cutting keyseats

In spite of the fact that many guys cut such things in one pass, I do not endorse the method. You are at the mercy of too many variables that can result in a less than acceptable result. It has always been a good idea to rough dimensions before finishing them. My policy when cutting key slots, unless they're cut with a Woodruff cutter, is to cut them undersized, then go to final size. I rarely rely on a cutter the same size of the keyway, either. It's much nicer to use an undersized one so you can control center and size. Properly cut, keyways have a relatively close tolerance, something I don't like to risk ruining because an end mill doesn't cut size.

Harold

I will assume that what you mean by "slot drill" is an end mill. I can also assume you ended up with an oversize keyway?

When new, they are generally on size, neither over or under within a pretty tight tolerance.

My thoughts are that he's wrong too. If I want, or need, a tight fitting keyway, I'll do it with a smaller cutter, but instead of using an "on size" cutter for the next cut, I move over to each side, slightly less than half the difference between the size I want and the cutter I have, then nibble the last thousandths out until the gage blocks just barely squeeze in. End mills can, and do, walk, and it's too easy for one to go oversize. "You can always take another cut, but that putting on tool is a bitch to grind."

Greybeard.

According to Machinery's Handbook...............

For milling keyseats, the total depth to feed cutter in from outside of shaft to bottom of keyseat is M + D, where D is depth of keyseat. For checking an assembled key and shaft, calliper measurement J between top of key and bottom of shaft is used. Where J=S-(M+D)+C

The page shows a diagram of a shaft and key and the following annotations are made.

C = height of key D = depth of keyseat (or the wall height of the keyway) E = width of key J = distance from bottom of shaft to top of key M = the distance between the top of the shaft and a line passing through the upper corners or edges of the keyseat S = diameter of shaft

M is calculated using the formula M=1/2(S-sq.rt.(S^2-E^2))

However an easier approximation to give M within .001" is

M=E^2/4S

As M varies with shaft diameter it is hard to say whether or not your method is correct without knowing the diameter of the shaft in question.

What I can say though is that I personally have never heard of using two sizes of cutter to cut a keyway in a shaft and when I was an apprentice (many moons ago) I was taught the same method as you for a down and dirty 'shop method'. If the keyway was large (i.e. around 2" wide) I would probably have taken it in a couple of cuts rather than hogging it all out in one pass.

The point of all the formulas for M is to get D = C/2; in other words, the shoulder of the keyseat is bearing in the center of the face of the key. By using an endmill with a diameter equal to the width of the key and gaging the depth from the point that the tool cuts full width, Ted is getting the correct depth (within the limits of how closely one can tell when the endmill is cutting full width) without having to do the calculations.

Of course if you accept Harold's advice to use a smaller cutter, and I do, this technique doesn't work.

Ned Simmons

I had taught my son the same method as Greybeard. When he went to work, they reamed him for taking too long to do a job. For fast work, get a new cutter and do it all one pass. Need a rigid machine , my old Bridgeport wouldn't do this. His trick for depth saves time also. Feed endmill down by hand until outside edge of endmill just makes contact with shaft - a perfect flat - then go on down 1/2 of key width. No need to look up in table.

I must say he did this at an auger manufacture place. Lots o' production, no real tight tolerances. Now, he's working at a tool making outfit. They threaten to fire guys that f*&^ up a part. All one of and VERY tight tolerances. Bet quick and dirty wouldn't go at the new place.

Karl

Thanks to everyone who replied to this post. The advice will be put to good use; sooner than I thought:

The part I made with the keyseat turned out to be scrap ( through no fault of mine, for other reasons ). I will have to make a new one tomorrow. I am going to use the approach of going in with a smaller cutter first and then finishing with the final size of the keyseat.

Fingers crossed,

You are quite correct... it is likely that the keyway will be oversized...

The only time I have used more than one pass was when the machine lacked rigidity or the keyway (I was taught that keyseats are inside "the hole.") was of a size larger than about 1.5 inches. And then the size difference was not so dramatic.... say.... first pass with a 1.25 end mill about .850 deep... second pass with a 2.0 end mill with full depth...

