Lathe turret loading strategy?

[Donkey Hotey]

I'm in the process of setting up a new SL-20 for an engineering prototyping lab. My desire/intent is to load the machine with a bank of 'standard' tools that stay in the machine (about half the turret positions).

As an example, on a mill, I like to leave the probe in the last pocket and four drill chucks in the next pockets down (19,18,17,16). Whenever I program an new part, I always know that the first drill is 19, the second is 18, etc.

I was going to use a similar logic on the SL-20 but, while cleaning all the cosmoline off this morning, I started to realize that boring holders in adjacent, non-current tool positions are still in serious danger of clanking into the chuck.

Where do others load the tools when they need 3-4 boring type tools held in a lathe turret? At least: a drill holder and a tap holder (ER-16 or ER32 depending on size) and a dedicated position for a boring bar (it won't always be loaded but, you'd always know it was going in 'that' pocket).

[MadMax]

What kind of parts do you typically do? Or perhaps more to the point: what are you typical "difficult" parts?

If you do a lot of deep boring with difficult to clear length to dia ratio, chip clearance and headstock clearance will become a more dominant issue. You may want to alternate between long and short tools to keep your tools clear of the headstock.

I find that I have been getting very good use out of double tool gang holders. If you can afford the space, you can put a lot of tools on a changer with gang holders. Gang tools are excellent for pilot-drill applications because you double up on a station and it saves time for toolchanging.

I had initially hoped to keep half of my tools the same in my machine, but I find that interference is just difficult to work around unless you have a machine that is oversized for your typical job. Instead of setting up for a generic range of tools, we have been trying to keep the tools that are the most time consuming to set up in the changer. We try to leave long boring bars which have to be accurately set up for height (Y direction) in the changer and swap out the short turning stuff and live tools (which have to be changed anyways) instead because setup on one of the long termpermental boring bars (6") protrusion takes longer than moving in and touching off a few short turning tools (which aren't very fussy).

[Geof]

Don't put a drill adjacent to a parting tool unless you want a very short drill and a lot of noise.

[Donkey Hotey]

What kind of parts do you typically do? Or perhaps more to the point: what are you typical "difficult" parts?

We don't have 'typical parts' and that's the challenge. I would like to set up the Mastercam library with the standard tools and always start from there. Maybe I'll just 'mentally' allocate the pockets but, not bolt the holder in there until it's needed.

Don't put a drill adjacent to a parting tool unless you want a very short drill and a lot of noise.

Yeah, that's exactly what has me worried. I actually removed all the boring holders except one and that still has me worried. I'm not going to be the only one running the machine and if somebody doesn't move clear of the chuck or fat-fingers the 'turret fwd' button at the wrong time, it's going to cause some damage.

[MadMax]

Unfortunately there's not much you can do. I used to think that I could have set up a more or less generic tool set on my turret, but part after part I find there are just too many clearance issues to deal with. If you had a machine that was bigger than you needed for most of your parts, you could probably use a more generic tool set, but my parts just barely fit into my machine. I've actually removed my subspindle chuck and altered the parameters to push it back further. I've squeezed out every extra inch I can out of my machine.

About the only thing I can do now is get a collet chuck which would have a lot more nose clearance, but a shorter collet chuck might have me banging the wall at the mainspindle with some of my longer tools. Some of my tools clear the spindle wall by about 0.400" and my turret comes about 0.05" from my chuckless sub spindle pushed and extra inch or so back. I wish I had gotten a 230 sized lathe instead of my Puma 200, but we're still getting production out at least.

Unless you've got a larger than necessary machine and a low profile collet chuck, I think you're just going to have to be careful with tool changer setups that are unfortunately variable. CNC lathe work is an expert users tool. You've got to be diligent about things because they're so crash prone (and destructively so). Mills can crash, but you typically don't have anything nearly as energetic as an 8" chuck with huge grabby jaws whirling around at 5000rpm. The tool crib is easier to maintain because you don't have to have the whole thing moving around close to the workpiece at the same time.

About the only generic tools that I've managed to keep in my turret are my combo bar pull and parting tool, and an outside turning tool. Both of these tools are not difficult to set up so they really don't save me much by remaining in the same place all the time.

About four other stations are not usually changed because they're boring bars that are annoying to set up, but they're only useful for one particular part. The rest of the stations move around between setups, either because they get in the way of things, or because they're live tools which have to be changed for different parts anyways.

For prototyping simple one off parts, or rework, I tend to find that it's often much faster to run them on my manual lathe which doesn't have the same tryout issues as the CNC beast. Sure it's annoying to do curves or off angle chamfers, but I find that prototyping is usually fraught with a lot of niggling issues like poor chip clearance that can destroy a tool that's being driven along blindly with a CNC machine.

I'm actually puttering away retrofitting a bit of a junky Syil C6 CNC lathe. There's something really nice about a lathe which doesn't have very fast rapid rates and a manual wedge post. It's not powerful enough to terribly destroy itself, and the tool crib is very easy to manage because there's no turret bristling with chuck snagging protrusions. For one off parts, super fast tool change isn't a big benefit, so I'm looking forward to a cheap and cheerful Fisher Price lego lathe to protoype with. You can also run a huge tool crib that's all touched in with a manual tool post.

