H-Frame Screw Press

So the project starts with a friend and neighbor, George, asking me to redesign an old arbor press because it is not able to do the job of compressing the springs on an automatic transmission.  The transmission cluster is just too tall to fit.  So we begin to chat about cutting the bottom off the arbor press off and adding a plate, but as the discussion progresses it becomes apparent that the arbor press is not the right tool for the job.  It would be easier and actually more effective to make something else.  I am thinking a simple H frame screw press.  The total pressure desired is only twenty, less than fifty pounds and a one ton arbor press is overkill.   Plus it would be nice if you could use the press to hold pressure.  With the arbor press George just tightened up the gib screws to jam the press in place or someone had to hold the handle.  With a screw press, the screw holds where ever you want it to hold, till you unscrew the press.  

Screw presses are about as old as technology having been used for everything from binding books to pressing grapes and olives back in Roman times. The Archimedes Screw described back in 250 B.C. along with the lever and formulas for area and volume by the famous Greek mathematician.   While Archimedes used the principal of the screw (as a water pump), the principal goes back a lot further to the incline plane, Tower of Babel, who the heck knows.   A caveman pushing the dead carcass of a slain deer up a slope.  Getting way off subject, it is good enough that we have screws, I don't think any one cares where they came from, at this point.  

Any way............ 

First step is to dig through the piles of personal materials with an idea of what you want to build, dictated by the limitations of what you have available in materials. I found an aluminum disk, half inch thick 6061, we got a bunch of these all bandsaw cut and I save four pieces for myself years ago.  I figured this would make a pretty good base. The dimensions for the opening had to be more than eight inches, because that is the diameter of the transmission cluster housing.   The capacity as to height needed to be about nine inches.   These disks are about eleven inches in diameter.   The next find was some 1" x 2" x 1/8" wall aluminum box tubing.  The piece I had was about three feet long but had been run through a rolling bender and had a slight arc to it.   You might be able to see this in the pictures that the uprights are not quite straight.   The cross beam was a piece of aluminum flat bar 1" x 3" which is kind of tall but it would offer good rigidity to keep the screw from twisting, and using  aluminum, all this extra thread would spread the threaded load.  

The screw is a piece of 5/8" x 11 tpi threaded rod which already had a nut on one end that was chiseled in place.  So while we would have to work around this nut, I was going to add a nut to the end of the screw anyway. 

The plan was to mortise the box tubing to fit the flat bar. In this way the mortise would hold the flat bar and screw vertical and control twisting and also the mortise would hold the pressure of the press.  This was accomplished using a two flute 3/16" center cutting mill that could be plunged.  I could have used other methods but the mill was available and it was nice to have the control make the hole a press fit to the cross bar.   

There are two 1/4"-20 x 1" stainless bolts with star lock washers that affix the uprights to the cross bar.  These actually pull in the uprights putting the H under strain. 

The uprights 1" x 2" box tubing are mortise at the plate end with a three degree over cut causing them to tilt outward.  These are also affixed with 1/4"-20 x 1" ss bolts. So the when the cross piece was ready to be installed the amount of draw to tighten everything up was about 1".

I thought about documenting all this better as an after thought so the number of pictures are limited to the end product and not so much the process of building, sorry.

The base plate was attached a face plate on the lathe using four holes you can see three of the four in the picture above.  Then the OD turned just to true it up and make the edge nicer.

One of the last steps and the only part besides having to purchase eight lock washers, was the bearing on the end of the screw. I wanted to have something that would not exert a turning motion to the item being press.  This press, most likely, is going to used against a flat bar across a spring cluster or some other adapter fitting and the rotation of the screw would either mar the surface or rotate off the surface, so I wanted something, a bearing to cancel out this rotation.  My first thought and one that I had in my junk box of treasures, was a tapered roller bearing.  This would take the pressure but tapered roller bearings are two pieces not attached. You have the actual bearing and then the race or cup the bearing runs in.  Plus the ID of the bearing needs a bearing surface also. So after a bit of thought the tapered roller bearing just is not ideal.  Plus all the ones I have, are larger OD than I wanted.

So a trip to Florida Bearing was in order, these guys are helpful and while asking a lot of questions they suggested a ball thrust bearing. The only thrust bearing that I am familiar with are flat series of roller bearings that form a disk. 

I had never heard of a ball thrust bearing which basically looks like a regular ball bearing except part of the race is along the axis to take thrust.  A normal ball bearing will tolerate some axial load but the Florida Bearing guys were not sure how much a regular ball bearing would take so the thrust bearing was a natural.

I wanted a press fit so I turned the screw end down to about .010" over the .503" measured ID of the bearing.  This would be the first time the press got to press something, an exciting moment.  

Turned out that while the bearing might look like a piece of stamped metal, it is actually hardened as you would expect a bearing race to be and .010" was too much.

A loud pop and I cracked the inner bearing race.  You can see the crack.  Back to Florida Bearing, $6.40 later I had a second chance to press the bearing onto the shaft.

I re-turned the shaft to .003" over.  I took my time with a file and some 220 grit wet/dry paper.  More than once I have needed to remove just a tiny bit and over done the job and ended up with an unacceptable loose fit. What you see on the shaft OD is grease and not a flaw, plus some added grease the bearing and this time the bearing pressed in just fine.

Another challenge was the tapping of a 5/8"-11 hole in the cross bar three inches thick.  This is a lot of thread. I started out with a hole a little larger than normal giving about a 70% thread.  Then the tap was started using the drill press as a tap guide. Not having it dead on perfectly perpendicular in this project would have been very bad.  Then as I ran out of tap I found an extension, turns out the tap end is 3/8" square which fits the end of a 3/8" extension bar used for a socket wrench. This and a pair of vicegrips got me that last half an inch of thread I needed. 

I put the threaded rod on the a wire wheel to clean off any burrs and then even a minute or two on a buffing wheel.    I wanted the screw to be very smooth running.  I added some special 70% Teflon grease and just two fingers will turn the screw the entire length and yet the amount of run out, wiggle, is very little.  Two things going for the design is that the thread while in aluminum is a full three inches which will spread the load considerably, and the amount of pressure in normal operation is less than fifty pounds, so I think this press, for its intended use, will give trouble free service for a long time.

Hope you enjoyed reading about this simple little project as much as I had building it.