Fixture for boring soft jaws on a lathe

I recently bought a set of three soft jaws for the 8" Pratt-Burned chuck on the Clausing 5914. I used these jaws to hold the flanged sleeves
being machined for the cursed Ryobi bench grinder. What was lacking was accurate boring of the jaw tips, so that a long round piece of stock can be held.
There have been various fixtures proposed and used, but one thing caught my eye about these soft jaws: The wells where the 3/8-16 cap head screws go were 0.760" in diameter, far larger than needed to accommodate a hex cap screw head. I suppose that this could be to allow one to use a hex head bolt, but another possibility occurred to me, to use these holes to mount a fixture.
So I made up a fixture from some scrap 6061 aluminum. It's a ring about 6" OD and 2.625" ID and 0.75" thick, with three 0.75" dia pins equispaced on a 4.25" circle mounted perpendicular to the plane of the ring.
The three holes to accept the pins are about 0.725" dia, having been made with a resharpened 0.75" dia end mill used as a drill, the hole having been roughed out with a large twist drill so the endmill had little material to remove. The endmill was used to get a smooth round hole, better than a twist drill can do, without going to the trouble of boring the holes.
The holes were then measured (they are not all the exact same size), and the pins were machined to be about 0.002" larger than their holes, leaving a little shoulder, and were trimmed to not quite go through the ring plate, and to stick out about 0.75" when mounted in the ring.
Press-fitting aluminum into aluminum is a bit dicey, as it tends to gall and jam, so a shrink-fit was instead used. A 0.725" diameter hole in aluminum will grow in diameter by (0.725)(350)(22*10^6)= 0.0056" for a 350 degree F rise in temperature, so the ring was heated in an old toaster oven, and the pins were simply slipped into place by hand, and the assembly allowed to cool.
At room temperature, those pins are quite rigidly held. I didn't try to push or pull one loose, but it won't be easy.
The ring mounts on the face of the chuck jaws, with the three pins projecting parallel to the spindle rotation axis into the inner three screw wells. The jaws are tightened against the pins by trying to close the jaws until stopped by the ring. The boring tool enters through the center hole in the ring.
Boring of the jaw tips was uneventful, and the jaws now hold a 0.75" dia bar quite firmly, concentric with the axis of rotation.
Joe Gwinn
Actual metal was cut in the preparation of this posting.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Sat, 02 May 2009 13:14:50 -0400, Joseph Gwinn

A useful device for a common problem but it has to be made very accurately if it is to work as intended.
The trouble is that it is over constrained - all three pins cannot simultaneously load all three jaws. With a RIGID jig, as soon as one pair of pins is contacted and pressure loading the corresponding two jaws, the chuck key cannot be tightened further and the third jaw is left unloaded. This is easily demonstrated by removing one of the pins - there will be little change in concentricity.
The solution is to deliberately introduce some circumferential compliance in the pin ring so that it is possible to continue to tighten after metal to metal contact has been reached on the first two pins.
I take this philosophy to the extreme on my own 6"chuck. I use three strips of 1/2" x 0.080" spring steel wedged between small detents in the sides of the soft jaws in the form of a triangle. As the chuck key is tightened these bow out and provide equal radial force on each of the three jaws.
Jim
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
snipped-for-privacy@yahoo.com wrote:

You are right of course, and this jig is quite rigid.
I think that many of the proposed fixtures will suffer from this as well.
My assumption is that the three well holes (and the three pins) are not that far from their theoretically correct locations, and so the fact that the pins (0.750" dia) are a bit smaller than the wells (0.760" dia) will allow things to shift around and equalize well enough. If 0.010" isn't sufficient slack, one could simply make the pins a bit smaller.
This should work in both directions, tightening on closing, and for tightening on opening.