[Donkey Hotey]

CNC lathe work is an expert users tool. You've got to be diligent about things because they're so crash prone (and destructively so). Mills can crash, but you typically don't have anything nearly as energetic as an 8" chuck with huge grabby jaws whirling around at 5000rpm.

I was going to send a link to this thread after reading that, then I got to this:

For prototyping simple one off parts, or rework, I tend to find that it's often much faster to run them on my manual lathe which doesn't have the same tryout issues as the CNC beast.

That would kind of wreck it all for me; I just talked them into this thing.

I fear that what you're saying is very true. Actually, I find the same thing true of the TL-1 and (with a single toolpost) I don't have the extra tools to worry about. Well, I do have a manual Dorian turret I haven't installed yet...

We still have a 13x40 manual Acra lathe for a basic turning or facing operation.

In our case 'prototype' means we're making one assembly of something, not one part. That assembly might have five of some turned part and it might have fifty (or five-hundred ). They're usually precise enough or weird enough in shape to make manual parts too costly and difficult.

I'm guessing most parts will be around an inch in diameter and will be close to the chuck. We got a 5C drawbar closer for general small stuff and a 3J dead-length for the 1-2" parts or the parts with higher accuracy length requirements.

I really wish there were soft stops you could set in the control (one per tool number--in the offset table). You could dial the tool changer right up to the collision point and 'origin' that location in the offset table. Anything in that range would error-out and give you an over-travel alarm.

[Vern Smith]

I guess there are always multiple ways to skin the cat but I took a little different approach to the same type of problem. Keep in mind that this is my first CNC lathe so most of this is probably old hat to you guys but I doubt if many of you have the travel limitations I have to deal with. The GT-10 has a total travel in Z of 8” and that is to the face of the supplied power chuck (pictured). Whatever the heights of the jaws are, you subtract that from the 8 inches. I’vie attached a picture of the machine with the turret fully retracted in Z and the supplied chuck installed with the chuck makers standard hard jaws. The center to center distance between adjacent mounting holes in the external holders is 3.5” so problems with adjacent tools holders banging into something gets problematic very quickly.

I opted to pass on the $9,000.00 tool eye option and came up with a system to set up tools to a common length and be able to establish the common length measuring point no matter which of the three chucking deceives I have in the machine. The red standoff pictured on the supplied power chuck is part of that system. It has a 4 lb pull magnet on the bottom and extends to the common tool setting length, which also happens to be the very front of the 5C collet chuck. I have a manual chuck not shown that has its own stand off that also extents to the standard tool measure length. The genesis for a lot of this came from other posts on this forum and contributions from forum moderators and resident Guru’s

The GT is a lot more entertainment than revenue generator so 100 plus part runs happen once a week at most, the rest of the time I want to use it like I have used a small engine lather for 40 years. As Donkey is lamenting, it’s easy to get turret collisions on the SL-20 which are magnified on my “baby” as one of the more prominent forum contributors refers to it. So assuming the safest way to go is not carry a bunch of internal ( is this the proper nomenclature?) tools in the turret on a recurring basis I set out to develop a system where most of my tools were in holders that would return to the same tool position register every time they were put in the turret. Simple stops in both X and Z were installed. Photos are attached. My ER 25 collet holder’s protruded well over an inch from the face of the turret holder reducing the usable length of any drill/bar. I made up a bunch of holders and bored out the front of two of them so I could recess the collet nut into the holder. You can see it mounted in the turret in one of the photos. The ER16 holders were less than 1” ID so they will slide into the tool holders with out modification.

Each tool is measured to the common Z length position and the numbers are recorded along with the X setting. Each tool and its length registered holder is marked and numbered. Haas provides 50 tool number positions so I use those above the 8 turret positions to retain the X and Z positional readings for each tool. All the Z positions are in machine coordinates. As a Newbie operator and well beyond the optimal age to be operating a 700 ipm machine I’m paranoid about crashing the little monster. All of the above I hoped would help keep the inevitable from happening.

Using the common tool length measuring point makes establishing a Gxx offset for each particular part type necessary. I use CAM software because I’m not proficient in G code so the Gxx positioning is necessary anyhow. There are two ways to do it, the traditional (I think) procedure using the end of the stock with tool #1 and the Z measure offset button. The other is to position tool #1 on the front of the chuck jaws, do the Z measure offset button, and manually change the result by the largest Z number in your program plus your comfort margin.

This gives me the capability to stick any of my tools in the turret, enter the Z and X numbers from further down the list into the turret position the numbered tool happens to occupy,, and have confidence that things will be where they are supposed to be and offsets properly established. Then I run at 5% to keep my pants pristine.

I know most of you have wrestled with all these problems many times and I would appreciate any suggestions you may have to improve my simplistic approach to the problem.

[Geof]

.......I really wish there were soft stops you could set in the control (one per tool number--in the offset table). You could dial the tool changer right up to the collision point and 'origin' that location in the offset table. Anything in that range would error-out and give you an over-travel alarm.

