Conceptual Design of DIY High Precision Dual Turret Slantbed Lathe

[Hezz]

Hello everyone,

I am starting this thread as a design log for my high precision dual turret slantbad lathe with live tooling. This is kind of an ambitious project. One with perhaps higher asperations that I will be able to attain. It is going to be a long haul type of project since I still need to figure out how I'm going to build such a machine. Much of the ground work has already been laid out in my head but a lot of CAD design still remains. It is intended to be a heavy hobby lathe with approximately a 30 x 48 inch footprint on a roll around heavy duty kart that allows for storing near a wall in a tight garage. Still it will be heavy enough that it would be optimum to have it left in the same location. My goal is for it to weight no more than about 1500-2000 pounds. It will really be a mini turn/mill center with integrated cooling and and enclosed cabinet.

I also want it to have an automatic chuck if possible. But I may forego this feature in the interest of cost and complexity. What I am interested in is a small footprint milling lathe that can do pretty complex parts without too many setup changes. Ideally, the operator will only have to change the part around for the second operation and align a new piece of material in the chuck jaws at the beginning of the turning cycle.

I will seek to make it as accurate as I possible can with the tools that I have or will be able to obtain.

The base and frame will be a hybrid welded steel and concrete and EG constructed frame. With perhaps an integrated box type space frame connected to the main monolithic weldment and casting to function as the cabinet structure and to add additional stiffness.

Here is a picture of the first conceptual idea for the base and subframe. The main plate is not shown in this picture. But depending on how this part is manufactured. The main slantbed base plate my be constructed of thick steel or epoxy granite.

The manufacturing process will be slightly different for the two methods. With the steel slant base the entire weldment will be built up and welded carefully to the underside of the slantbed plate. Then the back of the weldment mold will be left of and the whole assembly will be turned over to be filled with EG or concrete.

The other method calls for welding the entire base weldment mold together except for the slantbed plate and filling the weldment with concrete up to an inch or two from the top and then a second mold is fastened to the main weldment and a slantbed plate of EG is poured on top of the concrete and finished off by lapping to form the plate for the headstock and axis rails.

Similar Threads:

• High Precision Mini /Small Lathe
• New Machine Build- DIY Mini Router first conceptual design
• Need Help!- v26 lathe grooving and dual turret posting
• Help me decide which high precision CNC lathe to buy!!
• VRMesh v4.1 released for reverse modeling and conceptual design

[123CNC]

Nice, looking good so far. I like your idea on the 2 part molding if you will. Although I don't know of anyone every doing it that way. Any concerns with how well the interface boundary binds or interacts?

Looking forward to seeing the top half, or the slant bed surface plate/mold portion of your model. I'm guessing you'll imbed your spindle mount hardware as well any linear rail (guessing) reference ledge or mounts as well.

You might want to make 2, so that you can practice or prototype out on the first and master on the second, Then I can haul the prototype one off to scrap for you .

[lcvette]

I think you may have a problem with bonding epoxy to concrete. Concrete has a powdery type surface finish which may not allow for a solid bond over time..

I would opt for a full EG pour as you will be using alot of aggregate anyway so that small of a machine should require alot if expensive epoxy.

You could also forego the heavy surface plate with a full EG pour and just use some 1" X 2" bar stock welded to the ribs for your linear rail mounting and some small plates for bearing supports for your screws. A big thick steel plate will be difficult and expensive to machine consider the small amount that will actually be critical for rails, screws and headstock mounting. And then just make the mold top about 3/8" down from the tops of the bar stock pour the mix so it covers the rib tops by the remaining 5/8". This will save you alot of coat and be just as strong if not stronger since you will have 1" thick rail pads.

For the dual spindles, you should design in a dedicated rail for automating moving the work from the front chuck to the back chuck. I don't think that would be too difficult with some looking into how the big boys do it. At least not mechanically, control wise... Way beyond my scope... Lol

I daydream about builds like this all the time, it was the reason I settled on my 3in1. Because it offered a huge table 6.26" X 19" with a 17" swing and 31" between centers. And it is a full dovetail bed with closed top. This offered the ability to convert to linear rails and still have over 9" swing over the table and be able to use a large diameter tool turret with plenty of bed left for gang tooling.

