[Bowman]

Hi all. I was just trying to cut a small pocket with on my Sherline CNC mill with a 1/8" endmill running wide open @6 ipm with a .025" depth of cut, 33% stepover.

I notice what looks like lost steps and may have been but what really has me concerned is the way I could see the entire headstock deflect when cutting in the -x direction. It would get to the corner of the pocket and I could see the entire headstock pivot a substantial amount.

Is there something loose I am overlooking? I had the chips cleared so it wasn't build up. With the amount of motion I was surprised the cutter did not break being 1/8". Anyone else have this problem? It only seems to want to twist when cutter -X and I have not seen it happen with larger cutters taking deeper cuts.

Any input appreciated


Mojo

[escott76]

Use a DTI against the head to make sure you know where things are deflecting. Then you can see about the cause. Sounds like the tool is whipping though if it only happens with smaller tools and not larger ones. A DTI will confirm what is moving, start there.

[Bowman]

Thanks for the input but I don't need a indicator to tell me the headstock is moving. I can see it moving visually and on the workpiece in the cut area. The whole spindle/motor lifts or rather pivots up on the right hand side as the cutter moves left into the corner of the pocket. It is not just a tiny bit of movement that why I am wondering if something is not as tight as it should be or what.



Bo

[Crevice Reamer]

Hi Bo. Take your T handle Allen wrench and check tightness of the screws at column base left and head left side. (If these screws are loose--then your mill is probably out of tram.)

Also check the four screws that hold the column to the base.

CR.

[escott76]

Thanks for the input but I don't need a indicator to tell me the headstock is moving. I can see it moving visually and on the workpiece in the cut area. The whole spindle/motor lifts or rather pivots up on the right hand side as the cutter moves left into the corner of the pocket. It is not just a tiny bit of movement that why I am wondering if something is not as tight as it should be or what.



Bo

<Sigh> So why would this happen only on smaller tools and not larger ones which have more load? I'm not saying it doesn't happen, but you can't quantify where and how much without taking measurements.
I don't know if anyone here can tell you any better that something is loose any better than you investigating to see whats loose and making sure the mill is in proper adjustment and tram.

[Bowman]

I am going to tighten up everything I can and try to measure what deflection I can. It may very well be the L shaped motor mount bracket is a bit loose as well.

Will report back my findings hopefully something simple I am overlooking.



Bo

[ihavenofish]

<Sigh> So why would this happen only on smaller tools and not larger ones which have more load? I'm not saying it doesn't happen, but you can't quantify where and how much without taking measurements.

actually, a smaller tool "stalled" at the spindles maximum torque would exert more force on the workpiece than a larger tool. usually small tools arent pushed as hard, so you are generally correct, but in a stall situation, the small tool will "kick" more.

if the thing is visually "pivoting" though, they machine is either very very weak (unlikely cause home made mdf machines arent even that bad), or theres something very loose in the head/spindle/tool.

my kx1 which by most acounts is tiny and flexy will snap the cutting bit and rip an aluminium workpiece out of the vise long before exibiting visually noticable flex. in a situation like that (which has happened to me once) its going to flex .01" at most before something breaks (in my case, it pushed the workpiece out of its way). the sherline is alot "weaker" than a kx1, but so is the spindle. i really dont see this being a mechanical flex thing.

[Bowman]

Ok I measured the deflection off the spindle and while a bit jumpy on the dial it is deflectiong off .01". It appears when looking closely that the motor and spindle itself are the only things moving. Do the spindle bearings require tightening now and then? I got this mill 2nd hand and retro fit cnc to it myself so I am not intimately aware of all its issues.

Also I tried to cut again this time removing climb milling and cutting CW. Now when the cutter gets to the lower right corner and makes its move to the left the head flexes and stays that way until reaching the top left corner of the pocket. I also finally broke the cutter lol and slowed the feed down to 4 ipm.

So I tried again with a .25" endmill and still encounter the same problem. Although it does not appear that the column or saddle are moving at all. It is almost as if the spindle is loose in the headstock but if that were the case it would happen in all directions I would assume.

I am just getting the idea that this little mill has to take super light cuts to avoid these deflections and loosing steps. Man I can't wait till my daughter gets off my dime and out of college so I can get some real heavy equipment. This maybe the hobbiest idea of milling but I find it frustrating as all hell. You finally get one little issue resolved to only find another and another and in the end you still only have this little mill. I want to make parts not watch them being made countless hours or trying to figure out how to mod the mill to make it work right. I guess hobbiest like tinkering with the machine itself as much as cutting parts but man I would really like to just cut some parts and not worry with the rest

Have to hand it to those that can sit there for hours turning those little handwheels making those tiny cuts they have more patience than I do.


