PDA

View Full Version : Chatter in my Craftsman Atlas lathe



k2man
10-12-2014, 10:40 PM
I just got my lathe up and running. I'm turning some aluminum, and I'm getting some chatter. I noticed that the v belt pulleys are not tight to the spindle shaft. There is a small amount of play - it moves maybe or 1 degree of rotation.

Can I tighten the pulleys onto the shaft? How?


Sent from my iPhone using Tapatalk

doctordoctor
10-12-2014, 10:45 PM
I just got my lathe up and running. I'm turning some aluminum, and I'm getting some chatter. I noticed that the v belt pulleys are not tight to the spindle shaft. There is a small amount of play - it moves maybe or 1 degree of rotation.

Can I tighten the pulleys onto the shaft? How?


Sent from my iPhone using Tapatalk

There are about a trillion things that can cause chatter, especially if you are new to lathes (are you?), and especially if its a new-to-you ancient Atlas lathe. I just got an Atlas 10f as well, and had to do some things to get good finish.

Please describe everything about your setup in excruciating detail, with pics if possible. WE WILL OVERCOME

k2man
10-12-2014, 10:48 PM
Thanks drdr! Here's a video. I'm new to turning metal, and this is the first project on my new to me lathe.

https://m.youtube.com/watch?v=hSXAmWi5k90


Sent from my iPhone using Tapatalk

k2man
10-12-2014, 11:00 PM
In the video, I have a carbide tool borrowed from my neighbor. I was using a HSS 5% Cobalt bit that I ground properly (undercut so only contact was the point). Set the tool height right on center. Cutting only .015 or less per pass. Was running at slowest speed, moved spindle belt up to fastest pair, motor belt still on low speed pair. I do not know the alloy of the aluminum. I first cut 6" off of a 30" piece of tube using a ⅛" parting tool. Seemed to cut ok, but took forever (1 hr +). I was cutting at slowest speed though. I tried to grind the cutting tool with some undercut on the sides, but found that to be pretty difficult hand held on an 8"'grinding wheel. Got it cut ok, don't remember the chatter then, but as I said, this is my very first lathe project.


Sent from my iPhone using Tapatalk

doctordoctor
10-12-2014, 11:03 PM
Thanks drdr! Here's a video. I'm new to turning metal, and this is the first project on my new to me lathe.

https://m.youtube.com/watch?v=hSXAmWi5k90


Sent from my iPhone using Tapatalk

If you are brand new to running a lathe on metal, there are quite a few things you need to learn to make a successful cut. Turning the inside of a 4" aluminum tube is probably at the highest level of difficulty possible for a beginner.

I would put the tube aside, get some round brass, plastic, or aluminum bar, and try turning the outside of it.

Get ready to take pictures and upload them somewhere like photobucket so you can post links to them in here and we can help you. Videos are too slow too upload and too slow to watch most of the time.

Where did you come up with the RPM for that tube? Is it just a guess? Because it looks wayyyyy too fast.

Gary Paine
10-12-2014, 11:05 PM
The looseness you are seeing is because the pulley and gear/spindle are connected only through the drive pin on the right side of that big gear. Pull the pin out and the two will spin freely. That will allow you to throw the back gear into engagement. I do not think that is the cause of the chatter.

That large workpiece is sticking WAY out of the chuck. The cut is out from the headstock further than the distance between the two headstock bearings. ANY looseness in bearings, gibs, and so on will be highly conducive to the kind of chatter. You may improve things by tightening up the tapered roller bearings a bit. Easy process, but follow the directions so you do not have them too tight.

Best bet by far is to rig up some kind of a steady rest at the outboard end of the work.

Doozer
10-12-2014, 11:23 PM
I watched the video.
You can actually hear the toolbit rub.
When doing ID work, set the tool ABOVE center.

The pulley does not really have backlash, it is just the
pin mechanism to disconnect the pulley in order to
run in back gear. Very normal on Atlas.
--Doozer

J Tiers
10-12-2014, 11:27 PM
Well, I was waiting for that thing to jump right off the chuck.

