Straightening a lathe feed rod

[Paul Jones]

I have a used lathe so I am not exactly sure how the feed rod for my 12” swing gear head lathe was bent. I used a dial indicator with an adjustable arm and magnetic base for measuring the rod runout. Measurements were mostly at 12” intervals and the runout measurement written on the rod with blue marker. Also, attaching the DI mag base to the side of the lathe tailstock and sliding the tailstock along the lathe ways greatly simplified the measurement process.

I discovered a maximum runout of 0.055” at about one third of the distance from the headstock and a gradual decrease in runout towards the headstock and tailstock. I had read about solving the runout problem by either removing the feed rod or working with it in place. I decided to do the work in place. I also watched several of the shaft straightening YouTube videos by Keith Fenner. Seeing Keith’s videos gave me confidence that the shaft has to be bent way beyond the 0.055” runout in the process of straightening the shaft.

I used a scissor jack and measured the upward vertical extension movement with a DI. It took almost 0.500” in movement in the opposite direction to the bend to eventually decrease the 0.055” runout. When you do this for the first time, and see the large bend in the feed rod, you think you going to ruin the feed rod. Upon slowly releasing the jack and measuring again, you realize it is going to work. Just work slowly and methodically, measuring and marking the improved runout corrections. Eventually, I corrected the runout to a maximum of 0.008" along the entire length.

However, after applying the bending corrections many times there comes a point when the runout begins to rotate (spiral) the along the axis of the feed rod. I think this is when it would be better to remove the feed rod and use a hydraulic press (or even spot heating method with a torch) and measure the corrections at the workbench. My significant improvement in the feed rod runout is okay for now. Eventually I plan to rebuild the gears and bearings in the feed power mechanism in the lathe carriage apron. This will be an opportunity to remove the feed rod and make the final corrections off the lathe.

Thank you for viewing this,

Paul Jones

[NortonDommi]

Paul,
I just have to ask why you are worried? That is a long slender bar and the lead into the drive worm has lugs either side that will force the shaft into line. Unless the rod had a definate kink that was causing problems I would like to know the reasoning behind this. The 'spiral' can be a nightmare to chase even between centers and straightning with V-blocks in a press.

[Gramps43]

I have to believe that Keith Fenner would be having a cow seeing this. The damage you could do to your lead screw drive and heaven knows what else. Don't be lazy, remove the screw and rig up something with a hydraulic or your scissor and straighten the screw off the lathe.

D. Sheridan

[olderdan]

Paul, One method of straightening which gives more control is to chuck the shaft and work your way along it in increments correcting as you go, do this for as long as it takes end for end until you are happy with it, this way you do not get the kinks or spiral errors.
I do this often on bar stock that is close to the size I need.

[Paul Jones]

Olderdan and Gramps43,

I am sure Keith Fenner would be having a cow seeing this but don't worry because the final corrections to feed rod vibrations became almost perfect by taking the chance of doing this in place. The spiral errors are in within a few 0.001" and the feed rod is now running almost perfectly true and probably better than it is was when it was new in 1987. The proof is in the machining finishes showing almost no surface variations and even better than before I made the feed rod correction ( http://www.homemadetools.net/forum/l...7293#post23627 ). The correct approach would have been to completely remove the lathe feed rod and do it off the lathe and systematcilly making the correction with a hydralic press like Keith Fenner has shown in his YouTube videos.

Regards,

Paul Jones

[Toolmaker51]

Having corresponded with Paul Jones on various topics and instances, I trust his decision to work the shaft in place. Removal for straightening in a press is not always an option, where being suspended in its own bushings is good business. If discrepant bends were in separate 'waves', the issue would be of different concern.
Fact vibrations are diminished, and surface finished improved [along with other measures] kind of proves the point. There are feed rods and lead screws world wide that exhibit what used to be wrong with this one...

Having straightened innumerable aircraft torsion tubes, .045 wall 4130, I was often tempted to build a fixture, holding by the finished diameters at each end. Length and diameters [mulitple part numbers] varied too much to make that practical. I used UHMW Vee-blocks and a travel indicator for deflection and runout in a hydraulic press. The bends, nearly always centered in the length reacted well, with out straining the friction welded fittings at each end. Both the rotary swage forming and welding operations caused imperfections, addressed before each step. I got them with in .002 in 30"+ all the time, no-one else got close.