This would be a good idea and it should be possible to implement it fairly easily; Haas-Apps, hint, hint.

So far in this thread I have been pre-empted twice; Madmax mentions that an SL machine is not really conducive to one-off prototypes and Vern gives a solution. Thank you Vern; I was thinking about writing something not as involved as yours while at dinner, now it is not needed.

Here is one additional hint which can make any crashes far less damaging: as far as possible program the rapid move into position with the chuck stationary. Ramming a tool, even under rapids, into a stationary chuck is far less damaging than having it encounter all that rotational kinetic energy.

And another hint; learn how to interpret the graphics so that you know what a safe envelope for the trace is on the screen, and use the machine coordinate display to explore possible interference. I often write down the most negative Z machine positions for each tool and then jog to these points and spin the chuck by hand to check clearances

All these procedures add time to getting a setup running or add time to a cycle by stopping the chuck at every toolchange, but intially they are a good insurance policy. At the business in our early CNC days we did have two crashes that needed turret re-alignments but nothing more serious and maybe half a dozen broken or bent drill bits or boring bars; you do your best to try and make sure the inevitable is going to be as mild as possible.

[Geof]

..........I know most of you have wrestled with all these problems many times and I would appreciate any suggestions you may have to improve my simplistic approach to the problem.

For What Its Worth I don't think it is simplistic I think it is elegant and I wish I could get myself energized and organized enough to do something similar.

[Donkey Hotey]

Vern, thank you for that detailed write-up on how you run your GT-20, and thanks to everybody for the tips. It looks like the best thing is to keep the boring holders out of the machine.

I'm not so worried about myself (though that's still a concern). I'm supposed to be the 'super user' of this new lab we're setting up and any crashes will probably be blamed on me for not providing adequate training or documentation. I'm trying to out-guess all the things that could happen and trying to carefully choose the tooling and write procedures to minimize the possibility of crashes.

[MadMax]

Vern has a good suggestion if you typically make smaller parts close to the chuck. Having a common Z offset on all of your tools would allow you to set a travel parameter limit which would prevent crashes with the headstock. I once did something similar for my manual lathe before I figure out to program different offsets on my DRO.

As geoff points out, almost all of the destructive potential in a crash pretty much comes from a rapidly spinning, very heavy chuck. Kinetic energy is directly related to moving mass and the SQUARE of velocity or rotation rate of said mass, so you can appreciate how much energy is embodied in something that's quite heavy and spinning at 5krpm.

A rapid feed into a stationary headstock would require a change of insert and realignment of your turret. The same feed into a 6" chuck flying at 5krpm can crack the bed of your machine and put all of the bits that still work on Ebay.

It's a plus and a minus that you're going for a Haas machine. A moderate crash can spall a linear ball slide which are not as impact tolerant as a ground boxway. however, a linear slide is much more easily replaceable than regrinding a boxway. I've heard that Haas interfaces are more user friendly than Fanuc interfaces which might help you reduce crashes too.

If you can afford to set all of your tools to the same offset (i.e. typically short parts), than I suggest the following measures:

-Z travel limit preventing tools from crashing into jaws
-reduce max rapid feed, and cutting speeds through parameter adjustment
-use softjaws

If you can set all of your tools to the same physical offset, then you can prevent any headstock crashes with a parameter adjustment. If you can't, you could limit max rapid and cutting feeds so your tools would end up cutting the soft jaws instead of slamming into them which would be a lot less destructive.

I'm not a master operator myself. I'm quite dependent on my tool maker who does my manual G code programming and most of the tool setup. All I'm really useful for is helping to troubleshoot difficult operation plans (fixing nasty chip nesting, etc). All of my ideas pretty much come from the measures I take myself when we were getting used to our lathe. I made parameter adjustments so we'd accidentally machine our jaws instead of destroying things.

I strongly suggest you keep amateurs from using your new lathe without close expert supervision. These things can be very dangerous to those who lack discipline. It is not difficult to override things like door interlocks (tryout mode etc) and reach in when you want to pull a nest off of a tool when you think the lathe isn't going to move onto another operation. Two asshats operating the same CNC lathe is an even more dangerous situation. Be very careful allowing engineers who think they're technically proficient to use your equipment. If possible, review every operation plan that is to be run on your lathe. It should be possible for you to check a tool call and see if any moves would take it into a crash, possibly even with the help of a validation program that you could write in C or php.

I once tutored a class for mech eng students which involved writing a G code program for a plasma cutter. Making sure all moves were in the safe cutting zone was a simple text validation (one tool) but I think it should be simple to use tool calls to determine safe move envelopes.

[Geof]

.....I really wish there were soft stops you could set in the control (one per tool number--in the offset table). You could dial the tool changer right up to the collision point and 'origin' that location in the offset table. Anything in that range would error-out and give you an over-travel alarm.

I had second thoughts about this being a good idea, or even practical.

It should be possible to implement soft stops for the active tool but that is not the important one(s); these are the (longer) tools on the adjacent turret locations, especially the rear tool. Soft stops could even give a false feeling of security because you may be tempted to set and forget them and then blithely install a long tool in the adjacent location.