PDA

View Full Version : MT3 centering stub adapter



darryl
03-13-2011, 07:58 PM
Here's the first experimental adapter I made, following the idea I put forth a few days ago.
http://img.photobucket.com/albums/v136/heinrich/mt3adapter-centeringstub.jpg

Sorry about the 'unique' quality of the image- it was taken in my office with dim light, against a dark background. I 'creatively enhanced' the image so the details would show up.

Anyway, the threaded rod is the drawbar, and the two discs epoxied onto the rod are both tapered to match the MT3 spec. I used Supermend epoxy for this, which has proven to be a very durable compound. To assemble, and assure the correct position of both discs, I dry fit the inner one first, lightly snugged the drawbar nut, then ran the outer disc onto the rod. I rotated it until it matched the spindle taper and became snug, then withdrew the rod. Measuring the distance between the discs and keeping that in memory, I degreased it all, wiped clean the threads both internal and external, then applied the epoxy in the correct spot for the inner disc. Running it back and forth a bit got the epoxy distributed into the threads.

Then I applied epoxy to the second position, again running the disc back and forth a bit. I then left the second disc a bit further from the first disc than the measurement came to. Wiped a little light oil onto the ODs of the discs, and into the spindle bore. Inserting it into the spindle, I snugged the drawbar nut and gave some time for the epoxy to squish while the inner disc found its metal to metal contact with the threaded rod. I left the drawbar nut just barely snug. At this point I ran the outer disc along until it became seated in the spindle bore and found its own seating on the threaded rod. With a slight tightening of the drawbar nut at this point, I left it overnight.

Today I machined the stub end and polished it up. As I repeatedly mounted and removed this adapter, I found that it fit perfectly- there's no play whatsoever at either disc, and it comes to a solid stop everytime it's inserted into the spindle bore. This was my main design goal- to have both ends of the adapter fit the spindle bore at the same time, not one end loose while the other is tight or vice versa. I can't detect any runout on the stub with repeated remountings, so that proves to me both that my spindle bore is true, and that the adapter is true.

This was the first experimental model, which I will use since it worked out fine. I have two more to put together now, and both have thicker discs, especially the front disc. Individually, the discs seem to fit the taper, since there is no discernable rocking once a disc is touching the taper. Somehow I must have gotten the taper angle correct.

One drawback- because the drawbar is integral with the 'business end', it takes more clearance to the right of the spindle to insert and remove it.

philbur
03-13-2011, 08:05 PM
Which thread! Link maybe?

Phil:)


Here's the first experimental adapter I made, following the idea I put forth a few days ago.

Carld
03-13-2011, 08:05 PM
Sooo, what are you going to use it for?

John Stevenson
03-13-2011, 08:17 PM
Clever idea.
I use something similar but made from the solid to go in the headstock on the small CNC lathe with the end machined to 4mm.

This 'pin' then locates those push on air connectors so I can just grip by the hexagon and thread the existing thread down to a special size.

Just realised after seeing yours that instead of making a solid adaptor I just need to thread one of those soft MT3 adaptors and fit a piece off all thread / studding and machine this for the pin size. Then it becomes adjustable and lockable by means of a lock nut.

darryl
03-13-2011, 08:37 PM
http://bbs.homeshopmachinist.net/showthread.php?t=46979 Here's a link to the beginning of this idea.

randyc
03-13-2011, 11:36 PM
Clever idea !

darryl
03-14-2011, 12:57 AM
What will I use it for? Whenever I need to make a large disc from wood, plastic, or aluminum, I usually rough it out on the table saw. To do this I drill a center hole, 1/4 inch diameter, and use a pin poked into a hole on my table saw skid to rotate the blank on as I pass it through the saw blade. Once all the corners have been knocked off, the blank is then rotated on the pin to allow the blade to clean up and leave the disc round. When this disc needs some lathe work on it, it gets mounted to a faceplate. The centering pin centers it while the holding hardware is added. If I need to remove it and reinstall it, the pin is able to re-center the part accurately.