Bo

[ihavenofish]

Ok I measured the deflection off the spindle and while a bit jumpy on the dial it is deflectiong off .01". It appears when looking closely that the motor and spindle itself are the only things moving. Do the spindle bearings require tightening now and then? I got this mill 2nd hand and retro fit cnc to it myself so I am not intimately aware of all its issues.

Also I tried to cut again this time removing climb milling and cutting CW. Now when the cutter gets to the lower right corner and makes its move to the left the head flexes and stays that way until reaching the top left corner of the pocket. I also finally broke the cutter lol and slowed the feed down to 4 ipm.

So I tried again with a .25" endmill and still encounter the same problem. Although it does not appear that the column or saddle are moving at all. It is almost as if the spindle is loose in the headstock but if that were the case it would happen in all directions I would assume.

I am just getting the idea that this little mill has to take super light cuts to avoid these deflections and loosing steps. Man I can't wait till my daughter gets off my dime and out of college so I can get some real heavy equipment. This maybe the hobbiest idea of milling but I find it frustrating as all hell. You finally get one little issue resolved to only find another and another and in the end you still only have this little mill. I want to make parts not watch them being made countless hours or trying to figure out how to mod the mill to make it work right. I guess hobbiest like tinkering with the machine itself as much as cutting parts but man I would really like to just cut some parts and not worry with the rest

Have to hand it to those that can sit there for hours turning those little handwheels making those tiny cuts they have more patience than I do.


Bo

.01" of play in the system (as opposed to flex) should be "feelable" if you grab the spindle nose and try to shake it. you should go around poking at the spindle and head area to see what is loose. it could be bad bearings, or a loose mount.

.01" isnt alot of deflection, not something youd typically see with your eyes when its happening especially since it would be gradual. the sherline might easily flex that much in use especially in agressive climb milling, but the way you are making this sound, it seems to be alot more than simple flex.

make sure you are using correct feeds for the material and cutter though. if the spindle speed is only 2000rpm and your cutting metal with a 1/8" bit, 4ipm might actually be much too fast.

[vlmarshall]

Hmmm... I just tried a cut, full width of .250" endmill, .025" deep, 20IPM, and 3200 spindle speed, and didn't get much deflection at all.
I set up an indicator on the spindle body, with the magbase stuck on my Sherline's thin sheetmetal tool shelf, and still only saw about .005" deflection.
Here's a poor video of it.


I DO need to tune my new spindle card though, it doesn't seem to have the power that my old one had.

[Bowman]

vlmarshall - I measured the same way with mag base on the steel enclosure off the headstock but from the other side. When cutting in the same direction as you I was getting the .01" reading. Thanks for the video. Btw my enclosure vibrates more than yours for sure lol must not be as solid.

Wish I had a higher speed spindle. Just recently upgraded to the new Sherline motor from the old DC motor they used to use.

The fiddling continues I guess until I figure out how to get more deflection out of the cut.


Bo

[ihavenofish]

Hmmm... I just tried a cut, full width of .250" endmill, .025" deep, 20IPM, and 3200 spindle speed, and didn't get much deflection at all.
I set up an indicator on the spindle body, with the magbase stuck on my Sherline's thin sheetmetal tool shelf, and still only saw about .005" deflection.
Here's a poor video of it.


I DO need to tune my new spindle card though, it doesn't seem to have the power that my old one had.

on a full width cut, the primary deflection forces would be perpendicular to the cut as thats the direction the bite of each tooth is largest. if you measured the deflection in the y direction... or just ran the cut test in the Y youd see something very different i think.

[vlmarshall]

Yeah I tried to duplicate as much of your setup as I could, although I was indicating the spacer block between the spindle and the column. After running it again I found a better spot to set the indicator, and get even less vibration.

Check to see if you have the key(s) in the keyway(s) for the spindle mount. I'll bet that the head would deflect without them, even if the clamping setscrew were tight.

I hope you get it figured out easily, Sherline mills really aren't bad machines!

[vlmarshall]

on a full width cut, the primary deflection forces would be perpendicular to the cut as thats the direction the bite of each tooth is largest. if you measured the deflection in the y direction... or just ran the cut test in the Y youd see something very different i think.

Hmmm... good point, I didn't think about more than Bowman's test!

Here's one with the indicator on my Z stepper, measuring deflection in the Y direction. Still a .025" cut, full width of a .250 endmill, -X at 20IPM, 3200 RPM.



Quit laughing at my cheap indicator.

[Bowman]

After yet more fiddling I with larger cutters I am not seeing the deflection but I am now seeing what appears to be missed steps as each pass fails to go back to X+ before making the corner. So I am getting a stepped cut on the right side of the pocket of equal width with each pass. That in turn is causing the cutter to run into and cut the entire depth of the sidewall of the pocket on the left side.