For me this was an awesome compromise between having to build and having a platform that allowed me to convert in my terms without compromising my goals.

I'm almost done with the solid modeling for the conversion. I will be setting it up as a slant bed for better chip clearing and flipped the carriage so the long end was on the backside to mount the turret on.

The bed is long enough to add a rear headstock for dual spindles too... BiG sucker for the money.

The best part is I grabbed it in unused new condition on craigslist for $900. And it solved so many obstacles a full build from scratch would present. It also came with a milling head that I can use for another project a cnc drilling machine most likely.

I really wanted to undertake a project like that but when I started adding up the costs and considering the time.... It got overwhelming.. So I decided to get a leg up and just focus on the other items I wanted to add and figured I had more chance for success in completing the build this way. I don't want to dampen your drive or spirits, but if you were line me and were simply trying to have functionality not available in a normal manual Lathe conversion. Perhaps take a look at the 17x31 3in1 machines.. They are always on craigslist and I think I have been seeing them new for around $1700 lately... Very good deal for alot of machine!

Chris

Sent from my XT1080 using Tapatalk

[Hezz]

Nice, looking good so far. I like your idea on the 2 part molding if you will. Although I don't know of anyone every doing it that way. Any concerns with how well the interface boundary binds or interacts?

Looking forward to seeing the top half, or the slant bed surface plate/mold portion of your model. I'm guessing you'll imbed your spindle mount hardware as well any linear rail (guessing) reference ledge or mounts as well.

You might want to make 2, so that you can practice or prototype out on the first and master on the second, Then I can haul the prototype one off to scrap for you .

If I do decide to go with the two part casting the top EG will be just a flat bed or just an enclosed volume to hold and damp vibrations for some steel rail beds. I don't think think that I will cast the headstock into the EG bed part. I don't think that I would have the necessary skill or tools to get it accurate enough that way. The Headstock will have to be it's own weldment or casting with alignment adjustment built in.

[lcvette]

Or bed it and then drill and bore it using the X and Z axis so it is perfectly aligned to the finished bed. That is how many machines are finish honed for perfect alignment.

Sent from my XT1080 using Tapatalk

[Hezz]

I think you may have a problem with bonding epoxy to concrete. Concrete has a powdery type surface finish which may not allow for a solid bond over time..

I would opt for a full EG pour as you will be using alot of aggregate anyway so that small of a machine should require alot if expensive epoxy.

You could also forego the heavy surface plate with a full EG pour and just use some 1" X 2" bar stock welded to the ribs for your linear rail mounting and some small plates for bearing supports for your screws. A big thick steel plate will be difficult and expensive to machine consider the small amount that will actually be critical for rails, screws and headstock mounting. And then just make the mold top about 3/8" down from the tops of the bar stock pour the mix so it covers the rib tops by the remaining 5/8". This will save you alot of coat and be just as strong if not stronger since you will have 1" thick rail pads.

For the dual spindles, you should design in a dedicated rail for automating moving the work from the front chuck to the back chuck. I don't think that would be too difficult with some looking into how the big boys do it. At least not mechanically, control wise... Way beyond my scope... Lol

I daydream about builds like this all the time, it was the reason I settled on my 3in1. Because it offered a huge table 6.26" X 19" with a 17" swing and 31" between centers. And it is a full dovetail bed with closed top. This offered the ability to convert to linear rails and still have over 9" swing over the table and be able to use a large diameter tool turret with plenty of bed left for gang tooling.

For me this was an awesome compromise between having to build and having a platform that allowed me to convert in my terms without compromising my goals.

I'm almost done with the solid modeling for the conversion. I will be setting it up as a slant bed for better chip clearing and flipped the carriage so the long end was on the backside to mount the turret on.

The bed is long enough to add a rear headstock for dual spindles too... BiG sucker for the money.

The best part is I grabbed it in unused new condition on craigslist for $900. And it solved so many obstacles a full build from scratch would present. It also came with a milling head that I can use for another project a cnc drilling machine most likely.