If I understand, you have machined grooves in both sides of the soft jaws, the grooves being parallel to the rotation axis of the spindle, and put three pieces of flat spring steel between the three jaws, with strip ends in the grooves, the spring steel strips forming a triangle. When you tighten the jaws, the spring steel pieces bow away from straight and provide the force necessary to firmly seat the jaws, firm enough to allow the jaws to be machined, despite minor variations in everything.
The only thing that worries me is that the spring force due to the three strips will vary widely, depending on the exact length of the strips, the exact depths and locations of the grooves, and how far the strips deviate from straight. But if the force is large enough, it probably isn't necessary that the forces be exactly equal, so long as all three jaws are firmly held against the scroll.
Joe Gwinn
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
snip------
But if the force is large enough, it probably

Not quite true, Joe. You must eliminate any pressure to the side, which will cause one or more of the jaws to lean to one side or the other, resulting in error that will report as eccentricity in the part being held.
While the setup is nice, it is far too limiting for practical use. You are far better served to use an adjustable stop, one I call a spider, so you can set the jaws to a given diameter and remove only a trace of metal to achieve the desired diameter. That way your soft jaws will enjoy a much longer life.
There are three issues that are important if you expect the jaws to run within a half thou. One of them is loading the jaws equally, so they don't have any reason to move sideways in the slide. Second is to use one, and only one, of the holes that accept the chuck wrench. Each hole will load the scroll differently, so one may be better than the other two, assuming your chuck has three. Mark it and use it routinely.
The last consideration is to block the jaws such that they spring outwards, which is the condition they operate within when in use. Needless to say, the farther you move away from the location of the scroll, the better assured you are that the jaws will load against the slides properly, so the use of the fixture you created serves well, if you're willing to settle for the longer bar required to bore the jaws, and willing to sacrifice jaw material to achieve the desired size. I would recommend you reduce the pin size to no larger than 1/2", however, to allow the fixture to seek center, eliminating any chance that you will load the jaws with side pressure. That assures jaws that do not run true.
Harold
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Which gadget is this spider? I do recall a plate with three big holes bored in it, in the rough shape of a 3-petal cloverleaf, but this may not work all that well with the soft jaws I have, as the jaw tips were formed with a big sander, used freehand it appears, and are far from uniform.
I suspect that I won't consume the jaws all that fast, based on the rate at which I have been machining things.

I wasn't quite to the point of worrying about 0.0005" errors and side loads on the jaws, but your analysis is certainly correct.
Well, I'd quibble with one thing. I'm loading the jaws against a pinned ring on the face, so the jaws tilt outward slightly, and then boring the entire depth of the jaws, to hold long cylindrical workpieces. In use, the jaws are loaded differently than during boring, as the jaws no longer tilt outward. This effect can be somewhat remedied by shaping the pins to have balls or disks at the ends, so the load point is closer to the center of the jaws.
I'd also be loath to depend on the jaw tips and tip faces, because this is exactly the area that will be nibbled away as the jaws are machined to hold this or that. Using the wells seems more durable.
One solution to the over-constraint problem that occurred to me overnight was to use high-spot blue and a small file to ensure that the ring pins are evenly loaded. This would also require marking the ring and always installing it the same way on the chuck, the ring would fit only this specific chuck.
The chuck already has a distinguished chuck wrench hole, the one next to the Pratt-Burned logo.
The more complicated solution that also came to me overnight has a 1018 steel ring, three 1018 steel pins, and six short bits of 0.125" diameter O1 steel rod.
The ring has six radial holes, in three groups of two. Each group consists of two threaded holes that are at the 1/4 and 3/4 points in the thickness of the ring.
The six 0.125" rods are threaded to screw into the holes on the inside of the ring.
There are three identical pins. Each pin is cross-drilled twice (to accept two parallel 0.125" rods) and once lengthwise. The lengthwise hole is threaded at both ends to accept two setscrews that come from opposite ends of the pin and clamp the two 0.125" rods. Each pin is machined to have a ball or disk-shaped bulge at the end that will go into the jaw's bolthead well. The purpose of the ball or bulge is to prevent tilting of the pin under pressure from causing the pin to lever its way out of the well. The reason for two parallel rods is to reduce the tendency of the pins to tilt in the first place, by forming a parallelogram.
One could braze the ring to the rods, and rods to pins as well, or use 3 strips of O1 (instead of 6 rods) and slots in ring and pin, and braze everything together. We don't really care if the springs take a set, so long as they align correctly when used, so heat treatment isn't needed.
The intermediate solution that came to me in that same busy night has the pins on swinging arms, with the 3 arm axles turning in holes in the ring, the axles being parallel to the lathe spindle axis. Again, the pins would have disk-shaped tips.
But making a simple rigid ring with 0.500" diameter pins seems easier, and the prior movable-pin solutions may be overkill. Pins probably need to be steel, though. And having disk ends may be useful. I'll have to figure out the tradeoff between pin diameter and tolerance for imperfection.
Joe Gwinn
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Hey all,
I didn't think this was going to turn into a big deal, so I didn't jump in sooner. And I knew I'd have a hard time to find my jaw turning ring gadget to take a pix anyway. But now you've forced me into at least looking and made me mad because I can't find the damn thing. Heck, you'd think I'd be able to just go right out there in the shop and reach out and pick up something I haven't used in five years wouldn't you?!?!?! Well, I can't, so......
Anyway, mine is made by Royal, but I can't find it on their web-site, so maybe after the Chinese got involved they quit making what is undoubtedly a low-volume device. But go have a look at this PDF:
<http://www.schunk.com/schunk_files/attachments/j_backen-ausdreh-vorrichtung_DE_EN.pdf
Note the bosses or spigots on the inside end of the three in-ward mounted arms or "jaws" in the pix. These move together radially and concentric on a scroll housed in the ring, and are adjusted by hand turning the knurled outer ring (no chuck key required). It sure isn't apparent in the pix either, but these bosses are knurled too.
After the soft-jaws are newly fastened tight on the lathe chuck and jaws set by the lathe chuck scroll to approximately where you want them to be while turning/cutting them, this gadget is placed over the jaws and the spigots go in the mounting holes in the chuck soft-jaws. They are adjusted by the scroll of this gadget to apply an outward force to the chuck jaws if the jaws are to be used in normal "out-side clamping" fashion, or "in" if the chuck will be gripping an inside diameter. Once tightened securely, the lathe is then turned on and the required turning/cutting done to the soft-jaws. The "gadget" is then removed, leaving the chuck soft-jaws turned true, centred, and concentric, ready to accept the work piece.
The difference here to what has been described earlier to do the same thing is that this is adjustable over a decent range, so a job-specific "new one" doesn't have to be made for the next time, and it's QUICK.
Take care.
Brian Lawson, Bothwell, Ontario.
ps....sorry I've forgotten who the OP was.
.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

