View Full Version : Tilted keyways?

06-14-2010, 03:12 PM
Just cut two keyways with a 1/4 end mill in some 6061. Both keyways are tilted. That is, the sides of the keyway are not perpendicular to the surface of the work. It's as if the mill was slightly out of tram. The work consists of two flat pieces that were to be a sandwich with the key to keep them oriented to each other. The consequence is that they won't sit flat to each other when the key is inserted between them.

One keyway was cut on the x axis and one on the y axis, and the mill is dead square and in tram. I pre-cut the slots with a 3/16 2-flute end mill and came back with the 1/4 2-flute to open them up. Both slots slant in the same direction relative to the direction of the cut. The only thing I can figure is that the end mill flexed. But there is quite a slant, so much that you'd think I was using a rubber end mill. Of course, I can go back and open up the slots and mill out each side square, but that means an oversize key and more fitting. An oversize key and fiddling with the fit negates the whole point of doing this, which was to be an easy solution to keeping these parts in line.

Has anyone else had this problem?

06-14-2010, 03:28 PM
What kind of milling machine are you using?

06-14-2010, 03:37 PM
Sounds like the end mill wasn't on center..Check the verticals. They should be equal.

06-14-2010, 04:16 PM
Is your vice wore out where the movable jay may have lifted the work piece when tightened?

06-14-2010, 04:30 PM
I think the problem is you thought endmills where not made of rubber.

You should of opened it up by using the 3/16" and just taking two finishing passes with the same sized endmill
Using a 1/4" in a precut slot will result in two flutes trying to dig in at the same time, one in climb, one in conventional. Im betting the result is the conventional cut skimed along as the climb cutting flute dug deeper.

Never use an endmill to cut a slot to the endmills size and expect accuracy. Allways rough it out with a smaller endmill, then use.. well, any endmill smaller then the finish size. Maybe not the same one as you roughed it for verious reasons, but anyway.

Allways expect heavy endmill flex in any roughing operation, thats one of the many reasons you take roughing *and* finishing passes

06-14-2010, 04:38 PM
Black Moons,
While what you say is good advice and true, I do not see how flex is going to create any off perpendicular slot condition such as the OP mentioned.

06-14-2010, 05:17 PM
Moons is dead on. Reread his post he explained it well.

Rich Carlstedt
06-14-2010, 05:46 PM
Moons is right, when climb milling the cutter deflects, just like a spinning baseball will curve, a full size endmill will deflect.
The proper cutter for a keyway is a keyway mill, not an endmill.

When you rough with a 3/16 endmill. and then follow with a 1/4 endmil , EVEN THOUGH YOU ARE ON THE SAME CENTERLINE, the cutting forces are different. one side is cutting in conventional mode (deflects cutter away) and the other side is climb milling (pulls cutter)
Such a set up can result with the 1/4" cutter pulling more to the climb side . Now add to that that the 3/16" cutter did the same thing, and you can envison such a keyway may be off .020" easily.
The faster the feed, the higher the effect

06-14-2010, 06:03 PM
THANKS, guys. Makes perfect sense. I'll remember this for next time.

In the end, I solved the problem with my old standby, brute force. Since the parts were 6061 and the key was steel, I just put it all together and smooshed (technical term) them together in the vise. I had nothing to lose since the alternative would have taken the rest of the afternoon. The parts fit together tight and square which was the goal. A couple of cap screws to seal the deal and we're in business.

The part, by the way, is a fixed 90-degree miter gauge for the table saw. The ones that come with the saws suck and the aftermarket ones are too expensive for what I need. A few bits of 6061 from the scrapyard and the problem is solved.

Thanks again.

06-14-2010, 08:30 PM
I agree with Black Moons assessment. This is a common problem discussed in most machine shop operations manuals. The intended square cut keyway becomes a trapezoid due to flex and deflection of the end mill. It's best to use a a 4 flute mill for the finish passes to minimize flex and deflection.

Rich also hit the nail on the head when mentioning feed rate. Most machine shop manuals recommend reducing the feed rate by 25%-50% to minimize the deflection toward the climb cutting flute.

06-14-2010, 08:51 PM
Agree with the assessment,
Greg has more going wrong here then is answered in these suggested reasons and replies. There was just too much missing information for such an absolute reply.
We did not see his setup, machine, speeds, type of cutter (other then 2 flute 3/16 and 1/4), condition of cutters, feeds, or depth of cut. There have been many successful machining operations that were not done by the book, simply by taking it slow and easy. His improper methods is one case where I believe this would fully apply, it could just as easily been successful had he done things differently even though not by the books.