And here I thought I had just gotten the resonance issues resolved and had everything cutting accurately enough for my needs. Now I am back to trying to get rid of the lost steps. Perhaps some motor tuning is in order since I added the flywheel dampeners. Even checked the backlash again and it was the usual .004" in X. Was trying backlash comp on the Z since I have .01" on it in backlash and no adjustment lever like the newer models. Then I wondered it constant velocity was causing issues and tried turning it off but no dice still not cutting as it should.

Oh to have a fullsize machine.. Dreamssssss of something like the IH mill are filling my head.

I know the Sherline is capable of many things I am just not seeming to have much luck getting mine to perform any of them. Or should I say I think I have it for a few cuts that look promising and then all of a sudden something is out of whack causing trouble. Maybe I am just expecting to much from the little bugger but I don't think so when I read others saying what they cut and how well it does. Mine doesn't have the latest Sherline stepper mounts with the bearings etc it is just oldham couplers and all the backlash adjustments are the old flavor as well. Thankfully I only have $600-700 tied up in the whole mill including the retro fit CNC components, if I had spent 2 grand and had all these issues I would be really disappointed.

The other day I was cutting with a .375" endmill at a depth of .05" per pass and 6 ipm feed and everything cut nicely. Today is another story. At little consistency is all I am looking for, I will go easy on it but the cutter has to come into contact with the metal to make a part and thats where things go bad.



Bo

[vlmarshall]

Hmmm... mine has the older mounts, leadscrews, and backlash adjusters as well. D&M used manual Sherlines for their conversions, so I didn't even have thrust bearings on the leadscrews to begin with!

I have noticed that if I forget to lube the table and column ways with Teflon spray, I can get some missed steps, at least while rapiding around.

I guess that being able to do my heavy stuff at work takes a lot of the stress, and expectation, off of my Sherline.

[vlmarshall]

Speaking of the lack of thrust bearings, when I discovered it, my X coupler had galled and welded itself to the thrust plate. If your mill has the manual parts as well, check your leadscrew friction. It could be the cause of your missed steps, especially if the gibs are a bit too tight to try and make up for backlash that is actually in the leadscrew.


I didn't want to lose X travel in my mill enclosure, so switching to the Sherline stepper mounts was not an option. I used some miniature bearings from MSC, .250" ID, .375" OD, in counterbores on either side of the thrust plate/leadscrew mount. A few little modifications and it worked well.

[ihavenofish]

so i did a little math, and you(vlmarshall) are pushing .0032" per tooth assuming the end mill is a 2 flute. thats way to agressive and is probably flexing the bit and maybe head from pushing the bit through as much as it is from cutting torque.

even still though, thats using approximately .04hp if its aluminium, and resulting in 0.79 lb-in of torque, or on a 1/4" bit, about 6.3 lbs of lateral force averaged in all directions. probably 50% in the +y direction in your cut and the +-x directions canceled out. with a 2 flute, all the force is always in one direction at a time, so my numbers mean that its full load in the +y for 50% of the time, not that theres only 50% of the load. thats not much force at all though, and of course the machine doesnt appear to be flexing much in the videos - just a few spikes to .004ish here and there and that could be indicator bounce error.

Bowman is cutting at a more reasonable feed though 6ipm is still on the fast side for a 1/8" cutter especially if he doesnt have 3200rpm. at 2000rpm (stock sherline speed i think) a 2 flute end mill should be running 3 to 4 ipm and creating virtually zero cutting loads (under 2 lbs averaged in all directions).

basically what im getting at is.. there must be something loose in bowmans machine to be causeing vissible motion in the head.

[vlmarshall]

so i did a little math, and you(vlmarshall) are pushing .0032" per tooth assuming the end mill is a 2 flute. thats way to agressive and is probably flexing the bit and maybe head from pushing the bit through as much as it is from cutting torque.

Yeah, it's a .003 chip load, but I don't think it's flexing a carbide endmill.
I started at Bowman's full spindle speed and 6IPM, and increased the feed to stop the chatter.

Probably too aggressive, but like I've mentioned elsewhere, my biggest problem is coming home and getting out of 'production' thinking habits that I've had longer than this D&M Sherline.

Your force-vector physics stuff is interesting, thanks for that cool bit of insight!. HP requirements and material removal rates are about as much math as I apply to endmill physics... and I use a chart for that!

[ihavenofish]

Yeah, it's a .003 chip load, but I don't think it's flexing a carbide endmill.
I started at Bowman's full spindle speed and 6IPM, and increased the feed to stop the chatter.

ah, carbide might be better with that kind of feed, not sure, but what i meant was that the forward feed force might be outweighing the cutting force in terms of deflection causes. most 1/4" bits ive seen are recomended for .0015 feed, but ive seen specialized ones that allow up to .04" in aluminium! i ran a 1/4" m42 3 flute bit in aluminium at .007" feed per tooth (5krpm, 100ipm) and it cut fine with minimal chatter. tried the same with an hss 1/2" bit and one of the tips chipped off.