The sole exception to this method was a tapered driving shaft for a rotary compressor that had a "pocket type" .75 keyway tolerance (position and width) of something like +.0001.-.0002. The material was 4340 at about a Rc50 and we profiled the keyway in a large CNC machine using (rough and finish) carbide endmills.

Let us know how it turned out.....

????? Sorry, I don't agree-----

How so? In the hands of a skilled machinist, the odds of achieving proper size and location are far better when rouging undersized and then taking additional cuts. The slot will be on location, straight, and uniform, all of which are *very* difficult to achieve when taking a single pass, especially on light duty machines. Regardless of the opinion of anyone that is more interested in making money than doing the job properly, cutting keyslots in one pass is generally a poor way to get it done. Tool pressure is always a factor, as is chip load.

Harold

Ok, if you are going to introduce the concept of cutting a keyway on an undersized machine.... (or machine lacking rigidity) as I already posted.... more than one pass may be in order.

and that is a point.... a good machinist sets the machine up to account for pressures and chip load. Another point to consider is that a 1/2 drill and a 1/2 endmill don't cut on size.. if anything they cut larger than 1/2 inch.... especially if there isn't sufficient chip loading....

In the 35+ years I have spent as a machinist, I have found reasonably few instances where more than one pass was required... and my keys

... if your experience varies from mine... I suspect that it is due to variances in keyway size vs. machine mass....

...another arguable point is that an endmill is a poor choice for cutting a keyway, unless it can't be performed otherwise.....

After reading your last post, I realize that you and I aren't necessarily talking the same keyway. My experience is in job shop and missile industry, where cutting a keyway was not something that was done repetitively, or routinely, so, naturally, the tools used for the job were those that were at hand. The tools of choice for me, and the shops where I was employed, just happened to be end mills, and I suspect that unless most folks are engaged in machining key slots routinely, that would be their tool of choice, too.

While one may provide a little "Kentucky windage" when cutting a keyway, the amount of deflection is up for grabs, regardless of the type cutter used. Feeds and speeds create a myriad of conditions that can't be anticipated. I think my point is that those of us that prefer to "do it right" and not abuse, or ignore, tolerance, would not make the cut in one pass, nor would we expect it to be as good as one cut by our preferred method. Key slots are tight tolerance, too tight to risk cutting in one pass as far as I'm concerned. I agree, the key likely fits tightly when the seat is cut in one pass, but not because of the precision involved, but more so because of the variations in the key seat. Overall the key is a press fit, but is likely hitting on the highs alone. The key seat is likely not on (proper) location, not square, nor a constant width. That kind of work wasn't acceptable in the industry from which I came.

I took a quick look at your site. Nice job! I'm impressed with your background.

Harold

I'm getting in this late, I can tell, but as an old practicing engineer from a job shop background, I was taught that part being assembled (generally the pulley) should be a light push fit while the part over which the part was being fitted (generally the shaft) should be press fit to the key. With this kind of fit a single pass with an end mill will not get the job done on any kind of consistent basis. This kind of fit is the only way to prevent wallowing of fits under reversing loads..which are most kinds of driving loads anyway. The amount of time taken fitting the key, keyway and keyslot directly reflect on the life of the joint, as much or more than the fit of the bore. A one pass keyway, broached keyslot and a chopped off piece of keystock, filed to fit, will get a joint that fails that same as an oversized bore will fail. Some places, the joint requirements are not very high and an oversized joint can be tolerated but even then, it kind of reflects on the craftsmanship of the person making the joint. R. Wink

There certainly are different classes of work.... I've never actually taken a keyway/keyseat very seriously..... it isn't, by definition, a precision assembly, nor is it particularly well suited to do anything but bind two parts together that aren't subject to too much relative motion......

Having worked in the aerospace industry, myself.... I can't think of any critical assemblies that use a keyway/keyseat.... splines are generally employed for that purpose.

It isn't that a "precision" keyway/keyseat couldn't be drawn or milled... as I noted from the example of the rotary compressor drive shaft... it's just trying to make a silk purse out of a sow's ear.