I really wanted to undertake a project like that but when I started adding up the costs and considering the time.... It got overwhelming.. So I decided to get a leg up and just focus on the other items I wanted to add and figured I had more chance for success in completing the build this way. I don't want to dampen your drive or spirits, but if you were line me and were simply trying to have functionality not available in a normal manual Lathe conversion. Perhaps take a look at the 17x31 3in1 machines.. They are always on craigslist and I think I have been seeing them new for around $1700 lately... Very good deal for alot of machine!

Chris

Sent from my XT1080 using Tapatalk

Hi Chris,

Sounds like you got a really good deal on the Craigslist lathe that you converted. There is a great deal of merit in doing what you did. It sounds like a good solution for your machining needs and it saved a lot of money and time. My problem is that I have very little space and very little money that I can come up with at one time. Also, I want to see if I can advanced the state of the art for a small DIY CNC lathe. I just got spoiled working on Mazak CNC lathes a few years ago and now I want something nearly as nice in a small package.

I was under the impression the epoxy would bond very well to concrete if it is cleaned and etched with acid and made to be very porus at the boundry layer. Also, I think that other means could be used to bond the EG to the concrete. i do think that a full EG fill would be ideal but It would be too expensive for me to do it that way. I still have a lot of thought to put into exactly how I'm going to do this. I'm thinking that if I do an EG top which is more likely due to the high cost of thick steel I would put in a whole lot of rebar sticking up out of the cement in a grid of about 4 inches apart. Between the EG grabbing the rebar and flowing through the holes in the ribs and bonding to the cement to some degree, I think it would work.

I still like the idea of having a thick steel surface plate rail bed if possible because I think it will be easier to perfectly align and mount the rails that way. But the only way that is going to happen is if I find a piece of surplus 4140 or something like that for cheap. It could be done with 1/2 inch or 3/4 inch thick mild steel. I can buy a piece big enough for 80-120 bucks but it will be cut on a shear and not flat. In the end I will probably be forced to use the cheapest method which is something like you describe.

I will forgo the movement on the tailstock or second spindle. The dual turrets is mostly so I can have a lot of tools in optimized positions with smaller turrets rather than to turn on the second spindle. That would be too costly for me to build. But I will want to be able to turn longer pieces of material between centers. I will try to build just a high quality tailstock spindle that can perhaps mount a chuck if needed to hold heavy long pieces.

Because my goal is to have live tooling in these smaller turrets. It makes more sense to me to have two turrets that can move on the same rails but move out of each others way. One for horizontally oriented tools and ones that are perpendicular to the Z axis. I've got some thought to put into that.

I want one turret to have an additional axis for milling on the other axis.

Maybe I should design some type of sub-weldment for the rail bed that accurately and stiffly holds them into place while about 4-5 inches of EG mix is poured to create a volume to bond that assembly to the main big one.

[lcvette]

You are correct that the best way is to acid etch and use some stickout rebar to link the two. I would use a tile trowel to adds ridges to the concrete mating surface to increase the surface area of the bond and add some irregularity so it is as strong as possible.

You could also get some heavy gauge nails and bend the tips 90 degrees in a vice with a hammer and just plug the 90 degree ends into the wet concrete in a grid pattern with a nominal spacing of 1.5"s... This would really help lock it all together.

I really like your live tooling idea!

I mentioned this earlier in Siams thread because of what an awesome project i thought it was, and how budget oriented it was. The mini Lathe headstock with a 5c spin indexer spindle swapped out. This could double as both a tail stock and second spindle with interchangeable chucks. You could cnc it so it moved on the z axis on its own rails and add a X movement to it for your z axis live tooling. The X movement and main spindle could move together to generate linear interpolated cutting paths.

If you put it on a horizontally mounted rotary table with a vertical movement it could handle all your X and Z axis machining ops and be your secondary spindle and tail stock. Just send your main turret to the location you currently have your live tooling head located. Its table could share the main turrets rail and sit on a dedicated tailstock rail for best rigidity.

For saving space on a small machine, this would add some really slick functionality. So many ideas...

Can't wait to see what you come up with!


what if you used something affordable like a mini

Sent from my XT1080 using Tapatalk

[Hezz]

Last night I couldn't sleep for thinking too much about this project. After thinking about the realistic budget aspects about this design I realized that I would probably have to pour an EG top since the thick steel top will be too expensive to buy and have machined.