And expensive one assumes. Can't say I've seen any Chinese knockoffs, though I haven't looked either.

Very interesting. What is the diameter of the little spigots that engage the soft jaws? They look like 3/8" from the photo, but it's hard to really tell.
Not that I'll be cutting my own scrolls.
I would guess that because the moving parts with the spigots can move sideways in the tracks, and because the spigots are small, the spigots can accommodate slight inaccuracies, and so will not be over-constrained in practice.

I started the thread for sure.
Joe Gwinn
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Hey Joe,
I've answered your questions throughout the reply reply:
On Sun, 03 May 2009 17:35:44 -0400, Joseph Gwinn

Well, as I mentioned in my post, I can't find the darn thing, but they are a loose fit in the jaws mounting holes.

Understand that what you are trying to do is to "pull" each jaw "out" against the portion of the lathe chuck scroll that each jaw has contact with at that exact point, and to take up any "slop" in the jaws from wear, so "constrained" isn't maybe the best term here. You just want to make it so the new cut on the soft-jaws will be concentric with the spindle.

Yes, sorry about that Joe.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

In a separate posting, I did list a Taiwanese source.

I understand that I'm trying to impose forces on the jaws that approximate those felt in use clamping a workpiece.
Pentagrid's point about over-constraint is fundamental. Unless everything is exact, a rigid triangle of pins cannot simultaneously make perfect contact with the insides of three holes in a rigid triangle. Only two pins will make contact, and the third will not quite touch. If things are elastic (versus rigid), all three pins can make contact, but one pin will be far looser than the others.
If the machining is accurate enough, the ratio of loads on the pins may be equal enough.
An example of a rigid but perfectly constrained system is a "kinematic mount". Google will yield far too many references.
Joe Gwinn

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Sat, 02 May 2009 19:53:09 -0400, Joseph Gwinn
SNIP

So long as there is sufficient clearance the jig will always fit but, whether the clearance is small or large, a rigid jig can only only tighten on two of the three jaws

