What's new
What's new

Chatter with 12" Craftsman lathe

Status
Not open for further replies.

rhoward

Hot Rolled
Joined
Apr 10, 2003
Location
Everett, WA, USA
Hi all,

I have a very nice 12X36" Craftsman lathe that I pretty well restored from the ground up. New spindle bearings, new belts, spiffy paint etc. I fitted it with an AXA style QC post and bought some very nice Kennametal carbide insert tooling designed for a Swiss CNC machine. I also fitted it with a 1hp DC motor and speed controller. It is beautiful!

Trouble is that I never use it because it chatters so badly. I have tried everything I can think of but even light cuts in soft steel produce a terrible finish. And yes I have done all the obvious things, nice and tight spindle bearings, tight gibs and tool on center etc. It chatters with carbide inserts, cemented carbide, and HSS bits equally well!

Two possibilities I have noticed may contribute to this problem.
1. The DC motor is pretty heavy and since it is cantilevered off the back of the machine it may contribute to flexing of the bed. I have tried several brands of V belts but there is always some vibration and bouncing going on.

2. I made a long taper attachment for the machine using Bishop Wisecarver V wheels and track. When I tested the attachment I was surprised to see that the chatter problem was greatly reduced. Hmm. Maybe my cross-feed nut and shaft are too loose? I have run several badly worn lathes before with terribly sloppy cross-feeds and not experienced such bad chatter as this though.

So what do you all think? I know this is a light machine but have heard that the Craftsman/Atlas should do better than this.

Randy
 
Randy,

How did it cut before all the repair work you did?

Brass is pretty forgiving to machine. Same symptoms?

How's your workholding? Can you try cutting between centers with a faceplate and dog to eliminate chuck issues?

Is the machine sitting rigidly on the floor? Try bolting it down through some anti-vibration machine pads.

When it chatters, do you physically feel it in the compound? Cross-slide? Bedways? Headstock? Motor? Machine base?
 
You could try some incremental things and see if any help. Like lock up cross slide gib screws. Does chatter diminish? With them still locked, grab a flat way in each hand and jamb your thumbs against front of carriage as it cuts. Does chatter diminish? Some of this may serve to give an idea of what is what. If none of the above does anything, I would strongly suspect spindle bearings. You may have missed something...

John
 
No, I have not cut between centers but I have cut with a chuck and live center. Same story. I have put an indicator on the chuck and pryed the body with a bar between the bed and chuck. Close to zero reading.

Yes, when it chatters it definitely is felt in the tool post and carriage. If left unattended it will back off the feed screw. I have tightened up the gibs to the point of making it difficult to even turn the feed screws and it still chatters.

The spindle bearings are perfect. I have tightened them with proper preload so that they run a little warm when spun up at high speed but do not get dangerously hot.

Could the chuck be bad? It seems unlikely as it is a nice looking unit. It locks the work well and has stout two piece jaws that are reversible.

It almost seems like a casting is cracked but I have examined it closely and can find nothing amiss. This is very frustrating!

Randy
 
"Yes, when it chatters it definitely is felt in the tool post and carriage. If left unattended it will back off the feed screw. I have tightened up the gibs to the point of making it difficult to even turn the feed screws and it still chatters."

Seems to me the cross slide screw and/or nut are sloppy to the point they cant handle the cutting load any more. I may be wrong, but the gibs and dovetails serve to keep the compound from moving side-to-side and lifting up, while it's the helix of the screw that resists the thrust from cutting pressure.

Do you get the same chatter when the carriage is locked, and you are taking facing cuts from either direction? If not, it's prolly because now you are "controlling" the backlash in the screw either by hand feeding or powerfeeding (if the machine has this).
 