06-14-2010, 09:55 PM

Well, no pix of the setup but below is a shot of the finished part. The machine was a Clausing 8530 which was trammed square. Cut #1 was in a piece of 3/4 x 3/8 6061 held in a Kurt vise, stock sitting on matched parallels. The cut was on the x axis. It is possible but unlikely that the stock was not square and tilted in the vise although a tilt to the degree that the slot was off would have been noticed.

Cut #2 was to the bottom surface of a fabricated angle plate (two pieces bolted and pinned together) and held in the Kurt vise. The cut was on the y axis. I doubt the stock was tilted in the vise as the vertical piece of the angle plate, which was in the vise, sat down firmly on the base of the vise. In this case, Black Moons answer makes the most sense.

As I said, the error was in the same direction of cut for both pieces. If you were to sight down the cut like you were looking down a bowling alley, and moving the cutter away from you, the tilt was toward the right. This would be the side that was getting the climb cut. Thus the Black Moons answer adds up.

While there may have been better ways to fit these two parts together, it seemed that the easiest way to get them square and keep them that way was with a keyway. I know my mill cuts square and straight so there would be no alignment issues like I might have had if I had tried to clamp the two together, get them square and keep them that way while fitting the whole clamped-up gob into a drill press vise to drill for roll pins or dowel pins.

In the end, all is well as it is dead true and I've already made some cuts on the table saw that are dead square.

THANKS AGAIN guys. I appreciate all the help.


06-14-2010, 10:06 PM
Ken, with that attitude, let me guess. Do you work for BP?

06-14-2010, 10:07 PM
"In the end, all is well as it is dead true and I've already made some cuts on the table saw that are dead square."
In the end, that's pretty much all that counts :)

Good looking part there.!

06-14-2010, 10:14 PM
Ken, with that attitude, let me guess. Do you work for BP?

No TD I don't, nor do I have any attitude. What I said is reasonable and very true. You seem to be the one with an attitude, and you can keep it. I am also reasonably experienced, just not a snob about it, if you get my drift.


J Tiers
06-14-2010, 11:37 PM
I think many of us missed the flat pieces comment and were thinking that you were cutting a keyway in a shaft. If you are the least bit off-center, the sides are "tilted", meaning they are not parallel to a radius through the middle of the cut.

It's a lot less easy to see why an end mill would cut off-square in a piece of aluminum.... unless the clamping was not so good. Or there was a LOT of deflection in that end mill.

06-14-2010, 11:46 PM
Ken , my attitude is that on a forum where most members are hobbyists of limited experience a cavalier attitude to doing things correctly is at best a way to substandard work and quite possibly a recipe for disaster in the form of injury or death.
We are not talking about toys here. Good work habits aka "by the book" should be the order of the day.

06-14-2010, 11:51 PM
Greg, if you're one to make stuff to improve your table saw experience, you might like to build a skid. Here's a few pictures of one I made for my saw.




From the third pic you can see the laminate/hardboard/laminate construction, also the long steel strip that guides in the table saw slot. I don't recommend using a confusing patterned laminate as I did- it's what I had at the time. For the bottom it doesn't matter, but for the top a plain light color would have been better. Anyway, it's a dream to cross cut accurately using this. Once I had the hardboard laminated both sides and flat (not the easiest thing to do by the way) I mounted the guide piece as squarely as I could and made sure it all slid nicely in the groove. Then I added the front and back square tubing using countersunk bolts, making sure to avoid the area inline with the saw blade. Then I carefully raised the blade through it and made the slit from front to back. I used my thickest blade for this. After that I pre-drilled a series of mounting bolt holes for the low fence, making sure to get one bolt close by each side of the saw cut. With one only of those bolts in place, I aligned the fence best as possible using a square, then clamped the left end of the fence to the skids edge. The skid extends an inch past the left side of the saw table, so I was able to do this. After a series of careful cuts in some mdf scrap, I was able to reach an ultimate alignment for the fence, then kept that clamped while I drilled and tapped the remaining mounting bolt holes. Bolts inserted and tightened before removing the clamp-

At any rate it works great. It's nice to not have to skid the workpiece across the saw table, just hold it to the fence and push through. So far, knock on wood, I have avoided cutting through either the back or the front tubing which holds it together. By the way, once the one piece fence was accurately positioned and bolted, I sawed right through it. In testing to achieve the alignment, I used a nicely cut wood strip ahead of the fence so I would be able to cut right through the test pieces without going through the fence.

Not to brag, but a point of interest- I said earlier that I used a square to get the initial alignment of the fence. The saw cut was the reference edge. The final position of the fence was off from where the square said it should be. This skid allows for a truer squared cut than my squares can measure, and it continues to be able to do so. Part of the reason for that also has to do with the length of the guide piece, and the closeness of its fit in the tables groove. By the way, I would recommend a steel guide over an aluminum one. This is mostly for longevity, but also it doesn't stick in the groove like aluminum can and will.