You've correctly described the system I am using but the usage is a bit different. Although the spring steel strips I am using are initially straight I tighten the chuck on them until they are substantially bowed - approximately a 1/4 or more of a circle. The detent grooves are located close to the outer ends of the jaw sides This allows a long strip length which provides a radial preload which varies only slowly with jaw position. .
This flexibility of jaw position is particularly useful when using soft jaws as bar jaws holding long stock. With the strips installed it is easy to tighten to a few thou less than the stock diameter before machining the jaws to the desired ID. One set of strips covers a wide range of workpiece diameters.
Because the force only varies slowly with jaw position, strip length is not particulary critical However good practice is to grind to equal length and be careful with the symmetry of the detent grooves.
Jim
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
snipped-for-privacy@yahoo.com wrote:

I went and measured my jig. One jaw is always looser than the others, and which jaw it is follows a particular pin on the ring, although the degree of looseness varies slightly as the ring is rotated through the 3 possible positions.
Nor do the pins form a perfect equilateral triangle. (Nor was I being careful when I made the ring.)
In the best position of the ring, the loosest pin is in jaw #1, so I've numbered this pin as #1, and the other pins also take the number of their jaw. I then measured the distance over the outsides of the pins in pairs using an electronic caliper. The pin diameters are 0.754" or so, being the as-rolled stock rod surface.
1-3: 4.476"
2-3: 4.499"
3-1: 4.483"
This is enough deviation to alone explain the loose jaw.
So, smaller pins more accurately made and located would help greatly.

Ahh. That makes sense.
I gather you've been using these springs for years without drama. What are the dimensions of the grooves? And groove shape? It would be a real problem were one of those spring strips to pop loose while the chuck was spinning. Or just sitting there being tightened. I don't know the force needed to bend the strips into arcs, but it has to be substantial, and those strips are quite light. Those little strips could fly too fast to be seen, let alone dodged.

Good point. The ring is quite rigid, in every sense of the word.

Now I see it. With the strips forming 90-degree arcs, slight variations will have only slight effects, and the side forces on the jaws should be balanced and thus cancel out. Harold worries about side forces on the jaws causing inaccuracies, but this approach shouldn't have that problem so long as the strips are the same length and the grooves are located at the same position with respect to the bottom jaws (which engage the scroll).
Joe Gwinn
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Sun, 03 May 2009 11:58:49 -0400, Joseph Gwinn
SNIP

Nothing special about the grooves - just rectangular slots about strip thickness in depth and twice as wide, milled to the full axial length of the jaw. Although the spring forces are quite large it is this force that securely retains them in the grooves. Because the slots are milled the full axial length of the jaw, when installed, the sides of the strips can lie snugly against the face of the chuck body - this automatically gives them the correct axial location with respect to the scroll face of the jaws.
Although I normally remove the strips, once the bore or the chosen face diameter has been machined, for reasonable spindle speeds, it is possible to leave them in place and immediately chuck a workpiece.
One minor correction - although I described the strips inititally straight this is not very convenient as a dead straight strip will randomly bow inwards or outwards when the jaws are tightened. To avoid this, after grinding to length, the three strips are bent a little beyond their elastic limit so that a small permanent set remains. Very little set is needed, 20 thou or so deviation from flatness will ensure consistent outward bowing.
On a more general note, although this is a very simple kludge, I believe it produces more repeatable results than the majority of conventional designs. The ones that I have seen are basically a plate, disc or ring structure conformed to try to simultaneously constrain all three jaws in a pretty rigid configuration. Such a structure can only provide full restraining force on two of the three jaws with limited or no restraint on the third.
Even the expensive piece of kit suggested by Brian Lawson (<http://www.schunk.com/schunk_files/attachments/j_backen-ausdreh-vorrichtung_DE_EN.pdf ) suffers from this limitation. However I have no doubt that, made with typical German manufacturing precision, the difference is academic rather than practical.
Jim
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
snipped-for-privacy@yahoo.com wrote:

The strips are 0.080" thick and 0.5" wide by several inches long. This implies that the grooves will be ~0.100" deep by ~0.200" wide.

Yes.
The price is certainly right too. I like the simplicity and adjustability. I will implement this.

I'd venture to guess that the German response will be that one must of course use a chuck of commensurate precision.
Joe Gwinn
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.