Yes, facing is much better but not exactly super. The bad thing about the Atlas of this model is that it has a coarse feed screw for crossfeed. It reads actual travel of the tool point instead of thousandths off the turned diameter as is more usual. That means when you advance the crossfeed dial .001 it actually takes .002 off the diameter. A lousy setup IMHO. That also means that the screw has less of a mechanical advantage and cutting forces can back it off much more easily. So maybe it is the screw and nut? There is no provision for takeup on the nut. I wish I could replace the whole screw, nut, and dial with something decent that reads in .001 of turned diameter.

Randy
 
Randy,
I don't know about your experience level, so forget about this if it doesn't apply.

Have you considered your tool geometry? When I was first learning (on a 12" Atlas lathe, in fact), I remember reading in the books that the pictured shape of the recommended cutters always had a very large rounded nose. When I tried this, I had nothing but chatter - too much surface area contacting the material. Once I tried a sharp-cornered tool I had a lot more success. Ideally, a slight radius on the tip will stay sharp longer (especially with carbide), but a truly pointed tip will still do the job. If I'm making a cutter, I grind it to a sharp point and just hone a slight radius with a few swipes of a whet stone.

If you haven't already, I would experiment with tool geometry and see what works for your lathe. I found my Atlas tended to be more "chattery" than any lathe I've used since, but I made a bunch of stuff while I had it.
 
I look at it like this. Try to eliminate one possiblity at a time. Since you say the spindle bearings are good, take the chuck out of the equation. Go to a four jaw, or a face plate.
Inspect the taper, or in your case the thread for the chuck while you have it off.
Cut some brass.
Does chatter still exist?
Try to remove the BXA tool post. Go for a rocker style.
Try your cut again.
Try to use the tailstock.
Try without.
Try different tooling.
Try to cut in reverse. (away from the chuck)
Try to cut in reverse gear. (tool upside down, or on the back side)
If none of these work to eliminate the chatter,
then buy a Monarch! LOL :D
That is all I can think of right off hand.
Doug.
 
Yep, tool geometry is important alright as cutting forces go waaay up with just a small increase in tool point radius. Nope, that isn't it. I have even tried zero radius HSS cutting tools with poor results.

Randy
 
Randy,
The pitch of the screw may be the same whether reading radial or diametral travel. Sometimes the builder just cheaps out with a smaller dial where they couldnt make the graduations spaced closely enough. I worked for years with a Powematic lathe like this. Kind of a PITA, but you get used to it.

Doug is on the right track: change one thing at a time. And pull out the screw and nut and have a look-see for excessive wear or damage.
 
Hmmm. Sounds like you're stuck with process of elimination. You might set a dial indicator on one of the ways with the tip against your tool and try to wiggle different parts from your tool down to the carraige.

I remember thinking that damned taper spud arrangement for mounting the compund to the carraige was a constant annoyance for me. A suspect for the weak link for sure.

Good luck.
 
Ah yes, thank you Mr. Antiquos for pointing out the screw pitch / small dial relationship. This machine has an incredibly small dial that is almost worthless to see.

Okay, I see the trend. Do one thing at a time and be scientific about it. And here I was hoping for some magic fairy dust or something quick to fix the problem.

Thanks to all for the many good suggestions, it has helped to get me thinking again!

Randy
 
I once had a little Atlas 6" lathe that had a lot of the same problems you are describing. Tried everything that you have tried. Nothing worked. A bit later I bought a 9" South Bend that needed a bit of scraping on the cross slide, which was my introduction to scraping. On the off chance, I spotted the Atlas' cross slide on the surface plate. Brand new cross slide was only spotting at 3 corners. I scraped it, and the cross slide ways in, and never had any more trouble.
Harry
 
This may be very simple but can easily be overlooked. You say you fitted it with a new toolpost. Sometimes the bolt that holds it down could be bottomed out so it feels tight when you tighten the bolt, but the tool post may still not be clamped to the cross slide.
 