Most applications will not use a ground key nor a precision broached keyseat. The keyway may actually be cut into a piece of material turned in a lathe and not finish ground. To give one of the (3) features undue attention without attending to ALL with equal precision....? The designer is just fooling himself. And, IMHO, to bother with all of that precision hoopla for a lowly keyed-type drive suggests that your money would be better spent on generating a spline....

My $.02.....

Blush..... thanks!

You're very welcome. You earned it.

I think it gets down to interpretation. I'm from the era where your workmanship is a mark of your ability. I was judged, often harshly, on my ability to do it right, per print, and avoid MRB if at all possible. That would include cutting the odd keyway, should a demand arise. We rarely tried to make a silk purse from a sow's ear, as you suggested.

Truth is, a keyway *is* (and should be considered) a precision cut. It should be a snug fit, not a sloppy one. It should be at right angles to the shaft, and should be centered properly. They may be treated poorly by the masses, but R. Wink hit it right on the head.

Keyways that are not snug are prone to premature failure. Cutting them randomly is not a great idea, nor is cutting them in one pass. That industry does so and gets away with it is more a comment on the level of lowering the bar than one of unimportance. It's like the food industry. How many rat turds are you willing to eat? Industry sets standards that permit a small amount. Personally, I'll have none, thanks. Same with machining. I take no shortcuts, nor do I accept anything as "good enough".

Mind you, I'm not trying to tell you how you should be cutting keyways. I'm simply stating that just because they're treated with contempt by many doesn't mean they are deserving of the treatment. Same philosophy as threads and thread measuring. I do it with wires. Most folks think I'm crazy. I know I'm not, not about threads, anyway!

Different strokes, apparently.

Harold

I think you and I are on the same page with this... I certainly wouldn't condone sloppy work... I was able to start my business and stay in business by beating my competition on quality....

Let's put some number to this.... digging through my old drawings...the first drawing relating to keyways that I came to was a

Keystock tolerance is

Frankly, I don't really consider + or - .0010 very precision work.... but, for keyways, it is industry standard.

After I posted, yesterday, I went to do some errands and a few attempts (by manufacturers) to get keys to work in reversing and/or cyclic loads came to mind. (1) Two keys.. on opposite sides of the shaft (12 o'clock and 6 o'clock) or positioned at 90 degrees (12 o'clock and 3 o'clock), (2) round... not square keys, and (3) my personal un-favorite, two keys at 180 degrees(12 o'clock and 6 o'clock) with the keys, themselves, rotated 45 degrees...... so that you have a "V" cut on the shaft and a "V" cut in the bore. That one had absolutely NO tolerance for error.... especially since the bore moved back and forth on the shaft (and keys).....

This is sort of crazy, to use an end mill when the correct tool (woodruff key cutter) is so widely available. In a horizontal machine one can do the job much more reliably and accurately. I wouldn't think of of using an end mill for the job.

Jim

Depends on the job, Jim. We did a lot of keyways that had to be full depth for their entire length, and 1/4" past the end of the key was an oil seal. Woodruff cutters aren't always an option. It's actually pretty common.

Greybeard.

What Greybeard said. You don't always have the option for the radius. Otherwise I fully agree.....and actually prefer a Woodruff key in place of a square one when it's possible The added benefit of using the Woodruff cutter is they typically cut the proper width--which is normally reasonably close tolerance.

Further, does every shop have every width Woodruff? I seriously doubt it-------but I have a generous supply of end mills, starting with 1/32" going up to 1". I can cut any size keyway with no trouble, on size, on location. I use my skills in place of tools.

Harold

LOL. Harold, the same trick you employ using the end mill works with the keyseat cutter. You use a narrower one and then just move it over to finish the cut.

Because I own a horizontal mill, I've grown rather fond of woodruff key cutters. They actually are handy for a variety of tasks.

Jim

Ayup.

Gunner

"At the core of liberalism is the spoiled child - miserable, as all spoiled children are, unsatisfied, demanding, ill-disciplined, despotic and useless. Liberalism is a philosphy of sniveling brats." -- P.J. O'Rourke