But per the chatter, I didn't hear the traditional screaming chatter, you'd have to look at the cut to see if it is scalloped.

But it is reasonably normal with a tool stuck out that far, to have chatter when the width of the cut is increased, such as when your ten thou cut suddenly gets much wider/deeper after you hit the back of the "pocket" you are cutting. Common with boring bars, and a good reason for having the biggest one that fits in the hole.

As for the stickout, and why I was expecting the part to come off the chuck.....

The ID seems small, the OD large, and you are cutting near the OD. Lots of leverage relative to the "hold". Far better would be the OTHER type jaws, the ones that are stepped down as they go in vs the ones stepped down towards the OD of the chuck. With that type the "hold" is on teh OD of the part.

The term "stickout" has a lot of meanings. The two I like best are:

1) distance out on the part from the chuck, vs diameter of the part. After 3 or 4 diameters it gets to being a bit iffy. For larger diameters , it really should be the distance from the nose bearing of the spindle to the end of the work. It's the unsupported length of the part, vs diameter. I include the chuck when the part is larger than the spindle diameter, because teh spindle is more likely to be sprung by the work.

2) the distance out on the work vs the diameter of the "hold" you have on it. This is the same as #1 for holding on the OD of the part, but is far more descriptive of the situation for parts where the "hold" is on the ID of the part, as in your case.

Both of these have to do with leverage. The leverage the tool has against the stiffness of the part, and the security of the hold the chuck has on it. The best "hold" is nearly always with a 4 jaw chuck, and they only need one set of jaws to do either and OD hold or an ID hold. The jaws reverse easily.

When teh part is flexible and the "stickout" is long, the part can flex. This may mean it pushes away from the tool, and is NOT cut as much at the end, giving a taper. It can ALSO mean that it is "sucked in" by the cut, making it deeper, so that you end up with either chatter, or in bad cases, the work "catching", bending and causing a "crash".

In your case I fully expected to see the part pulled off the jaws by the cut, due to the smaller diameter of the "chuck hold" vs the longish part and large diameter of the cut. When the tool hits the back of the pocket, the forces go up a lot, very suddenly.

doctordoctor
10-13-2014, 12:00 AM
Well, I was waiting for that thing to jump right off the chuck.

But per the chatter, I didn't hear the traditional screaming chatter, you'd have to look at the cut to see if it is scalloped.

But it is reasonably normal with a tool stuck out that far, to have chatter when the width of the cut is increased, such as when your ten thou cut suddenly gets much wider/deeper after you hit the back of the "pocket" you are cutting. Common with boring bars, and a good reason for having the biggest one that fits in the hole.

As for the stickout, and why I was expecting the part to come off the chuck.....

The ID seems small, the OD large, and you are cutting near the OD. Lots of leverage relative to the "hold". Far better would be the OTHER type jaws, the ones that are stepped down as they go in vs the ones stepped down towards the OD of the chuck. With that type the "hold" is on teh OD of the part.

The term "stickout" has a lot of meanings. The two I like best are:

1) distance out on the part from the chuck, vs diameter of the part. After 3 or 4 diameters it gets to being a bit iffy. For larger diameters , it really should be the distance from the nose bearing of the spindle to the end of the work. It's the unsupported length of the part, vs diameter. I include the chuck when the part is larger than the spindle diameter, because teh spindle is more likely to be sprung by the work.

2) the distance out on the work vs the diameter of the "hold" you have on it. This is the same as #1 for holding on the OD of the part, but is far more descriptive of the situation for parts where the "hold" is on the ID of the part, as in your case.

Both of these have to do with leverage. The leverage the tool has against the stiffness of the part, and the security of the hold the chuck has on it. The best "hold" is nearly always with a 4 jaw chuck, and they only need one set of jaws to do either and OD hold or an ID hold. The jaws reverse easily.