[olderdan]

Paul,
I was not being critical of the method you used just glad it worked for you, “the having a cow” comment was not from me (I never criticize others work).

[Paul Jones]

Olderdan,

I know and don't worry because I did have a big laugh about "having a cow". That is exactly the way I started to feel after the first time I bent the lathe feed rod way beyond the 0.055” runout in the process of straightening the shaft. At first I had this bad feeling that maybe this wasn't such a good idea after all but then I could see the immediate improvements and remembered how much Keith Fenner had to flex a shaft during the alignment process.

I would not recommend doing this "in place" straightening to the finely made lead screw that is just above the feed rod. The lead screw would have to be removed and carefully protected before doing any strengthening work off the lathe.

I finally removed the lathe feed rod and lead screw when I made the thrust bearings for these parts (see http://www.homemadetools.net/forum/l...7792#post37667 ). I almost accidentally bent the rods again when removing the two roll pins from the bearing block at the right end of the lathe before removing the two cap head screws. The block is approximately one inch thick but the lathe builder used 3" long roll pins set 2" into the lathe bed casting as guide pins to correctly align the bearing block. Removing the bearing block by clearing the long roll pins was not easy (potential of bending the lead screw, feed rod and motor control rod) and I later replaced the long roll pins with much shorter versions when replacing the bearing block.

Regards, Paul

[petertha]

I can sympathize with your bent power feed shaft. I documented my rebuild which included sending my rod to Keith Fenner for straightening. Within my post is a YouTube link where Keith is doing (my shaft) it with a combination of press bending & his heat/cool method. Best money I have ever spent IMO, he is a pro & real great guy. https://www.hobby-machinist.com/thre...7/#post-540705

But as I mentioned in the post, now with a perfectly straight feed rod in hand, I wasn't done yet. It only eliminated one problem but highlighted other issues. So I disagree with post#9 presuming the 'lugs' will keep the shaft in line & all will be well presuming your in-lathe straightening works out. It might work out that way but also it might not. If the feed rod got bent from a bad choice of where lifting straps were positioned (very common unfortunately) then the other driveline components may be OK. But if you experienced a degradation in power feed & clutch complaining as I did, the bent shaft may be the after effect, not the root cause. If you get to the bottom of my post you will see the new bearing blocks & bronze bushing I made.

I have come to the personal conclusion that the worm gear block on this particular type of machine is not a great design & the lathe kind of ran despite itself. Basically power can mask many misalignments & friction sources... until something eventually gives way. This may not be your exact situation, but hopefully it will trigger a deeper investigation. Depending on the vintage of your lathe, good luck getting a replacement worm & pinion gears, so proceed cautiously. Similarly, you likely have a metric shaft. Yes you could get probably get an IMP shaft, but not with a metric keyway to fit the gear and/or sliding fit on gear ID. So if you botch the shaft or make it worse, you may be in for some down time.

Good luck, let us know how it turns out. I'm interested.

[Paul Jones]

petertha,

This is a very interesting comment and thank you for your insight. I think I may be the exceptional case of being luckier than smart because the in situ lathe feed rod straightening worked-out fine and no more problems that I can detect from the lead screw or the feed rod. Lucky the lead screw was not bent and I have used the lead screw to cut both fine thread imperial and metric threads without any further problems in being out of round.

However, one of these days I do want to totally dismantle the lathe carriage and repair a very slow oil leak coming from the carriage oil bath sump (I think the oil leak is from one of the thread dial attachment screws drilled too deeply and into the oil sump - I think I can counterbore the top of the thread hole and use an O-ring slipped over the attachment machine bolt to stop the leak). When I do the lathe carriage dismantling, I will remove the rods and use the measurement technique used by Keith Fenner for further straightening of the lathe drive shaft rods if needed. As you advised, I will check the worm gear block and if necessary, I can fabricate the replacement parts because none are available (see my case of replacing the lathe change gears http://www.homemadetools.net/forum/r...6967#post23027 ).

I have watched several of the rod straightening videos by Keith Fenner and also saw him perform this work live at the June Summer Bash events hosted by Stan Zinkosky in Rancho Cucamonga, California. Keith's techniques are exceptional and I highly recommend trying the methods.

I will let you know how this turns out.

Regards,

Paul Jones