Goofs & Blunders You Should Avoid.

In no case should you be taking a cut without taking out the backlash from whatever you move to set the cut, whether it is the compound, as in threading (for some folks) or the cross slide. It just isn't done:)
 
That's why it's called "goofs and blunders you should avoid."

Tony Wells said:
In no case should you be taking a cut without taking out the backlash from whatever you move to set the cut, whether it is the compound, as in threading (for some folks) or the cross slide. It just isn't done:)
Click to expand...
 
Tony Wells said:
Not really the best approach. Shouldn't cut with forces pulling the part away from the chuck or bottom of soft jaws unless you have to. Use M04 instead of M03, and left hand tooling. And unless you have a sufficient width thread relief to start at the bottom, there can be sync errors with the spindle encoder trying to get everything up to speed. But, ya gotta get by with what you got sometimes.

I've pulled parts out of chucks on manual machines cutting -Z- plus. I try to avoid it.
Click to expand...

Have done hundreds of 2"-4 TPI LH acme threads in HRS/CRS and have never pulled a part out of the chuck, I do them in pairs 1 LH 1 RH for turnbuckles used in ship building at the Philadelphia Ship Yard, the only change in the program is the beginning Z and ending Z all else remains the same, thread depth, lead and tool.

If you can pull a part out before the insert breaks the chuck is not tight enough, the tool will break well before that.

On a side note I set up a chucker lathe for a new guy to run a 100 part job on this week, 2 tools on a quick change post, an 80 Deg. insert turn and face tool and an 80 Deg. boring bar.

Turn and face, bore then flip the part and face the back, this requires 1 manual tool change per part, face, bore, face so the face tool is already in for the next part. I showed him what will happen if he left the boring bar on when the program called for the facing tool, "It will crash the boring tool right into the chuck said I, when you hit the start button it will rapid to .100 away from the start position said I, start the next operation without the tool and the spindle off" within 1 hour he called me to the machine and said that the insert had broken. He was right he had of course crashed it and this also rotated the tool post so I had to square it and reset the work shift and reset all of the other tool offsets, it was a long day.

Set up the same lathe for another guy early this year, 1000 parts of 3/8" CRS that have a .220 diameter X ,350" long bearing journal on one end so requires no tool changes, put part in chuck, push button, remove part insert next. After 45 minutes he told me "the insert broke", it wasn't broken it was destroyed. This is a very simple Bridgeport CNC lathe one must start the spindle like a manual lathe with a lever on the apron then push the program start button, after 30+ parts he turned the spindle on in reverse and hit Start, this approach does not work well.
 
Tony Wells said:
In no case should you be taking a cut without taking out the backlash from whatever you move to set the cut, whether it is the compound, as in threading (for some folks) or the cross slide. It just isn't done:)
Click to expand...
Well explained Tony, personally I only use the compound when threading very course leads on multiple parts in a manual lathe, 6 TPI or less, I set the compound at around 45 Deg. or so as this clears the handles nicely. Touch off the tool with the cross slide and zero the X dial or DRO then set the compound dial to 0, then dial in the cut with the compound, retract the cross slide and return to the start. Return the cross slide to 0 and advance the compound for the next pass without having had to move it back, repeat as needed.

Works a charm and makes threading faster.
 
In 40 years, I have only pulled one part out of a chuck while cutting away from the chuck, and it was because of a problem with the feed clutch. Bossman didn't believe me. It was cutting a fairly wide, deep groove in a part about 6" ID. Working off a travel indicator, I intended to stop the feed a bit short, and cut to the face by hand and feed back out of the groove to finish that wall. Well, I had bumped the indicator I guess, or something had moved....I power fed right into the wall (shoulder) of the groove and started taking a cut about 1/2" deep (per side). Naturally, first I tried to disengage the feed. It was relatively thin walled, and I wasn't running a chucking plug so I couldn't crank down on it. For whatever reason, I could move the feed lever to kill the feed, but it didn't disengage. I don't know why. It was like the lever was just flopping around and not even connected to the clutch. At any rate, once it started cutting and the feed got stuck, it didn't break anything....insert, tool, nothing. But it did of course scrap the part (groove too wide, to say the least), and it did manage to pull the part out before I could get the presence of mind to just stop the spindle. About 3 months later, running some other part, but again feeding out, boss had the same kind of incident. The feed just would not disengage. We never did figure out why it was doing that, but we sure were cautious about feeding out from the chuck on that lathe lol. I reminded him about my scrap part, and he finally had to admit it was the machine's fault. Weird thing was, on this lathe, there were two ways to get the feed to run -Z- positive. It was all in the arrangement of the feed selectors in the headstock and in the carriage. That was an 20 x 120" Edestaal(sp?) lathe. New in '79.

I am certainly not saying you can't cut things away from the chuck, but I am saying there are risks inherent to doing it, and if you don't really have to, you shouldn't. But there is a good argument in cases like your turnbuckle parts where it just makes sense to just edit a couple of lines of code and you have a workable program. You just have to make sure you have a grip on the parts, and that the part geometry will not give you any problems doing it that way. Especially I think if it is a manually operated chuck, where the tightness of the chuck is a bit subjective due to the operator, as opposed to the air or hydraulic power chucks usually on NC/CNC machines.

