Acceptable Amount of Play

My lathe is a 9" A V Carroll, which is very similar to a 9" South Bend. I am posting here because I can't find anyone else with my model. I am hoping you SB folks can help.

The problem I am trying to solve is terrible chatter on even a light cut on an 8" steel test bar. My experience is limited to a mini-lathe, which required a lot of tinkering to remove chatter and giving me some experience on the topic.

The Carroll is a bench top 9" with the motor behind the head stock powering an overhead pulley that connects to the head stock via flat belt. It has bronze split bearings on the head stock. I added a QCTP after the below picture. (The T-nut is not the problem.)

Here is a picture:


I am guessing the problem falls into one of 3 categories: 1. Excessive play, 2. Machine rigidity, or 3. Tooling.

Any comments on the following would be appreciated.

1. The head stock has .003" vertical play, but has about .006" horizontal play. The carriage has around .005" of play. I have tried tightening up the carriage and cross slides to the point they almost will not move leaving only the head stock play as the variable and this did not fix the problem. Most SB guys are adjusting vertical play with shims. I have not seen any discussion about horizontal play. I don't know if anything can be done about horizontal play. Do these numbers sound acceptable or bad enough to cause the chatter issue?

2. The lathe is mounted on a wooden bench reinforced with steel including 1/4" steel top. It is bolted to the bench, but the bench can sway a bit. The motor is mounted on a floating pin (like a hinge) that tensions the belt via gravity pulling the motor. I noticed that when it is running, the motor jumps around a bit and creates some vibration. Does anyone with experience know of this causing bed flex and/or chatter?

3. Proper tooling is a bit of a mystery. I would assume that the info in the South Bend manual would apply, which calls for significant positive rake and the cutter tip 3/64" above the center. That's based on using a lantern tool holder. I am going on the assumption that using a QCTP changes things a bit and that the cutter should be on the center. A HSS tool ground with positive rake still had significant chatter and poor finish. I have not tried a carbide insert yet due to not having a correct size for the AXA holder. I am a little hesitant to spend any more money on this thing until I can be sure it will work. Is there a tool setup that would be least likely to cause chatter?

MP

mpoore said:

3. Proper tooling is a bit of a mystery. I would assume that the info in the South Bend manual would apply, which calls for significant positive rake and the cutter tip 3/64" above the center. That's based on using a lantern tool holder. I am going on the assumption that using a QCTP changes things a bit and that the cutter should be on the center. A HSS tool ground with positive rake still had significant chatter and poor finish. I have not tried a carbide insert yet due to not having a correct size for the AXA holder. I am a little hesitant to spend any more money on this thing until I can be sure it will work. Is there a tool setup that would be least likely to cause chatter?

MP

Click to expand...

Going to insert tooling on the QCTP was night and day for me. I had HSS ground with various degrees of rake also in the QCTP, but the carbide inserts are giving the best results. Depth of cut is limited, but it no longer chatters.

I'm using CCMT/CCGT inserts along with some cheapo holders. The CCGT inserts are razor sharp with high rake designed for aluminum but they can take a decent cut in steel and give a great finish.

Your clearances sound OK, remember that backlash and play in the cross feed screws and nuts contribute. On mine I have found that gib adjustment is also critical. This is all assuming you are running the proper RPMs for the kind of work you are doing.

A couple of comments:

If you are cutting at the end of an un-supported 8" long bar is bound to chatter no matter what you do. The bar needs to be supported with a center in the tailstock. The lathe and the chuck simply are not rigid enough on that lathe to allow a chatter-free cut on an cantilevered work piece that long. A couple of inches of "stick out" is the most you should consider.

If your motor is mounted so that gravity tensions the belt, you also have to have it oriented so that motor torque and belt tension pull the motor down to add tension to the belt. If motor torque and the belt tension tend to pull the motor up, then it will bounce around and the belt may tend to slip under heavy load.

Dobermann said:

A couple of comments:

If you are cutting at the end of an un-supported 8" long bar is bound to chatter no matter what you do. The bar needs to be supported with a center in the tailstock. The lathe and the chuck simply are not rigid enough on that lathe to allow a chatter-free cut on an cantilevered work piece that long. A couple of inches of "stick out" is the most you should consider.

Click to expand...

No doubt tail stock support would improve the situation, but this was intended to be a test bar to align the bed. It's 1 1/4" thick. The chatter is worse at the end, but it is continuous right up to the chuck.

pavt said:

Going to insert tooling on the QCTP was night and day for me. I had HSS ground with various degrees of rake also in the QCTP, but the carbide inserts are giving the best results. Depth of cut is limited, but it no longer chatters.

Your clearances sound OK, remember that backlash and play in the cross feed screws and nuts contribute. On mine I have found that gib adjustment is also critical. This is all assuming you are running the proper RPMs for the kind of work you are doing.

Click to expand...

I take it from your reply that chatter was an issue with all HSS tools in the QCTP?

