two cast concrete paving stones might work ? Two bits of granite counter top would be an interesting option to think about... You could level these with non-shrinking grout or whatever it is tilers use to level before tiling (can't recall what its called) - or wedges. That would be easily removed later. Just my two cents FWIW.

 

[Uguessedit]

Thank you for the reply and I agree that the floor is part of the issue at one It is in rough shape where it would normally need its main support. I dismantled the base and suspended the lathe taking measurements of the end cabinets and they are definitely out of square. Looks as though the stand was welded together this way. That seemed to be amplifying my issue. After getting the two cabinets as close as I could I ran angle across the lower front and attached to the bottom of both cabinets. This way I was able to establish keeping it level left to right so I wasn’t fighting the front side. It was more difficult than you would think. I had to wait 10-15 minutes between adjustments tonallow for any flex in the sheet metal. Not a great design for a lathe stand. So I did let it all settle overnight and it has remained level side to side and atbthe momeny I’m dealing with the leveling front to back. Similar issue though where I lift the rear side (doesn’t matter which end) and instead of the stand tilting forward as I need it wants to lift the entire left or right cabinet suspending the front off the ground. My thought is to try a couple tapered shims and see if I can get those to manipulate the stand so to lean into its proper placement. If this doesn’t work I did pick up two floor anchors I may have to resort to on the front side despite the likliness of moving the machine in 6 months it’s the last resort. I believe that front anchors and tapered shims in the rear will force it to tilt. It just seems a little unatural butnif the cabinets are welded out of square I haven’t much choice. The top side of the left cabinet where the drip pan sits tapers almost an 1/8” on that cabinet top alone. I will have to shim between the cabinet and pan so the lathe doesn’t pull this together when I tighten it down to the stand. The weight of the lathe itself is offset toward the rear with motor which appears counter productive in getting it to lean forward. Ultimately this stand is awful. It’s always been somewhat wobbly, less now that the rubber pads are removed.. if this works as I hope it should get by a few months until I move it and can fabricate a heavier gauge stand with a wider base and slightly longer length for better stability. The angle I added atbthe base is facing out 1-3/4” inches and has improved the stability where adding that inbthe rear maybe enough to help. The wider the more apt it should be to tilt and lean to a level position versus lift the entire end. All great suggestions here and to the other gentleman about the rubber feet I’ve used and seen them used on other machines too. I think as stated it comes down to the stand quality here and not enough footprint to disperse the weight more evenly across the floor.

You want to move those aluminum bricks so the almost touch the bolts. I say that because it looks like the floor is coming apart! Other than shims there's machinery grout, a non-shrinking grout you pack in under the feet once the unit is leveled. Requires a nut on top and bottom of foot and a bolt that's embedded in the floor...

That's all I know. I've set some big 'uns..

Pete

 

[Uguessedit]

Photo of the modification I chose to use to widen the base and contour the floor taper. Also this makes it easier to attach with anchors and I can grind off the angle later if I don’t decide to scrap the stand when I replace it with a wider footprint. Setting the right side flat on the floor didn’t help stability any more than it was on the rubber pads but the angle has helped since attaching. This will require a 2mm shim in the right rear corner and thicker shims as you go to the left to just over an inch. After I get it blocked and level front to rear I will attach a 5ft section of angle on the back side like the front. Hopefully this alleviates the problem for the time being so I can focus on installing ballscrews, etc..

 

[Uguessedit]

The floor is problematic at this end (left side). Of course this is also where all the weight sits and gap needed to level it. Probably difficult to see in the photo but it’s uneven every direction each side of the cracks a 1/16 or more. Using angle front and rear lets me avoid having to contend with this. We’ll update...

 

[petertha]

And don't even think about adjusting headstock alignment until after you have got the bed straight by getting it level all the way along and ensured headstock bearings are good and tight. No point in aligning a headstock to a banana bed and shaky spindle.