When teh part is flexible and the "stickout" is long, the part can flex. This may mean it pushes away from the tool, and is NOT cut as much at the end, giving a taper. It can ALSO mean that it is "sucked in" by the cut, making it deeper, so that you end up with either chatter, or in bad cases, the work "catching", bending and causing a "crash".

In your case I fully expected to see the part pulled off the jaws by the cut, due to the smaller diameter of the "chuck hold" vs the longish part and large diameter of the cut. When the tool hits the back of the pocket, the forces go up a lot, very suddenly.

All good info.

But I think this gentleman is SO new to lathe work that we need to step back several steps and hold his hand until he can get a perfect finish on something.

Id say lose the ID turning..we cant see the tool very well and its alot more subject to bad workholding and machine tolerances than turning the OD of something smaller and shorter.

I would suggest a short solid bar of brass or aluminum.

And given what I've seen on my Atlas..definitely check the bearing preload. Thats job #1. I went from being totally unable to get a reasonable finish to getting a VERY nice finish in about 5 minutes. Its super easy.

Also, grab the parts manual, operations manual, and threading supplement for the Atlas. They are all online in various places, Id put links but I dont remember where I got them. Just google. The threading supplement may seem too far advanced but it has a lot of enlightening descriptions of how the lathe works.

k2man
10-13-2014, 04:19 PM
I actually have a little time on a metal lathe - my neighbor has a cheap, bad condition mini lathe that I have used. I also had a wood lathe as a kid that I used extensively. I understand the forces involved - good explanation J Tiers, I appreciate your help here.

I need to make this part as it is a mold for a silicone rubber part that I need to get done. It is part of a prototype of an invention I'm working on. It doesn't have to be too precise, as what I've used so far is PVC pipe, polycarbonate sheet, a drill press and a dremel. I just got the lathe (and also a Bridgeport 2J2 mill) so I am light years ahead of where I was.

Doozer
10-13-2014, 10:04 PM
Did-you-try-rasing-the-toolbit?

--Doozer

k2man
10-14-2014, 08:08 AM
Thanks Doozer. I will raise the bit. I had read that it should be set 2* above center. I had a HSS bit I was using to do most of the cutting. I had it set high, and I was running at the slowest speed.

I'll try setting the bit back up - today when I get back from my colonoscopy [emoji27]


Sent from my iPhone using Tapatalk

90LX_Notch
10-14-2014, 10:29 AM
Here are my observations-

QCTP is oriented wrong. Turn it 90 degrees clockwise. The way you have it has the cross slide at the far end of it's travel; so any loosness in the gibbs is going to be magnified.

You are feeding with the compound. Loose gibbs become your enemy the farther you feed with the compound. Feed with the carriage that's why it has a feed/thread capability.

It looks like you are cutting with a large portion of the side of the boring bar. It is acting more like a form tool at that point with a non ridgid (Atlas) lathe. You need to be cutting with a small section of the corner. Angle the QCTP to achieve this.

I agree with what others have mentioned. Stick out is going to magnify any problem and incorrect tool height will cause all kind of problems.

Tip: Color the cutting edges of the boring bar as well as the area below them with a Sharpie. This will allow you to see if the tool is rubbing anywhere.

If I were you I'd adjust all of the gibbs on the machine. That can't hurt but only help.


-Bob

dp
10-14-2014, 12:42 PM
Even with a perfect lathe I think that job needs a cat head and steady rest. There's too much stickout from a too-small chuck. Others have commented on good lathe mastery: gibs, cutter geometry and position, QCTP orientation. I'd also lock the fixed axis.

I wonder too if it is possible to make an all-thread arbor run through the spindle bore that can be used to pull the work against the chuck. It is easier to do than making a cat head. Attaching it to the work can be a simple cross-bored hole with a pin and clevis screwed to the all-thread. It looked to me like the work was going to leave the area during the cut.

BTW - that didn't sound at all like harmonic chatter so much as a dull cutter or inadequate clearance.