The belts are set up for the lowest rpm possible without back gears. Part of my test included tightening up the gibs to the point it was difficult to move.

mpoore said:

I take it from your reply that chatter was an issue with all HSS tools in the QCTP?

Click to expand...

No, I did not mean to imply that -- maybe I could have worded my post better. What I meant was that the chatter would start sooner with HSS in a lantern post or a 4-way post. I then got a QCTP and that improved things, because its more rigid. Then I went to carbide and the chatter is far less than what it would do under the same conditions with the HSS. Carbide will chatter too, if you do it wrong -- but it gives me a lot more leeway before it all goes bad. Especially with the sharp ground inserts.

pavt said:

No, I did not mean to imply that -- maybe I could have worded my post better. What I meant was that the chatter would start sooner with HSS in a lantern post or a 4-way post. I then got a QCTP and that improved things, because its more rigid. Then I went to carbide and the chatter is far less than what it would do under the same conditions with the HSS.

Click to expand...

Thanks for the clarification and sorry for misunderstanding your post. I was worried that HSS in a QCTP might increase chatter as it would not allow the ideal tool angle for these old lathes like with a lantern where the wedge allows a lot of rake.

Quote "If your motor is mounted so that gravity tensions the belt, you also have to have it oriented so that motor torque and belt tension pull the motor down to add tension to the belt. If motor torque and the belt tension tend to pull the motor up, then it will bounce around and the belt may tend to slip under heavy load."

Yes..See if you can take the bounce out of the motor mount some how..

Quote "The lathe is mounted on a wooden bench reinforced with steel including 1/4" steel top. It is bolted to the bench, but the bench can sway a bit"

See if you can take that SWAY out of the bench that should help some...

The head stock has .003" vertical play, but has about .006" horizontal play

Click to expand...

I am not going to say what is acceptable, but I will say this is the main contributor and if less the occurrence of chatter will be less.

Eight inches out unsupported is way beyond this machine's capability - even if the lathe was factory fresh as to fit of spindle bearings. Simply not enough MASS to support such goings on

John oder nails it.

I'd like to reduce to the play in the head stock, but it would appear the bronze bearings are .003" out of round given the clearance measurements of vertical at .003" and horizontal at .006". Can shims fix this or would this be more extensive? I am thinking the proper fix is shimming the bearings under size and line boring.

I understand that 8" unsupported is too much. What length would you use for a test bar to measure head stock alignment with the ways?

I wouldn't be turning anything. You need to either purchase a test bar with a #3 morse taper on one end so that you can insert that into the spindle. (Assuming it's a #3 Morse taper like a lot of similar South Bends) This removes alignment of the chuck jaws from the test mix.

Or

Purchase piece of ground finish drill rod, maybe 1" in diameter and chuck that up. Get a dial indicator to ride along on the carriage and indicate against the rod for 10 or 12 inches of carriage travel. Mark down the results from that test. Rotate the spindle 90 deg repeat the indicator measurement. Rotate 90 degrees again and repeat the indicator test. You can rotate back to the original position and re-check if you think there will be variation in your measurement technique.

From those measurements you should have a pretty good idea of the alignment of the spindle with the bed ways. Remember this isn't a Hardinge or similar high-precision machine. It's just a clunky ol' South Bend clone.

It is MT3. I didn't know they sell MT3 test bars until I read your post and looked it up. Problem solved. Thanks.

What about the head stock bearing issue? Does anyone know of an easier way to tighten up the horizontal clearance problem?

One other factor that might be contributing to the chatter is how far out your cutting edge is from the lantern tool post. Keep the cutter in as close and as tight as you can.

JMHO

-Ron

Thanks for the suggestion. I am using a QCTP with the tool tight to the post.

If your bearing are showing oval or “egging” then I would investigate your bearing caps and such, I’m not Familiar with your head bearing type so I can’t say specifics but if your bearings have expanders And caps then it’s possible your expanders are not tight allowing the bearing to clamp and then the bearing caps when tightened down could squish the tops and bottoms of the bearing making it egg out. Make sure the bearings are adjusted correctly

Homebrewblob said:

If your bearing are showing oval or “egging” then I would investigate your bearing caps and such, I’m not Familiar with your head bearing type so I can’t say specifics but if your bearings have expanders And caps then it’s possible your expanders are not tight allowing the bearing to clamp and then the bearing caps when tightened down could squish the tops and bottoms of the bearing making it egg out. Make sure the bearings are adjusted correctly

Click to expand...

I am going to investigate this to see if it can be tightened up. I was checking it out today and noticed end play of .0025". Nothing is hugely out of spec, but it adds up. Then you throw in a bouncing motor and I think it is not surprising there is chatter.

Be carful as your problem if it was adjustment related would be a combination of too tight and too loose, now ya gotta figure out if it is and what it is

If your motor is bouncing then THAT is your first and most important issue

To see what South Bend recommends for their lathes, see this bulletin. There's a section on spindle bearing clearances.

http://vintagemachinery.org/pubs/1617/3533.pdf

Ted