In all of 5 minutes you can insert a test bar in the spindle socket (thus eliminating chuck effects), set a DTI on the carriage, traverse it down the bar length & 100% eliminate the possibility the headstock is yawed in/or out as well as up/down. I bought a India made 18" long precision ground bar made with MT3 taper on one end & center drilled on both ends bar off Ebay for a whopping $50. Its also useful for tail stock alignment, a separate discussion.

If the head stock happens to be even a whisker canted & you measure 0.020" runout at the bar end, the amount of bed twist to compensate that to zero is substantial & not healthy for your bed. That's also chasing your tail for the wrong reason. The bottom line is both issues (twisted bed and canted head stock) are potential problems & can mask one another depending on which is out, direction & to what degree. They both need to be addressed. If you find B-then-A easier than A-then-B, that's fine. As long as you ultimately check both A&B.

Tom Lipton shows this in a 2 part video. The first segment is about turning & measuring a test bar & jacking the lathe bed (twist). As I mentioned above, this may be sufficient, or, it may only get you so far. Around min 25 he illustrates the headstock issue. The OP hasn't confirmed if his lathe is this style or not or whether he has measured. I'm just pointing out another parameter to consider.

 

[Uguessedit]

Well I fastened all four corners to the floor and it’s impossible to get it exact between the cabinet and the floor issues. Front to back took priority. As soon as I’d level one side the other would change. This is the best I could get the twist out of it. The floor inconsistencies make it impossible to configure a proper shim when it’s changing every 1/4” every direction. I will turn a part and see how Accurate it is. Without measuring actual twist and relying on the bubble I’d have to guess it’s fairly minute and probably not an issue. Besides I don’t turn anything as long as the bed capabilities. The small amount of twist occurs at the last 10” which still leaves me 27” of level bed. As you can see both ends are still in the center lines of level just offset one direction to the other. From side to side I lost the level adjust I had and will accept a slight downward slope to the left for now. Everytime id add a shim and settle the lathe I’d keep chasing it around four corners. If this gets it close I can always shim the lathe to stand for further accuracy which I have not yet done. I have to suspect the garage floor had settled over the past few months. Something or keep an eye on and if I need to make adjustments it will be a lot easier now. I appreciate the feedback and with 7 machines here this is the only one that has been problematic to level. I will admit I sat down a few times and watched a television show so I didn’t pull my hair out. Any further suggestions feel free to share. I will post results later if any taper remains using some 4140.

 

[Hopper]

In all of 5 minutes you can insert a test bar in the spindle socket (thus eliminating chuck effects), set a DTI on the carriage, traverse it down the bar length & 100% eliminate the possibility the headstock is yawed in/or out as well as up/down.

Not if the bed is twisted. The bed needs to be leveled first. Always. It is the reference surface off which all other measurements are taken.

 

[Uguessedit]

After test turning I am getting .013 (inches) taper over a 4” bar of 3/8 4140 after taking a .25mm cut. The smallest part of taper is at the lathe chuck. The tailstock is straight in line but I had initially suspected there is play in the tail stock quill despite the miles are low on a relatively new BT1337g lathe. I am using a Shars live center and to rule those out I decided to run the part without the tailstock and I have the same taper occurring where it’s smaller toward the chuck. Inspecting the carriage I see nothing jumping out at me causing issue. Coincidence part/tool deflection? I’m going to run some 6061 and see what happens although In the past I hadn’t any issue turning 4140. No visual deflection either... It had been 8-9 weeks since I last ran these same parts and suddenly the big change which started this releveling project. The vast majority of my parts the outside diameter is purely cosmetic so it’s a non issue but these parts are high rpm axle shafts that have bearings at either end. Obviously I don’t want a bearing tight one one end and loose fit the other. I realize once converted to CNC this can be compensated for but I’d rather get to the bottom of the mechanical issue. It seems it may be a few issues that have simply gone unnoticed until now. Double checking center hole alignment from tailstock looks to be accurate. I will verify again.