On that BP CNC lathe, would that be one of those ProtoTrak types where you can actually "teach" the machine by running a part manually and have the machine generate the code from how you ran it? I have always wondered how good those were. never had a chance to run one or be in a shop with one.

I know exactly the type of operator you are talking about. What they are really good at is "shrugging". I dunno what happened!!!
 
Tony Wells said:
On that BP CNC lathe, would that be one of those ProtoTrak types where you can actually "teach" the machine by running a part manually and have the machine generate the code from how you ran it? I have always wondered how good those were. never had a chance to run one or be in a shop with one.

I know exactly the type of operator you are talking about. What they are really good at is "shrugging". I dunno what happened!!!
Click to expand...

Bridgeport EZ Path lathe, a Romi manual with ball screws driven by DC servos using a Bridgeport control, the C axis is an encoder on the spindle that allows threading but can not be indexed to a fixed position, it has a mechanical spindle brake much like the clutch on an automotive AC compressor. The control is programmable in straight G-Code but it's strength and ease of use lies in the conversational mode, for canned routines such as threading it merely requires entering data in several fields, lead, thread depth, depth of first pass, minimum depth after the first pass, # of spring passes, start diameter and end diameter, (Tapered pipe threads or any other such thread), start Z and finish Z, approach angle and clearance distances between passes.

A threading program in conversational mode would look like this, field on the left input added by user on the right for say a 1"-8 thread

Thread
Lead .0625
Thread height .062
First pass .005
Minimum pass .001
Spring passes 2
Clearance .100 This gives one the option pulling the tool well clear of the part if chips become a problem
Approach angle 29.500 Deg
Start diameter 1.000
End diameter 1.000
Start Z 0.000
End Z - 2.500
M0

Boom, set tool, determine Z work shift, fill out a dozen or so parameters, close door, turn on spindle, push button and thread away, measure with gauge or over wires and adjust the thread height as needed. The input data is what an experienced manual machinist would already know, Bridgeport marketed them in the 90's as a bridge between turning centers and manuals so as not to scare off the GOMM's (Grumpy Old Man Machinists) and let them ease into NC.

By far the most simple conversational control that I have ever seen, has a peck drill screen, straight line feed screen and a 45 Deg. chamfer function. There are some things that the conversational will not allow because the software has crash protection built in therefore it will not produce an undercut, this must be done manually with line by line coding. There is someone that frequents this forum that has a BP EZ Trak knee mill, these are the same controls with canned milling routines rather then turning.

Have never used Mazatrol which is considered user friendly by some, we have some mills with Fanuc controls which are not terribly easy to learn.

As far as a teaching mode goes I have no idea but will have a look next week.
 
I ran a little Quick Turn (18T I believe) Mazak with their conversational control. I didn't care for it too much. Management at the time thought the operators could program their own next job while running the current one. They were wrong, in the case of the kind of parts we were running. We also had a couple of Webb knee mills with pretty good conversational controls. Everyone who ran them seemed to like the Dynapath controls. One was a 4 axis and it still was pretty straightforward to understand. I kind of wish I had one of those Webbs in my shop. We had an M4 rebuilt and a Dynapath control put on it. It didn't go over as well with the lathe operators as the Dynapath for mills. I didn't run it too much, but it seemed ok to me.
 
Stopped by a commercial machine shop last week to chat with a guy about a tap. While there, I saw a guy cutting on a lathe and he was pulling out the long "chips" with his fingers. Told him he should be using pliers, to which he replied, "It's around here someplace".
Today, as I'm cutting a piece and seeing some chips wrapped around the piece, I start thinking about the guy I saw and what I told him. Of course, I pull the chips with my fingers and get a very nice cut.
One should practice what one preaches. I don't know whether this is a goof or blunder, but it sure makes it in the "stupid" category.
 
Don't use an Allen wrench for a chip hook.....they're too short. I reached for a stringy chip just as the chuck jaw did, so there for a half second or so, we had a tug'o war. The lathe won. I still have no feeling in that finger. Cut to the bone and the chip broke. I had to go sit down for a few minutes.

Chip hooks need to be long enough to keep out of the way of the chips coming off the cut, even (or maybe especially) if they are stringy and shooting off pretty fast. They can change direction without notice, and get wrapped around the spindle in the blink of an eye. My favorite chip hook is a piece of copper wire, like used on the side of a power pole to ground it. It's stiff enough to work, but soft enough to bend before it drags you into the machine. I have it in a screw on file handle. After a while, the steel chips will wear through the bend in the end, and I have to bend a new hook on it, and it does get shorter, but I use only about a 1/2" hook on it. I have longer, stronger ones for cleanout with the machine stopped, but I'll use the copper one while running.
 
You must log in or register to reply here.
Back
Top