A square hole question...

[99Panhard]

These are the "crosses" that connect the axle to the hub in my 1910 Mitchell. Making them has been onr of or likely the most challenging part I've ever done, largely because it took me two or three years, on and off, to think of a way to do it.

Each has 4 arms with a lug at the end having a 3-degree taper to go into a matching slot on the hub. As you can see, I've gotten to the point where I need the swuare hole in the center and that I have no good way to do. So, I'm looking to the pros for suggestions. I was thinking of EDM but, as yet, I've found no local company that does it, The truth is that I've gotten so much good advice through this forum over the last 10 years that I'd rather give the job to a forum member than someone I've never dealt with. I need two crosses...the third one will work but the dimensions are slightly off so I saved that as a spare and/or for set up. I also have one original piece that the dimensions of the hole can be taken from. The square is a nominal 1-1/8" with rounded corners. The original seems to measure .006 to .008 over but it's 120 years old so I can 't tell how much of this is wear.



Here is how they work...

 

[guythatbrews]

You can cut the square on a Bridgeport with a rotary spacer. It is a bit picky but not impossible. Tilt the head both ways to be tangent to both inclined surface. Do one corner just past center both ways, then index to next corner.



This is a dihedral angle situation. For a three degree taper you won't tilt to 3 degrees but will be a bit off. I can't remember how to figure that, not even with CAD. But I'll bet someone will chime in.

 

[99Panhard]

I don't have a Bridgeport and the top speed with my B&S vertical head is not fast enough for really small end mills. In any case, the square hole isn't tapered...it's the lugs on the outside of the crosses that are tapered. Theoretically, I could do it with the slotting head that I have but that will require some modification and at least one new piece since it was made for a later machine. All that will take time and I still have reservations about whether I can do it well enough. I'd rather not experiment with parts that I spent the best part of a month making.

The center section is 1.25" inches deep.

 

[jaguar36]

You've only got to do two of them and they are AL? Just grab a file and be done with it.

 

[guythatbrews]

I don't have a Bridgeport and the top speed with my B&S vertical head is not fast enough for really small end mills. In any case, the square hole isn't tapered...it's the lugs on the outside of the crosses that are tapered. Theoretically, I could do it with the slotting head that I have but that will require some modification and at least one new piece since it was made for a later machine. All that will take time and I still have reservations about whether I can do it well enough. I'd rather not experiment with parts that I spent the best part of a month making.

The center section is 1.25" inches deep.

Oh I see now. I did a crummy reading job there. Sorry. BTW I figured out how to set the angle driving to work.

So you just need a square hole in the center. You can run HSS endmills about as slow as you want. It will just take longer. So the inside square is just a matter of time.

I guess the lugs are just tapered one way. The slotter will do this. It's not clear how close the taper needs to be. Just use a lot of care and do it.

Anyway, there is always risk ruining what you've already done any time you do a second operation. If you don't want to risk messing them up get them wired.

 

[mjr6550]

I have no advice for you, but wondered why the new ones have a keyway and four holes. I assume you modified a different part?
Also, what is the function of the square hole.
These questions may be important if I ever buy a 1910 Mitchell!

 

[William Lynn]

What's the corner radius? Common mill size? Do you have a boring head? If not, calculate the offset and make a boring arbor for that radius. Maybe offset the part in a 4 jaw and bore each corner in your lathe.

 

[jim rozen]

Is the corner a radius, or a flat? It's possible to "broach" squares like that using a lathe. Or, find somebody with (gasp) a shaper. Short of broachihng that's how you'd do it back in the day.

 

[Kingbob]

You've only got to do two of them and they are AL? Just grab a file and be done with it.

My thoughts exactly. I would imagine these would have been done with a die filer and a few jigs back in the day.

 

[99Panhard]

The hole with the keyway is there for the arbor I turned them on and to locate them on the rotary table that I milled them on. These started out as discs that weighed 8-1/4 lbs. The finished part weighs about 2-1/2 lbs. The tapers were cut with a fixture I made to hold the dividing head so I could offset it. That part is done...they fit as they should. They are made from 4140PH. The square holes are needed to match the axles which have square ends.

 

[Limy Sami]

Out of curiosity, wht is the corner radius and the top speed of your milling machine.

 

[Jim Christie]

I have a couple of thoughts on how you might mill it even at low speed depending on how large you tell us the corner radius is.
No matter what method you choose it might be a good idea to make one or more discs or bushings1.25" thick, bored to a little under 1-1/8 " and large enough in diameter to have a reasonably thick wall something in the 2-3" diameter range from some off cuts of the same 4140 material you are using and make one or more trial finished square holes of the required size.
While it may take a bit of time you will have made a couple of practice parts from perhaps essentially scrap pieces and have fine tuned your procedure and possibly reduced your stress level before starting to cut into the parts that you have many hours invested in already .
Jim

 

[99Panhard]

The radius of the corners is 1-3/8. The axles themselves are 1-3/8 with a 1-1/8 square end.
The top speed of the vertical head is about 1200 rpm

 

[Jim Christie]

I was thinking that the corner radius might have been something like 3/16" and so you could have gone in with a 3/8" endmill and 1200 RPM would probably have been more than enough for that size .
If the corner Radius has to fit the 1-3/8" OD of the axle and not just the 1-1/8" square then it is more complicated than I had first thought.
Perhaps posting a picture of a sketch or drawing of the hole would make things more clear.
Jim

 

[sfriedberg]

If you have up to 1200 RPM on the vertical head, just end mill that square. Support the part on something so the endmill doesn't cut your table, clamp it down, find the existing hole center, know your endmill diameter, and just get cutting. This is a stupid simple job on a vertical mill. Be sensible about feed and depth of cut, but frankly if you take 10 times as long as an "optimal" solution, the job will still be done pretty quickly, and without risk. You do not need some fancy pants 15,000RPM spindle to cut aluminum!
Ignore the original corner radius. As long as you make the same sized square with a smaller corner radius, the axle will fit. And if you make the radius as large as the current "chamfer" allows, there won't be an abnormal stress riser.
Use a 1/2" or 3/8" diameter endmill (or even larger), and switch to a smaller diameter (if necessary) to reduce the radius in the corners.