Edit: tailstock is off .002. Will adjust and recheck taper.

 

[Hopper]

First thing you need to get clear is that taper with a tailstock centre is a whole different issue from taper without a tailstock centre. The first one you need to get right is the taper without tailstock centre. After you have that right, then you can adjust the tailstock centre to turn a parallel test bar between centres (Not a bar held in the chuck with a tailstock centre.)

Second thing: With your test done so far, your 3/8" diameter bar of 4140 is waaaay too small in diameter to provide any kind of meaningful reading. What is happening is the end of the bar is flexing away from the tool making the bar larger diameter the further away from the chuck.

A test bar sticking out of the chuck four to six inches should be one inch diameter or more. Finishing cut should be a couple of thou for finest finish you can get. Give it a try with that and I bet your result will be much better.

Material can be mild steel or maybe good aluminum alloy such as 6061 T6. I wouldn't use 4140 as it is pretty tough and could flex a home hobby lathe a bit and blur your readings.

 

[BaronJ]

Hi Guys,

You need to solve the problem of the floor first ! All you are doing is adding bandaids to a problem.
I would shift the lathe so the area where you want to put it is clear. Make a timber frame big enough to enclose that area about 40 mm thick and tack strips of plywood on the inside to match the floor. The object being to get the top of the frame level in all directions. Go buy a bag of cement and some bags of sand to make a wet mix that will run into any cracks and gaps in the floor. Smooth it off using the top of the frame as a guide. You might have to use some heavy items to stop the frame from moving whilst you are pouring the concrete mix. When this has cured, in about a week, you will have a secure flat base to put your lathe onto. Now you can start to set the lathe up.

JMTPW

 

[KellisRJ]

Just for what it's worth, my brain struggles with out a line break or two . .
Point 1. The goal is to not have the lathe bed twisted. Ships are never level, but the bed can still be straight.
Point 2. The stand needs to be as stable as possible, then leveled as best can be managed. In your case you can do both at once.
Build a quick frame.
Put down some cheap plastic on top of the frame/floor.
Lay as thin as possible bed of floor level on top of the plastic. Thin so you can bust it up later.
Put down as thick as 4 ~ 6" strapping as you can afford, and put some rectangle tubing on top of that.
Bolt the stand to that.
Read up on Rollie's Father's Lathe Leveling Method. Some like it, some don't.

Now do the taper test and shim the lathe feet as necessary to remove any twist.

Just my $.02.

Ron

 

[BaronJ]

There is nothing wrong with any method of getting a lathe to turn straight, as long as it achieves the desired result ! The problem is keeping it that way !!!

 

[Anatol]

This thread is so useful and so basic (fundamental, not dumb). And I have a (12x36) lathe with a tail-stock alignment problem

The thing that fascinates me is that such a fundamental task - getting something straight, flat and level - can be so (intellectually) challenging. And the wonderful and terrible thing about machining is that 'almost' is not good enough.

Having a strong, flat floor is no doubt a great help! I don't. My lathe is sitting on a plywood floor (prefab shed). It probably bounces around. So I'm going to start from the bottom, like Ugeussedit is doing. But also like him, its not my shed, so pouring concrete isn't an option.

I've got good 3D thinking skills and good making skills, but still such problems challenge me. I'm sure I'm not alone. It would be interesting to distill this process into a FAQ or how-to. Or maybe someone knows of a set of really good "How to level a lathe" instructions somewhere?

 

[BaronJ]

The bottom line is to make the foundation for the lathe as rigid as possible ! Ignoring coolant running to the wrong end of the catch tray, is doesn't matter if it slopes a little. I know of a chap who is in a similar situation to you. He has his lathe mounted on several railway sleepers with a plywood sheet screwed down to them. But it means that he can use the rigidity to control the twist in the lathe bed.

My lathe is stood on a hard rubber mat on a concrete floor not bolted down, but I do check and correct the slight variations in twist caused by temperature fluctuations, if I need to.