Vise jaw deflection

What I always do with long soft jaws is that I put say 1X2X3 blocks in the part of jaws that will bend (usually ends of jaws) and close vise (not to tight). That will cause the jaws to deflect slightly, say 0.003". Then I machine my step and after my jaws grab the part better and more even. Without doing that I find my part still tends to vibrate at the ends. Jerry

Sounds to me like you are going about your project the hard way.

Since this is not a critical task, why not

1) Clamp the aluminum to the table. 2) Use a center drill to start all your holes. 3) Use drill press with a fixture to prevent grabbing or spinning and drill all the holes. 4) Use a piloted counter bore. 5) Use a single flute countersink to bevel top and bottom of the holes.

Your done.

Using this method you should be able to hold all holes to within .004 inches.

-------------------------------------------------------------- Home Page:

If you're not the lead dog, the view never changes Doubt yourself, and the real world will eat you alive The world doesn't revolve around you, it revolves around me No skeletons in the closet; just decomposing corpses

-------------------------------------------------------------- Dependence is Vulnerability:

-------------------------------------------------------------- "Open the Pod Bay Doors please, Hal" "I'm sorry, Dave, I'm afraid I can't do that.."

Vises are often a crutch for holding parts, like someone said already, just clamp the parts to the table. Then there shouldn't be any unsupported areas. I would bank the parts against several 1/2" dowels, clamp securely and cut the parts.

Vises suck, one jaw is rigid, the other jaw basically floats, doing precision parts in a vise can only be done right when you bore a set of soft jaws. But that's just me. I detest vises but I simply hate screen savers on a work computer.

Vises are quick and easy but just lack in work holding accuracy. If vises are your vice, maybe you should think it over and use a vise only as a second option. I run matching pattern halves and they will almost never match if done in any kind of vise (Kurt too), if you want to see shift, run some matching pattern halves in a vise and see what the vise is really doing.. Michael.

The rule of thumb I learned for minimum deflection of a beam supported by two columns is to put them 25% in at each end, I think the math works out to 27%.

Assuming you're concerned with counteracting the bending from the point load of the drill, my quick and dirty analysis says the ratio between the center span and the overhangs should be 4:1. In other words, 9.6 inches between the vises and 3.2" outside.

If anyone is interested in checking my logic, the formula for the deflection of a uniform beam has the general form WL^3/CEI where C is a coefficient specific to the beam's end supports. C = 3 for a cantilever (one end fixed) and C = 192 for a beam with both ends fixed; root3(192/3) = 4

Don't forget that steel is 3x as stiff as aluminum for the same cross section (E, Youngs modulus, in the formula above), and the stiffness of a cross section (I; moment of inertia) varies with the 3rd power of the depth.

The number will vary some, depending on the exact loading conditions and the type of connections at the supports, but it seems to me that's plenty close enough for the case at hand.

PV:

Myself, I'd put the vises wide (maybe with a couple of inches hanging beyond each end of the vises) and take some studs and put two nuts on each backwards (with the integral washer to the outside - to create some jack-screws) and support the solid jaw a few places between the vises. Supporting one jaw should be enough to keep the "sag" out of your part, but if not, put a few under the movable jaw as well. Shouldn't take long to snug them up with a couple of open end wrenches after each loading of your part. 5/8" thick? What size screws are you putting in? 3/8"? I'd helix in for the counterbore. Someone suggested a piloted counterbore tool, but for CNC's there is too much chance of a chip getting between the pilot and the hole and creating havoc.

Funny, this idea has always offended me, but it finally dawned on me that effing vises are really just C clamps, attached to a table!

And, I have room for a 3rd Kurt, but a c-clamp (or two) would actually be a lot more convenient!

This is not the first time this has been suggested, but its utility

*finally* sunk in.

The other comments were helpful as well, and the notion of vises being inferior to "true fixturing" to a table is inneresting as well.

Most prototype/mini production work, tho, pretty much requires the pretty good compromise of a good vise. And, some vises are better than others. Kurt makes a series more accurate than its D series, as well.

I wasn't worried about sag, but this is simple enough to do. I've done it in the past, and sometimes just stack material underneath, with some shims.

Coupling nuts give a lot of latitude for this. I even have a slip-fit base for threaded rod, that can be slid around.

Why not mount it into a sacrificial length of steel that will support it and mount the set into the vices......

Make a work holder and hold the work and holder with vices...

Mart> Awl --

I ran through the numbers on this and for a uniformly loaded beam with two point supports the best places are very close to 1/5 of the span from each end.