06-15-2010, 12:31 AM

That's a real nice one. Thanks for the pix. I particularly like the Al. fence and square tubes holding it together. I made one that was all wood and it was a pain to get the fence straight and parallel.

Right now I'm making a mantle clock and my sled is way too big for the little bits I'm cutting. My sled is about 2 x 3 feet and is a pain to lift off the saw when I'm switching between ripping and crosscutting frequently. I should make a small sled for crosscuts, but I think I've got it covered for now.

By the way, I learned today that a miter gauge or sled needs to be square to the slot in the table and NOT to the blade. If you stop to think about it, you'll see that this is correct. (Now I'll sit back and wait for the debate..... :D )

No shortage of things to do!

06-15-2010, 12:45 AM
Ken , my attitude is that on a forum where most members are hobbyists of limited experience a cavalier attitude to doing things correctly is at best a way to substandard work and quite possibly a recipe for disaster in the form of injury or death.
We are not talking about toys here. Good work habits aka "by the book" should be the order of the day.

TD, come on, get real, who works by the books all the time, you ?

BTW, where can one find this book of perfect machining practices? perhaps it is not a book at all but a person right here on HSM?

And now you have exploded my statements to injury or death, well, I can't top death. Makes me wonder how my cavalier attitude allowed me to get to my age.

06-15-2010, 12:51 AM
You're right about that- it needs to be square to the slot. On this saw, the alignment of the blade to the slot can also be done and it is, even though the adjustment room is all to one end, it was just enough to make it square.

I have a few other skids for this saw, all made from wood and smaller (job specific). This is the nicest one though. One feature I wish I'd built in is a sliding pivot point for when I'm cutting round discs. You can't really see it in these pics, but there's now a whole bunch of 1/4 inch holes off to the left of the blade, about centered from front to back. I'll set a metal pin into a hole in a piece of material, then pivot the material around as I raise the blade bit by bit. Makes beautiful round discs. I've got pivot holes for 3 inch discs, 4 inch, 6 inch, 8 inch, and some custom ones, but I'm running out of room for holes. Besides, once drilled, that's it- if it's not in the right spot- too bad. Drill another one-

I also made a nice aluminum stop block to clamp to the upright leg of the fence.

If I need to cut an angle, I also have some angled pieces to screw to the skid as a guide. I haven't used my miter in probably 20 years-

06-15-2010, 08:23 AM
My answer assumed (bad practice) that he was cutting a keyway in round stock. Flat stock should be held down on the mill table. If I was answering on the phone I would have asked a lot more questions.

I have cut many keyways in round stock successfully using four flute end mills, light feed and experience..

06-15-2010, 09:41 AM
Aw Shucks, Ken , you got me there. There isn't A book. But the textbooks and lesson texts from my apprenticeship stack up to over 37 inches. Add additional texts, manuals, and classic works such as Stanley and Colvin's and it's about 70 inches.
Notice that you were the only one to believe in your not by the book way. It was explained by two people very well and you refused to believe them. Insisting on unorthodox and flawed procedures is , indeed, a cavalier attitude.
There is a further reason that you don't cut a keyway with an on size endmill. it will be oversize. No machine and no toolholding system is completely free of runout. Therefor, the keyway which should be a snug fit will be larger than the endmill size even if it did not pull to one side due to the climb/conventional cut.
Doing things properly and doing things safely are, by and large, one and the same.
With your attitude you should be asking advice, not giving it.

06-15-2010, 10:19 AM
You have read way too much into what I said, and to make matters worse, incorrectly added from there when your ego jumped at the chance to prove someone wrong.

You said "you refused to believe them"
What I said was "Agree with the assessment"

You said "Insisting on unorthodox and flawed procedures"
What I said is"While what you say is good advice and true"

Last but not least
You said "you should be asking advice, not giving it."
Can you quote that advice I supposedly gave, I can't seem to find it.

I hope your machining skills are better then your reading recall.

The premise was that successful machining operations can, will and have been accomplished by taking ones time in spite of not being a proper or accepted method.
No one, you being the single exception, starts out or ends up doing everything exactly as it should be done.

Nothing more, end of story.

06-15-2010, 01:21 PM
Absolutely right, last time I sloted incorrectly the endmill broke off at mach 2 and killed 3 people in its path, and that was the last time I tryed things my own way to learn stuff.

Then you guys made me feel all bad about killing 3 people just because I did'nt wanna read 37" of textbooks :(

But seriously guys...
Problems been resolved, OP is happy, why are you still ****ing on eachother?