So now my thinking is to have the weldment mold poured mostly full of a high strength concrete and then finish the top off with an EG fill that will create the rail mounting surface and be a lapped flat reference surface for the rail mounting. I know that it is generally preferable to mount the rails directly to some kind of steel rail beds. And I think that this is ideal for a machine that makes heavy cuts since the steel will not deform as much as the EG. But my budget will limit the amount of horsepower that I have and so this machine will focus on making high quality and repeatably accurate cuts instead of heavy cuts. Thus, I think that having the rails mounted directly to the EG surface will improve vibration dampening.

I have added some 1 x 2 inch solid steel ribs that are welded into the weldment mold and will rest about 1.5-2 inches down below the top of the EG surface and will be located below the rails for drilling and tapping. Thus, the bolts mounting the rails will have a very solid material to grab into and the 1 x 2 solid steel ribs will have enough EG above them to really hold them down into the matrix.

Here's a snapshot of the current design concept showing a wooden temporary rim around the outer edge to hold the EG mix until it sets. Then once it sets hopefully I can hand lap the top surface to get a very flat surface with no obstructions in the way. By using this method I will not have to do any machining that I won't be able to easily afford.

[Hezz]

You are correct that the best way is to acid etch and use some stickout rebar to link the two. I would use a tile trowel to adds ridges to the concrete mating surface to increase the surface area of the bond and add some irregularity so it is as strong as possible.

You could also get some heavy gauge nails and bend the tips 90 degrees in a vice with a hammer and just plug the 90 degree ends into the wet concrete in a grid pattern with a nominal spacing of 1.5"s... This would really help lock it all together.

I really like your live tooling idea!

I mentioned this earlier in Siams thread because of what an awesome project i thought it was, and how budget oriented it was. The mini Lathe headstock with a 5c spin indexer spindle swapped out. This could double as both a tail stock and second spindle with interchangeable chucks. You could cnc it so it moved on the z axis on its own rails and add a X movement to it for your z axis live tooling. The X movement and main spindle could move together to generate linear interpolated cutting paths.

If you put it on a horizontally mounted rotary table with a vertical movement it could handle all your X and Z axis machining ops and be your secondary spindle and tail stock. Just send your main turret to the location you currently have your live tooling head located. Its table could share the main turrets rail and sit on a dedicated tailstock rail for best rigidity.

For saving space on a small machine, this would add some really slick functionality. So many ideas...

Can't wait to see what you come up with!


what if you used something affordable like a mini

Sent from my XT1080 using Tapatalk

Hey Chris,

good idea about the nails and using a tile trowel. I think that between using short pieces of rebar and bent nails and using a course tile trowel to get a really rough surface at the top of the concrete that I could get a really good bond. Maybe I could go over it with a broom to to put micro grooves in the cement surface too. Really rough it up as much as possible.

Using a horizontal rotary table for the tailstock with a swapable live center and chuck is an interesting idea. I will have to ponder on that one a little. It would add some more machining flexibility. But I really want to make this lathe capable of making high precision spindles of various lengths as a first priority so having a tailstock design that I can really dial in to high accuracy is a major priority. In fact, I want the main spindle to be an air bearing spindle eventually. I also want to have an air bearing spindle live tooling on one of the turrets for doing high precision grinding.

Will, this ideas really brings up something to think about. What if I was to think about making the tailstock a secondary spindle with an air bearing and drive. This would seem to lend itself well to functioning at times as a dumb passive spindle. This would make an excellent place for doing grinding operations.

Too many ideas!!!, but that is what this stage of the design process is about.

[Hezz]

I think that I want to make an air driven grinding spindle that mounts on the main headstock and tailstock rails and slides back and forth on it's own X axis between them and is mounted below the Z axis. It wouldn't need to be a large so it wouldn't get in the way too much. I like this added functionality since this is kind of meant to be a do-it-all kind of turning machine as much as possible.

