Cut Depth Affected By Feed Rod Rotation?

[barnbwt]

I'm playing with the PM1228VF lathe now that I have a few hours free (my measuring devices haven't arrived just yet, so mostly just making a mess and checking everything for basic function & familiarity). I have noticed a very strange, persistent behavior of the cutter that I haven't heard reference to before in my many hours of online research into lathery. Every 1/8" --every turn of the feed rod-- there is a smooth wave a couple thousandths or so deep. No matter what spindle speed, no matter what gear selection (I did fastest and slowest and the result was the same) there are these sinusoidal variations. They aren't spirals, but rings.

I have two theories, kind a of chicken/egg situation as to which might be the culprit here. The first is that the apron gear driven by the feed rod is oblong, and lifting/shifting the carriage as the high spot swings around, affecting the cut. The second is that the headstock gears driving the feed rod are tight at some point in their rotation, and this bogs the motor down subtly enough to affect the cut (this seems less likely, but there is a slight variation in RPMs every few seconds as the feed rod rotates, whether by cut depth or gear train)

Searches for the title I've chosen and pertinent key words are coming up a goose egg, so I thought I'd try here for advice. While I don't have measuring tools just yet, I can clearly track a paint dot on the feed rod, and it always sits in the same spot relative to the peak of the ridges being cut, so it's not just a frequency thing, but a feed-rod position correlation. Oh yeah, running the lead screw for a cut does not appear to leave the same ridges (harder to tell with threads, though)

TCB

 

[T Bredehoft]

My Clausing Atlas 6 by 24 had a similar problem, I've straightened some of it out. Once when I was changing the gearing, I realized that the feed screw had considerable bind on it. I had to shim the left end support out about .020 before it would turn freely. It still 'pulses' a little, but I keep a little drag on the hand wheel and it's a lot better. In fact, if I work at it, I can get a nearly perfect finish. My feed screw seems to be bent a little, I think the original set up had a free spot in the cycle, and the carriage would move ahead then and lag behind when it was binding.

 

[tomh]

TCB
Are you cutting cold rolled ? if so you will get a Finish similar to what you are describing as cold rolled has a life of it's own and can be difficult to turn , also what type of cutter are you using.and is the bit on center?
Tomh

 

[barnbwt]

The material is ~8" aluminum tube (I'm guessing 6061 or similar; it's fairly gummy and has a smooth but ugly finish & long stringy chips) in the attached photo. I also did a section of steel barrel stump (high carbon alloy most likely) stuck out only 1.5" and got THE EXACT same pattern of waves. I do not think it has anything to do with cutting at this point, and is symptomatic of machine geometry. I can even 'flatten' the waves out by aligning the witness marks on spindle and feed rod opposite the first pass when engaging the feed, or skate over the tops of all the ridges on a zero-pass by keeping it aligned.

Comically enough, the hollow tube, stuck that far out with only 1" engagement in the jaws, only varies ~.002 at the high spots by my calipers (fatter at the tail), so the lathe seems somewhat close to square at this point. Same waves using HSS or a brazed carbide cheapo (neither ground or aligned with the center carefully; I don't have my measuring tools, yet). I also haven't played with heavy cuts much since I only have so much practice stock and need to save it for measured tramming, but I think the waves get deeper with heavier cuts while maintaining the same spacing (makes sense if binding is the cause, but not if tool flex/chatter is the issue)

I will unhook the gear sync for the power feed tomorrow morning, and see if there's a hitch in the getalong. I can rotate the spindle with equal ease when unpowered, so I'm thinking this may be more of an apron issue, with the binding occurring only when engaged (makes sense, because this pulsing cut does not occur when manually or screw feeding, and it would if the mere rotation of the feed rod in parallel was the culprit). I'm guessing the meshing part that fits over the feed rod is bronze, and therefore cast, and therefore misshapen.

It's much harder to tell, but I think I can even hear a 'pulsing' when facing the end of a part with power feed.

TCB

 

[FOMOGO]

Quote: sinusoidal variations.

