lathe back plate machining to true it up

[Buickgsman]

The 3 jaw chuck on the clausing 4904 has a small amount of wobble to it which might make the theory that the lathe was dropped more of a possibility. Its a small amount of wobble but I was wondering if I could just face the backplate true to make the chuck not wobble. Is this a reasonable fix for this problem or any other suggestions on how I could fix it? I have checked the spindle and there is no runout.

Thanks

 

[Ray C]

I would strongly recommend finding the true problem first then, make calculated corrections.

Start by measuring the spindle RO and work backward or forward from there (if that makes sense). If you can more precisely define the wobble you're seeing, we can all pitch in and figure-out how best to diagnose.


Ray

 

[fastback]

+1 with Ray. You need to know what you are fixing before you can truly fix it.

 

[Buickgsman]

Ray, I measured the spindle runout and it was <or= .0005 (measuring the spindle turning by hand and the indicator riding on the front inside edge of the spindle taper.) It looks to me that the back plate is slightly bent causing the chuck to wobble. I do have to say that I am not an expert at checking runout either.. so if this sounds wrong please feel free to correct my measuring technique. Thanks !

 

[Ray C]

OK, half thou. -That's reasonable. When you say, "looks to me the the backplate is slighly bent"... How are you making that assessment? Can you visually see it wobble? Are you putting a shaft in the jaws and indicating off the shaft? Measuring the face of the chuck? Measuring the side body of the chuck?

I'd be inclined to remove the chuck from the plate and measure off the face of the plate as a starting point. If you do this, stamp a witness mark on the plate and body. Also, are you sure there's no swarf hung-up in the spindle threads or shoulder?

Do you have any other chucks to try?

It could be the chuck got bumped hard and shifted off center... Could be a lot of things and we need to go from left to right checking everything as we go.

In the Metrology section, there's a post about "D1-4: Measuring and Fitting the Chuck" (or something like that). Even though it's about a D1-4, some of the techniques of measuring the plate and chuck flatness on a granite are shown. -That's applicable here.

Still need more info to diagnose...


Ray

Ray, I measured the spindle runout and it was <or= .0005 (measuring the spindle turning by hand and the indicator riding on the front inside edge of the spindle taper.) It looks to me that the back plate is slightly bent causing the chuck to wobble. I do have to say that I am not an expert at checking runout either.. so if this sounds wrong please feel free to correct my measuring technique. Thanks !

 

[Buickgsman]

Ray, I took the chuck off to measure the runout of the spindle.. no visible swarf to booger up the threads so I belive it is sitting properly on the spindle. I can see the back plate wobble.... its very slight. I put a bar in the chuck and spun it and the rod is exaggerating the wobble the farther it goes from the chuck. I have a 4 jaw chuck I will swap on to confrim that the spindle is not the problem. The wobble in the back plate is slight. I will also remove the chuck from the plate to measure the amount of wobble and report back. I'll also hit up the posting you mentioned on measuring and fitting the chuck to see what I can learn there. Bob

 

[Ray C]

Good, keep at it. Keep in mind that if someone hand filed the plate while spinning or sloppily chamfered the edge of the plate, it may give an optical illusion of wobbling.

BTW: A shaft in the jaws won't tell you much. 99% of the time, the jaws are not square and a perfectly centered chuck will still wobble a shaft. What you described about the shaft wobble increasing down it's length is the classic symptom of misaligned jaws. To some extent, it's almost unavoidable but, can be minimized. I would still disassemble the chuck as you must proceed from left to right, checking everything as you go. It's the only way (in my opinion) to get a true baseline of the landscape. Others may feel differently about this but, I'd hate for you to take corrective actions and not get results. Once you lathe, grind, mill or file metal off something, it's hard to get it back on.

Ray

Ray, I took the chuck off to measure the runout of the spindle.. no visible swarf to booger up the threads so I belive it is sitting properly on the spindle. I can see the back plate wobble.... its very slight. I put a bar in the chuck and spun it and the rod is exaggerating the wobble the farther it goes from the chuck. I have a 4 jaw chuck I will swap on to confrim that the spindle is not the problem. The wobble in the back plate is slight. I will also remove the chuck from the plate to measure the amount of wobble and report back. I'll also hit up the posting you mentioned on measuring and fitting the chuck to see what I can learn there. Bob

 

[Ray C]

Oh, and we can't ignore the possibility that the entire head may be misaligned. Before you can make that assessment, the back and chuck must be trued-up. After that, you check bed alignment with either or both the Rollie or 2-Collar method. From there, you'll have a better idea if the head is misaligned. There's a long way to go before we get to that point.


Ray

 

[Buickgsman]

Tonight I confirmed that the back plate is indeed bent. I forgot that I got a drive plate with the lathe so I spun that onto the chuck and it turns perfectly true. I also have a 4 jaw that I spun to confirm and they are both perfect. I took the chuck off the back plate and used an indicator to check how much wobble there actually is. I have about .008 worth of wobble. Is this fixable or should I try to find a new back plate?

 

[Tony Wells]

Take a skim cut on both faces of equal amounts. Don't turn the step if it acts as a register for the chuck to sit on. If you do, and you may need to, take as little as possible, and when you mount the chuck, you'll need to indicate it in with a bar when you tighten the bolts to make sure it runs true axially. You'll be fine.