[lcvette]

Are you wanting to primarily make spindles? A grinder setup sure would be sweet for high Precision work. The more I think about machine builds, the hardest part I think isn't building the structure as much as getting everything perfectly lined up with minimal equipment..in some cases, the Precision components required to get the alignment perfect cost as much as a machine...lol

Sent from my XT1080 using Tapatalk

[Eddard_Stark]

Are you wanting to primarily make spindles? A grinder setup sure would be sweet for high Precision work. The more I think about machine builds, the hardest part I think isn't building the structure as much as getting everything perfectly lined up with minimal equipment..in some cases, the Precision components required to get the alignment perfect cost as much as a machine...lol

Sent from my XT1080 using Tapatalk

re: precision components for perfect alignment; you have probably seen Dan Gelbart's scratch built high precision lathe and grinder. If you haven't here is a link:

His method of using granite parallells for high accuracy rail beds was pretty neat I thought. Precision Granite USA 562.696.8328 - Manufacturer of custom granite surface plates, cases, Cleaner, Inspection Table, Master Square, Stand, Straight Edge, Tri-Square, V Block, Southern California. Somewhere he posted a link to the price list and they aren't that expensive for the accuracy you get. To give you an idea he says he has less than 10k in the entire lathe accurate to a micron. If you go through all of the comments he explains more about the build.

What I wonder is if you use two granite parallels held in compression (epoxied with the aid of a surface plate just as Dan did) and epoxy? threaded inserts in the parallels for mounting the rails and embed the entire structure in a larger EG frame how rigid the build would be. I would think you could also epoxy threaded dowels on the sides of the parallels for boding to the larger EG frame. I have no idea if it is a good idea or not, but it would allow one to sidestep one of the most challenging problems of a DIY build; precision rail mounting and alignment.

[Aaron Bruce]

So great job!

[Hezz]

Are you wanting to primarily make spindles? A grinder setup sure would be sweet for high Precision work. The more I think about machine builds, the hardest part I think isn't building the structure as much as getting everything perfectly lined up with minimal equipment..in some cases, the Precision components required to get the alignment perfect cost as much as a machine...lol

Sent from my XT1080 using Tapatalk

I want to be able to make almost anything of small or medium size. But I especially want to be able to make good quality heavy duty spindles. I really want to make a micro grinding machine similar to the one of Dan Gelbarts. The only problem is the granite parallels. The high quality ones are really expensive and he uses four of them. If he's using the really high grade one's he has 20-30 thousand dollars wrapped up in just the parallels.

[Hezz]

re: precision components for perfect alignment; you have probably seen Dan Gelbart's scratch built high precision lathe and grinder. If you haven't here is a link:

His method of using granite parallells for high accuracy rail beds was pretty neat I thought. Precision Granite USA 562.696.8328 - Manufacturer of custom granite surface plates, cases, Cleaner, Inspection Table, Master Square, Stand, Straight Edge, Tri-Square, V Block, Southern California. Somewhere he posted a link to the price list and they aren't that expensive for the accuracy you get. To give you an idea he says he has less than 10k in the entire lathe accurate to a micron. If you go through all of the comments he explains more about the build.

What I wonder is if you use two granite parallels held in compression (epoxied with the aid of a surface plate just as Dan did) and epoxy? threaded inserts in the parallels for mounting the rails and embed the entire structure in a larger EG frame how rigid the build would be. I would think you could also epoxy threaded dowels on the sides of the parallels for boding to the larger EG frame. I have no idea if it is a good idea or not, but it would allow one to sidestep one of the most challenging problems of a DIY build; precision rail mounting and alignment.

That is probably my favorite video on You Tube. I don't know where he got all of his granite parallels. A cursory search on Google shows average price for 1 high grade granite parallel at 15,000 bucks USD. Perhaps he is using lower quality ones or got some kind of educational deal. I'm sure there are cheaper ones but I didn't find any in a few minutes of searching.

What I really want is to make a small bench top micro grinder that has nearly the quality of Dan Gelbarts for not much money. All of the linear ways and spindles are air bearings. I wouldn't think that you would need tolerances any better than about .0005 an inch for linear air bearings. I might be wrong. But the main improvemnt in my view is just eliminating all of the vibration by having all moving surfaces close tolerance and air bearing. His Lathe /grinder combo is not really laid out the best for productivity. But his system really works well.