I had that once, but they gave me a shot and it went away. Sorry, that was truly awful. Mike

 

[brino]

I am not familiar with tnat lathe model, but to me it sound like a slightly bent feed rod/screw.
Do you only hear the pulsing with power feed engaged?
Does it happen with both horizontal and cross feed?

Can you move everything to one end to get a good view, then see if any rods/screws have any wobble.
You may need to wait for your measuring tools.

-brino

 

[barnbwt]

I'll need my tools. Being a hex rod, the feed rod is kinda hard to spot wobble in as it rotates. I only noticed the waves when using the power feed, though it would be hard to spot them even with a fine pitch thread cut (again, measuring tools). Effort to turn the disengaged feed/lead screw gear train seems consistent, which is why I am thinking it is an apron problem.

 

[RJSakowski]

I don't own this lathe but did look at the brochure on -line.

When turning, whether manually or with the lead screw, a momentary stop in lateral motion will produce a "ring" similar to what you describe. This happens because the tool makes a "spring cut" at that point, cutting very slightly deeper.

If they are uniform in spacing, it almost certainly is a problem with the lead screw. It looks like your lead screw has 8 tpi which matches the spacing you describe. A bend in the lead screw coupled with play in the half nuts could conceivably cause an issue like this. The bend could cause a binding in turn causing the half nut mechanism to "wind up" against the resistance and momentarily slowing or stopping the lateral motion. As the shaft rotates and the binding releases, the mechanism would unwind slightly accelerating the lateral motion. Another place to look is lead screw end play. Excessive end play could cause the the lead screw to move slightly back and forth if it experienced binding as described. Finally, it could be a problem inside the gearbox.

It is possible that a loose carriage could be twisting on the ways but I would think that very unlikely since the reference datum is the vee on the ways and the carriage would have to lift up to twist.

You should be able to sort this one out. Please post a follow-up. Good luck!

 

[barnbwt]

The frequency is the same when power-feeding regardless the change gear selection --that rules out the lead screw (this lathe has both a lead screw and feed rod). I suspect that the feed rod, or the rotating elements it drives are to blame. If the worm gear turned by the feed rod has its tooth axis non-collinear, it would drive the carriage advance at slightly varying speeds throughout rotation. I'll have to crack the apron open to find out this evening. I did get a dial indicator and base, today, so perhaps I can check my feed rod for variance relative to the ways.

I'm really, really hoping it's just some packing grease/dirt in a stupid spot or some super-loose gibs. Heck, I'll bet a nick or flat spot in the worm gear teeth would cause this, as well.

 

[barnbwt]

Okey-dokey, we're at least part-way there, now. According to my 'fancy' (their words) 20$ dial gage, there was about .01" of vertical play at the front and back of the carriage. Half the gib-screws on the back of the carriage were nearly falling out, and cinching those down not only did wonders for the smoothness of the handwheel action, but also took that play down to .005" or less (dial says .0005" but I doubt it's that good). Cut wobble went from .003" down to .001" --progress

I measured the amount of deflection in the feed rod and there is no way 5lbs or so of force due to it being bowed should be moving the carriage around. That said, it does seem to get more pronounced toward the chuck, but that could be workpiece deflection somehow damping out this cutting depth variance (I have like a 6" stickout just for measuring cut consistency at this time; .003" over 5", though the cheap dial and wavy surface makes a read +/- .0015" or so). It is still happening with consistent frequency across the entire length, and since there is a good .05" of play in the rod at the carriage, that rules out its warp/binding to my thinking, and my next guess is the gear box it turns has an eccentricity. If I can somehow tighten the clamp-up at the front carriage ways, I think it would be squashed to zero, but there doesn't appear to be a row of gib adjustment screws in the front like you see at the rear, so I assume they are not adjustable (and rely on the tool pressure to hold the carriage down).

I'll try a carefully-centered tool cut tomorrow to see if that may be the issue, and then I'll try to lift carriage off for a looksee (can't simply drop the apron down, it seems)

TCB