Here's a story about my Chinese lathe, 13" x 24", that will reinforce what everyone says about such machines. I have always had a terrible time with chatter when parting off. Recently I was turning a piece of 1" drill rod that was sticking out of the 8" 4-jaw only about an inch and a half. I could get a mirror smooth finish except for a barber-pole like spiral groove with a pitch of about 1/4". Changing tools, speeds, cuts, feed, etc. did not eliminate it but did change the pitch. Finally got disgusted and slept on it. The next day I decided to stick in a live center. Totally eliminated the problem.

I am now just about totally convinced that in an attempt to get good specs, in particular a large spindle bore, they made the thickness of the spindle wall far too low so that it flexes when rotating and any cutting force at all is applied. I have been thinking about turning a #4 morse taper with a 1" rod on the end and using a draw bolt to lock it in the spindle. This should provide some reinforcement when I don't need the bore. Does anyone think this would be time well spent? (I know; the better answer is get a new lathe.)

I don't know if the Craftsman problem is related to this or not, but it is worth considering.
 
Randy

My experience with a 10" Atlas lathe of similar construction was this:

The crossfeed screw which originally was Acme thread, was worn to the point of looking like a sharp V thread in the sweet spot where it had seen a lot of use. Of course this was after a boatload of agony.

I would first see how much backlash you have in the cross-slide screw, then take apart the cross slide and have a look for yourself.

My problems were that as I turned I'd get little "burrs" which came up on the turned surface under anything but the lightest cuts. The second problem was that I could not part a piece off without breaking off the parting tool.

My reasoning after seeing the condition of the thread was that the cross-nut was bearing only on the tips of the sharp V thread which were quite compliant, the resultant flexing let the tool point "skip" over a bit of metal when the pressure got too great, and when parting off, the tool simply dug in and ate up all the backlash in a split-second, causing the tool to break off, workpiece to bend, etc.

Hope this might help.

-Matt

PS: I think I would also have a look at the compound screw and nut while you are at it.
 
Thanks for the info Matt.

My lathe is exhibiting the same problem, a wooly finish on the part that has lots of tiny burrs on it. I looked at the cross feed screw and nut and they are not worn looking and still have nice square threads. I measured end play of the crossfeed carriage and it measures .017 when I physically push and pull on it. That sounds like a lot but I am still skeptical that it is the real problem. Even a new nut and screw will have some appreciable end play.

I measured deflection of the spindle near the headstock by prying with a crowbar. Most I could get was .001 with a fair amount of force. I next measured deflection at the work with a 1" piece of round stock chucked up. Here I got .005 which seemed like a lot.

So I swapped chucks and put on a small 5" one that I got with the machine. It is old and wimpy looking with tiny jaws but I immediately had better luck turning with it! Gone is the awful harmonic that caused chatter and now surface finish seems to be more of an issue of tooling and material. It still is not great but is very much improved.

So I guess I need to look at my 6" chuck and see where I went wrong. It is a quality Japanese unit that was intended to be "D" mounted with three pins. I removed the pins and machined a cast backing plate for it and attached it by the three threaded pin holes. I guess the system is too flexible so maybe I will see if I can contact more of the chuck backside with the backing plate.

Hmm. Should I counterbore a shallow face where the backing plate meets the shoulder of the spindle?

Hey, I am making progress! Thanks for all the suggestions.

Randy
 
rhoward:

The D back on the problematic chuch was made to pull up hard on the taper in the shallow counterbore (and adjacent outer face). If you are not accomodating this, the next best thing is a larger adapter plate with a boss fit into the chuck on as large a diameter as is possible

John
 
Right John.

Now I am wondering about how the 1-1/2 thread is supposed to align the backing plate/chuck to the spindle. The way I have it now is that the backing plate simply threads on until it bottoms on the shoulder of the headstock spindle. I guess with this method the threads just sort of self center (somehow). Is it intended to work this way or should a close-fitting counterbore be machined into the face plate that matches the major spindle O.D. so that this surface actually does the centering and load bearing when the face plate is screwed home?

Randy
 
Status
Not open for further replies.








 
Back
Top