Hendey spindle runout, help needed

I need some help with my 1898 Hendey. I have completed the restoration and have started working on getting the machine tuned in. Things have gone quite well but I'm stumped on a problem with the spindle. When I initially put the 3-jaw chuck on and measured runout it was way up there.. probably around 25 thousandths. With the help of forum member Joe, I worked on getting the bearings to fit properly. I am at the point now where I can put a 2x4 under the chuck and lift with all my might and not get more than 2 thou movement.

While I was working on the bearings I measure the runout of the spindle at the end of the threads (close the the headstock bearings) and was happy to see it was well under 2 thou. I then measured inside the spindle and was surprised to see it was 10 thou. It was (to me at least) interesting to note that there was a definite 'high' spot (10 thou) and a definite low spot (0 thou) and these two spots were 180 degrees apart. I worked more on the bearings, tried shims, did everything I could think of but each time I measured the inside of the spindle the reading was 100% the same, 10 thou out and high/low spots hadn't moved (I marked them).

At this point I thought that maybe someone had tried to change the taper (it seems to be close to a MT4) and messed the job up and made the taper off-center. Since the base of the threaded area was reading less that 2 thou I was hopeful that when I put the chuck on things would be OK. Unfortunately I noticed that the threads themselves seemed to be moving up and down but I wrote that off to an optical illusion. When I put the 3-jaw back on I was disappointed to find that the same 20 thou runout was still there. I measure the runout of the back plate and it appeared to be about 10 thou so part of the additional chuck runout is probably due to the chuck to back plate fit. It is quite possible that was always bad.

I'm not sure what to do at this point. I know that I can just live with the 10 thou as is and get a 4-jaw chuck. (right??) I could probably figure a way to mount an MT4 cutter and reamer in the compound or TS and re-cut the taper in the spindle. I plan to get an ER40 Collet Chuck holder with MT4 taper for holding small part so I'll need the inside of the spindle to run true.

Hopefully there is something I'm missing here? I can't figure out what happened to the the spindle nose. I have measured runout all along the spindle and it is always sub 2 thou, but the threaded part of the nose is 10 thou and visually noticeable. It looks like maybe the threads are a little rounded and I guess someone could have damaged it or maybe tried to re-thread it and got it off center then. Speaking of that... is it possible to cut the threads again? Maybe same pitch but a slightly smaller diameter? I don't have the skill to do that but I could probably find someone that could come over and help. I would assume that cutting them on the spindle right "in place" so to speak would be the way to go?

I will post a link to a short video showing the runout. Update: video is here: Hendey spindle runout - YouTube

Though it took huge forces, it appears the spindle is BENT between those two points - no other possibility

You can ignore such for this reason

A back plate MADE TO FIT your spindle will ONLY go on one way, and STOP in the same place every time. Since it will be FACED and TURNED to suit any plain back chuck, the Chuck will ALWAYS run true to the front spindle bearing - not to the THREAD or TAPER in the spindle

If instead of trying to USE the spindle taper to insert dead centers into for turning between centers - you CHUCK UP something that can be machined to function as a dead center, this perceived problem GOES AWAY

Thanks for the quick reply John. Just to make sure I understand... all I need to do is make a new back plate?

RCPDesigns said:

Thanks for the quick reply John. Just to make sure I understand... all I need to do is make a new back plate?

Click to expand...

If you have a known to fit back plate it would be interesting to see how true it runs with the chuck off - it it has one on it. In general - a back plate will have a face and a diameter that locates the chuck - those are the areas to check

If it has no dedicated chuck maybe you can reclaim it for a chuck TBD at some point

If it has a chuck and it is an old worn out three jaw, that will only bring you a whole host of unrelated run out issues

About 20 years ago I bought new 12" four jaw and three jaw China Cushmans. The three jaw has never been on the lathe because I use the four jaw exclusively - that way I am the only factor in run out or lack of same

The trued back plate and an adjust-tru back plate chuck will get you very well into spec...a knock in center then turned true with having a line up mark so goes in the same knock-in every time running dead...and a light tickle if needing .000x or so.
Regrinding the ID may be needed if intending to run collects.. but even for that a knock in and bored collect holder with a line-up mark can make a collect set run near true..A chuck held collect holder could be made.
I would agrree a bent spindle or a poor ID grind job..