It is pretty much a faceplate aid. There have been several times I've made custom holding jigs, but they are only useful one time- once you remove them from the faceplate, you can never get them back perfectly. With the centering pin they can be re-used. Most of these jigs are either mdf or pvc- relatively stable stuff if not damaged or subjected to extremes, or moisture in the case of mdf.

The first use of this centering pin will be when I make the form for my belt project. The form needs to go from the table saw to the faceplate, and will be in two halves. Because the machining will be right at the limit of my lathes capacity, I can't afford any extra material on the blank, which normally would let me true it and arrive at the required dimension. I need the blank to run true right from the start.

I do a similar operation on the mill, where I might need to true up the OD of a large disc. I attach a pivot pin to the table, then place the workpiece over it. I can rotate the workpiece by hand while an endmill for example works on the OD. I call this pivot point milling. In either case, the center hole in the workpiece is used as a reference point, and where the fit to the pin is snug, the runout is essentially zero.

I'll need to use the tool post grinder for the belt project, so the workpiece has to be perfectly centered on the lathe, and I have to be able to remove and remount it perfectly centered.

Sometimes I wish work wouldn't get in the way so I could spend more time in the shop.

Carld
03-14-2011, 09:54 AM
Hmmm, good idea. You could have made a spacer between the two discs and then you could screw the all thread in or out as needed to get more or less sticking out for different lathes or different jobs.

You could just snug the two discs against the spacer and have the right amount sticking out as needed.

fciron
03-14-2011, 10:31 AM
Cool! Glad to see that it worked.

I thought of this a bit late to put it in on the other thread. Why not just make a short tapered piece? I realize two points of contact at either end of the taper would seem more accurate, but wouldn't a 1-1/2" long taper be just as good?

Fewer parts, nothing to go out of adjustment when you're not looking, no waiting on epoxy, Occam's razor, blah, blah, blah. -

Rich Carlstedt
03-14-2011, 05:11 PM
A variation of an old machinist trick to match unknown tapers.
You make two discs with very sharp edges and centrally threaded to match a threaded rod. The threaded rod is center drilled dead nuts beforehand.
One disc is larger than the other, and both fit the taper, one deep and the other shallow. Put a lock nut on the shallow one, and fit to the taper by gradually turning in the large one. When they match the taper, spin in a second lock nut on the bigger disc.
Remove and place between centers of a lathe and mount a DI on the tool slide. When both disc sharp corners match on the DI, your compound has the taper setting to make a solid plug with the exact same taper.
This method was used in the 30's
Taught to me by Charlie , a real "Bull of the Woods" era machinist (RIP)

Rich

darryl
03-14-2011, 10:14 PM
Thanks, Rich. Being able to turn the exact taper is one of the problems that this method solves. Having a method to set the compound to the exact angle would be very handy.

I did a little checking of my compound last night. If I don't allow the dovetail slide surfaces to become exposed, I have a full 7/8 inch of travel. It can go up to 1 3/4 inch maximum at which point the leadscrew nut 'bottoms out'. In this condition, about an inch of the slideway becomes exposed, and the inner cavity where the leadscrew is has already become exposed. I don't like to use it like this, so I'm figuring the maximum travel of the compound to be 1 1/4 inch.

That is enough to make a stable tapered plug, even if it isn't the normal length of a morse taper adapter. I can always drill and tap for a drawbar, and an option there is to epoxy in a short length of threaded rod with a coupler nut, in which case I can still use the existing drawbar which is sized for the full length adapters.

I've wondered during this process of mine whether I should adapt a larger compound to this lathe, something with more meat and longer travel. I think I'll look into this. While I'm at it, I could arrange a more precise indexing for it. I would probably include a stop point at 90, one at 88.4 or whatever the MT spec is, a couple for threading, etc. A positive notch at these points would help a lot when setting up for various operations.