Think how cheap you could make ultra high quality spindles if you had a grinder like that. You wouldn't even need to buy expensive roller and AC bearings.

I think you could come pretty close just by using EG, steel reinforcement for the bed and build the ways from stainless steel. It would needs lots of grinding and hand lapping . The cutting forces are light so stiffness is not going to be as important as dampening in my opinion. As long as it is pretty stiff.

[lcvette]

I dunno.. Air bearings have to be tight enough to not allow an air molecule from escaping. If it isn't and they can escape then it is metal on metal bearing..lol. But it would still be a really accurate Way. I will leave the uktra Precision to you guys... I'm floored if I could hold .0005" and have it be repeatable on anything built for under $10k. More realistically for anything I work on .001" is beyond good enough.

Sent from my XT1080 using Tapatalk

[ebrewste]

Air bearings leak air in normal operation:

"The design of the air bearing is such that, although the air constantly dissipates from the bearing site, the continual flow of pressurized air through the bearing is sufficient to support the working loads."

Air Bearing Basics | NEWWAY Air bearings

I'm not saying they don't have to precise, but the containing every molecule thing isn't how they work.

[lcvette]

Good to know, that video was the first I saw of an example.. I went by him saying it had to be air tight.. My mistake!

Sent from my XT1080 using Tapatalk

[Hezz]

Air bearings leak air in normal operation:

"The design of the air bearing is such that, although the air constantly dissipates from the bearing site, the continual flow of pressurized air through the bearing is sufficient to support the working loads."

Air Bearing Basics | NEWWAY Air bearings

I'm not saying they don't have to precise, but the containing every molecule thing isn't how they work.

I wonder if air bearings could support the loads of a typical single point cutting operation. It's obvious that they are sufficiently stiff for grinding. Thanks for the link ebrewste.

Assume for the point of argument that we had air bearing ways on the X axis and on a live grinding tool. If we were to turn the main spindle for grinding at a very slow speed we could reduce vibration to very small levels. I wonder how close we could come to Dan Gelbarts precision. Interesting question. I know the really expensive big multi turn lathes do this.

[Eddard_Stark]

That is probably my favorite video on You Tube. I don't know where he got all of his granite parallels. A cursory search on Google shows average price for 1 high grade granite parallel at 15,000 bucks USD. Perhaps he is using lower quality ones or got some kind of educational deal. I'm sure there are cheaper ones but I didn't find any in a few minutes of searching.

What I really want is to make a small bench top micro grinder that has nearly the quality of Dan Gelbarts for not much money. All of the linear ways and spindles are air bearings. I wouldn't think that you would need tolerances any better than about .0005 an inch for linear air bearings. I might be wrong. But the main improvemnt in my view is just eliminating all of the vibration by having all moving surfaces close tolerance and air bearing. His Lathe /grinder combo is not really laid out the best for productivity. But his system really works well.

Think how cheap you could make ultra high quality spindles if you had a grinder like that. You wouldn't even need to buy expensive roller and AC bearings.

I think you could come pretty close just by using EG, steel reinforcement for the bed and build the ways from stainless steel. It would needs lots of grinding and hand lapping . The cutting forces are light so stiffness is not going to be as important as dampening in my opinion. As long as it is pretty stiff.

I linked to where Gelbart got his parallels, it was Precision Granite USA in CA. I found the price list he linked to: http://precisiongraniteusa.com/pdffiles/PGPRICELIST.pdf You can buy a 4 face Laboratory Grade AA (+.000025") parallel that is 30" X 1.5" X 4 " for 964.00; ergo he has around 4000.00USD in all four of his. As I said, he claims to have about 10k total in his lathe, so 15,000 for a single parallel seems wayyyyy off. If the link is messed up, as they are want to do on CNCZone, let me know and I can PM you the link.

I would love to follow along with a replication of Dan's lathe. He has a few posts (5 I think) on Practical Machinist where he talks a little about the lathe, but the best info is from his comments on his YouTube videos.

Anyone have any feedback on embedding the parallels in a larger EG structure and using linear rails for a machine capable of larger cutting forces?