Some day when you have an extra few minuets you can make a new spindle tube..

I took the chuck off and measured the back plate by itself. Attached is a picture.

The hub is a little beat up. The smoothest areas measured about 2 thou. The diameter varies a little and is anywhere between 2.98 and 2.99. I guess is was 3" at one point and has worn down.

The outside diameter is 7" and has two areas. If you look closely at the picture you can see part of it is turned slightly smaller in diameter. This area measured about 3 thou. The rest of the outside was right on 10 thou. I measured the face to see if it had any runout but it was 0.

Given these measurements it does seem that the back plate was turned on the lathe if maybe not made from scratch on the lathe. I think the most telling measurements are the outside where the main area has the same 10 thou runout I see inside the spindle but the inner section of that has been turned down to about 3 thou.

The chuck is a seemingly little used 10" diameter Bison. The diameter for the hub area is 2.9921" (I just looked this up on the MSC Direct website. Seems this is a $900+ chuck) and the outer (registration boss??) is 7.75" diameter. I note that this is a larger diameter than the back plate as such the chuck/back plate alignment is only on the hub and it is not a snug fit. There isn't a lot of movement (1 or 2 thou) but it certainly does move.

My guess is that the lathe had an accident at some point, a new back plate was made to make up for the runout. That 'new' back plate is quite old itself and has significant wear. The Bison chuck is a recent addition (I think the PO bought it for the lathe and never used it) and doesn't perfectly fit the back plate.

I think at this point a new back plate and a 4 jaw and I'm going to be in good shape. If I could find a 4 jaw that needed a less the 3" hub, then I could turn down my back plate and use it.

Guess you could make a tube stub to go on the spindle nose with having an out board ID small enough to bore for a collect holder...With it being a draw bar type or a nose nut type.

I've been using the lathe some and have started to get the hang of the controls. Turning things close to the chuck works fine because the runout is not that bad. I'm getting to the point where I'd like to make some longer parts and the runout is a problem. I can make sleeve for the backplate and then look at regrinding the spindle. I was wondering if I could MT4 taper rod in the nose, turn the spindle such that it far end of the rod is at the highest point of runout, heat spindle up and then try to push the bar down and bend the spindle back. I know it wouldn't be perfect and I'll still have to make a backplate or backplate sleeve and regrind the nose, but it would be closer. Is this feasible? Will I ruin the spindle heating it up? If not, how hot should I get it?

Another interesting measurement would be to run the indicator on the face and registration shoulder of the spindle along the spindle axis. If the spindle is bent between the in the threaded area, the registration shoulder should show little variation, while the face of the spindle will show variation. I don't know how this helps improve anything, but it could confirm the bending of the spindle, as opposed to wear or bad taper grinding.

wheels17: The nose of the spindle is so beat up it is impossible to get a good reading on it. I did measure the face of the back plate and it was true.

Previously I was thinking that the back plate had been modified (or made) after the crash. Since there was still runout in the chuck I was thinking that the chuck was added later. Today when I took the chuck off to measure as wheels17 suggested I again noticed that the back plate doesn't really register the chuck. It is close, but the "alignment" of the back plate to chuck is more the bolts than it is the hub of the back plate. I then realized I probably would have to drill/tap new holes once I turned and sleeved the hub. I then thought that, possibly, this might have been done already and that, basically, I just had the chuck on using the wrong holes. I put the chuck back on, chucked up a 12" test bar and measured nearly 100 thou runout at the end. I then removed the bolts and rotated the chuck one hole, tightened and then measured about 40 thou. Rotating to the next hole (the last of the three) yielded about 10 thou and that is out about 12" from the chuck. I could fiddle with one of the bolts (nearly back it out) and tighten one of the others more and get it down to about 3 thou. If I tighten it back it goes to 10 thou but I'm certainly happy with that for now. As such, I'm leaving it alone for now and will make a new back plate later (when my skills increase) as well as regrind the taper. I've made a couple of special bolts for the 1913 Metz Roadster and I'm pretty thrilled to have this new ability.