Some of the suggestions that have been made here seem to allow for a threaded rod to go through the mating discs that fit the taper, but don't take into consideration that the rod won't be able to repeatably center in the threads. Anytime there's a looseness that can be re-tightened, such as running up a locknut, there's a chance for some runout. I'm relying in large part on the threaded rod being integral with the tapered discs, ie, once epoxied there can be no relative motion between the rod and the discs. The repeatability is built-in, but if the rod can ever move independently of the tapered discs, the precision is lost.

Using locknuts and/or spacers between discs- I can see that making it very difficult to get things exactly right. For one, I've relied on the threaded discs being able to float slightly on the threaded rod, which allows the discs to self-align to some extent- a locknut would destroy that possibility. A spacer would do that as well, plus it would have to be very carefully trimmed to length in order to allow the parts to fit well. A thou to short and the inner tapered disc would be a loose fit- a tad too long and the outboard disc would have slop. My opinion only, but I think the way I've done it is close to optimum. Still room for improvement though.

My second adapter came out as good as the first- this one has wider discs for a larger contact area in the spindle bore. I have one more to glue up- the centering stub on it will stick out further, so it will be useful with setups that have to be further out from the end of the spindle.

fciron
03-14-2011, 10:50 PM
Darry, makes sense: that's not a lot of compound travel. Sounds like your disks are doing the trick, so more power to you. :)

Rich Carlstedt
03-14-2011, 11:45 PM
[QUOTE=............, but don't take into consideration that the rod won't be able to repeatably center in the threads. Anytime there's a looseness that can be re-tightened, such as running up a locknut, there's a chance for some runout......................[/QUOTE]

Darryl
Not really, The disks will center themselves.
Under load, the helix acts as a spiral cone, and cones do nest 100 %
The "secret" if there is one , is to have the threads perfectly perpendicular to the disc faces . This can be done easily by single pointing in the lathe, or tapping in the mill/drillpress using good down force to hold the hole perpendicularity.
In the lathe, after single pointing, you can turn the OD and face/cutoff with good assurance.
If the faces are "square with the thread they will automatically center when a load (ie locknut) is placed against them.
You can see this easily by taking a 6-32 screw and putting it into a 8-32 nut.
Lots of play/clearance, but when you pull or push, the screw centers itself and you will need to release force to dislodge it sideways.

If you do tap by hand, the best thnig is to turn the discs and face them while threaded on a stud, but loading them properly so they square up is really tough

Hope this helps

Rich

darryl
03-15-2011, 01:28 AM
When I threaded these discs, it was a bit of fun:( I cut off and faced the parts, then drilled the center hole using the lathe. Then set up a guide pin in the tailstock chuck to guide the tap. I got the tap into the discs as far as I could before they began slipping in the chuck, then without removing the tap from the disc, I transferred it to the vice. Finished tapping, using fluid, then deburred one side.

I mounted a grade 8 bolt in the lathe and played with it til I got the threads nearest the chuck to the least runout state I could. I have some stepped steel washers that are accurately made, so I put one of those over the threads and up against the jaws. Then each threaded disc in turn went on the bolt and got faced, deburred again, then reversed and faced again. At this time I brought them to diameter and turned the taper on them, rounded the edges and polished using purple scrubbie. I did this because even though the tapping seemed to start well , each disc had a discernable amount of wobble, different for each one. Had I single pointed them in the lathe, they would have been true from the start. It was simply a choice of which way I was willing to do it.

In any event, using a bolt as a mounting spud almost never gets things very true. I could have single pointed a stub on some scrap, and it would have made a better post to mount the discs on for truing.

Rich, what you're saying about threads aligning when brought into contact is true for the most part, but I still think you'll get runout if you do some adjusting then locknut again. Perhaps if the internal and external threads are single pointed, this effect will be negligible. It's probably worthy of some experimentation anyway. I haven't tried that in any serious way.

Don Young
03-15-2011, 10:13 PM
That is certainly a clever idea. I am wondering if it would not work just as well with only the larger disc and some sort of 'semi-precision" bushing and a nut to hold the back end of the rod approximately centered at the back end of the spindle. Sort of like a collet setup uses. With a long rod any minor centering change on the back end should be insignificant. Of course any bending of the rod near the headstock after machining would create an error.