 

[99Panhard]

Ah...but they aren't aluminum. They are 4140 pre-hard. I also don't have a dro so every cut would have to be very carefully done using either the dial or, more likely, an indicator attached to the table.

 

[Joe Michaels]

Since there are only two (2) parts needing that square hole, a slotter with rotary table would be one way to go about it. If the square hole has sides that do not taper, a slotting head on a Brideport (or similar) mill and rotary table would do the trick.

Wire EDM certainly would get the job done, particularly if the sides of the square hole have a taper to them. I do some engineering and weld inspection for a machine shop in Kingston, NY called Fala Technologies. My nephew is working there, learning to be a CNC machinist as well as handling the day-to-day scheduling and expediting of jobs on the shop floor. Fala routinely takes on the seemingly 'impossible' jobs other shops won't take on. They have wire EDM equipment, and do some amazing things with it, as well as with their CNC waterjet cutting table.Stuff for aerospace, medical instrumentation, defense work, and on it goes. If you hit a wall, contact Mr. Andrew Clearwater at Fala Technologies in Kingston, NY. They may be able to help you out, and they do take on one-off type jobs for special projects, restorations and the like. I have no idea what the job would cost to have Fala put the square holes in your hub parts. They may well opt to do it on a CNC machining center using carbide end mill cutters. Fala is a shop where working to 'tenths' or better is routine. You'd get a first class job if Fala took it on.

 

[sfriedberg]

Ah...but they aren't aluminum. They are 4140 pre-hard. I also don't have a dro so every cut would have to be very carefully done using either the dial or, more likely, an indicator attached to the table.

My error. But I'll stand by my advice. A sharp endmill (either HSS or carbide), conservative feeds and DOC with 1200RPM is ample to cut Rc35-40 4140.
Regarding no DRO, there is no question that a DRO is a tremendous convenience. But machinists counted turns and read their micrometer axis dials for well over a century before DROs became common. You can do this. You don't need to farm it out, and you don't need another machine.
If you want a 2nd confidence builder in addition to the axis travel indicators, stick pieces of masking tape along the two mating edges of each axis. Before you start cutting, mark the two extremes of travel by drawing lines across both pieces of tape. That will give you a 2nd visual indication of travel and tell you when to slow down and start watching your micrometer dials more closely.

 

[99Panhard]

Thanks guys...
I think you have convinced me to try it. I have two more pieces, the first prototype that I made out of aluminum and another round piece that I screwed up. That is why there are three finished pieces...I ruined the first steel one and went on to try again. The two best pieces were made after I'd figured out where I'd gone wrong. I don't see how giving it a try can go wrong...I can only fail and then I'll be in the same place I am now. This discussion has given me some ideas on how to proceed.

What sort of depth do you think I should be attempting for each cut if I'm using a 5/16 or 3/8 end mill? I was doing .050 for each cut. I'm thinking of drilling a hole at each corner, slightly smaller than the diameter of the end mill and working from corner to corner they turning the rotary table 90-degrees and doing the next one.

 

[Joe Michaels]

99Panhard:

Your information is that the hub is 1.25 inches thick. If you user an end mill to machine that square center, the thickness of the hub is a concern. A small diameter/long end mill is prone to break if too heavy a cut is taken. The end mill is a cantilevered beam. A small diameter end mill is none too stout when side-milling a cut over a 1.25" thickness of the hub. The heavier the cut, the more that cutter wants to deflect under the load from the cut. A 3/8" diameter end mill taking even a light cut over
the full 1.25" depth of the hub will want to deflect. Carbide end mills are even more prone to breaking from this type of side-loading/deflection. HSS is a bit more forgiving, not by much.

The fact the hub has that center hole bored and a keyway cut has hogged out a lot of meat, which is a good thing. Your idea of drilling the (4) corners with a smaller drill is a good one. My recommendation is to go with your idea of drilling the (4) corners. Then, I'd use the largest diameter end mill possible to rough out the square
hole. A roughing endmill ("corncob") would be my choice to rough out the square hole. Once I had the hole roughed to a square shape with the larger diameter roughing end mill, I'd finish with the smaller diameter end mill to take the sides and corners to final dimensions.

I am not ashamed to admit that I've had a few end mills break off due to excessive side load. I've seen 1/4", 3/8" and even a long flute 1/2" diameter end mill break off due to side loading. Small diameter end mills doing side milling on the order of 1 1/4" need to be babied. 0.050" is way too heavy a cut for a 3/8" end mill, working on the side of the cutter and 1 1/4" thick material. I'd be thinking more on the order of 0.015" - 0.020" depth of cut for those cuts where the full 1 1/4" thickness of the hub is being milled to finished dimensions.