Speaking of Broaching - Are Your Bearings Really That Bad

Apr 17, 2026 Last reply: 3 months ago 7 Replies

More than once I've read or heard people say their bearings weren't good enough for broaching on their milling machine. Wait a minute? Do you drill with your milling machine?



Okay, lets take a step back. Sure pushing a stepped cutter through a round hole to make a square or a hex takes a fair amount of force. I typically do that on an arbor press if I can. Its maybe not as much as people think though. Each tooth only takes off a little bit of metal. It might be an insane amount of force in a fairly thick piece of steel, but in most things we might be broaching is it really that much.



Still, I probably wouldn't do that sort of broaching on a milling machine. If for no other reason than you can't (or shouldn't) push the broach through the table below the part. Yes, there are work around methods depending on your ability to fixture the job, the available clearance, etc, but that sort of broaching is best done on some sort of a press.



Now here is where the cognitive dissonance comes in. A lot of broaching can be (is) done with a single tooth/flute tool, and engagement is easily adjusted to a suitable amount of force within the limits of the material being broached. Internal keyways are often done this way ona lathe. Lock the spindle and use the tool post to push through a cutter. Its not a high impact application unless the operator choose to make it one. A sharp tool cuts even alloy fairly well if you get the right amount of engagement. Sure a lathe headstock/spindle is typically more robust than a mill spindle, but its still axial force on the bearings.



So why not the mill? Old? Not as robust? Shock load? Maybe to all of those things, but for a single point/flute tool engagement is a choice. Its axial load which unless its total crap is the direction in which the bearings are strongest. How much axial load are you putting on those bearings when you are forcing a drill through a piece of steel? You know you have to push hard enough so it removes a chip, because way to many alloys will harden up on you and stop cutting if you don't.



In any case you have to do what you think is best, but maybe what is best is a little more nuanced than you think. Maybe think of broaching on the mill like using a clapped out old shaper. You are the one who chooses how hard the cut is.



More than once I've read or heard people say their bearings weren't good enough for broaching on their milling machine. Wait a minute? Do you drill with your milling machine?

---------------- A voice of experience wrote that the -two speed quill feed gears- on a Clausing mill aren't strong enough for broaching, especially with a pipe extension on the handle.

Another sought and found replacements:

formatting link
"The gears in the quill can easily be broken."

If it needs a pipe extension that was a choice of engagement. Broaching a square corner on a mill is done by nibbling. Not firing a full size net shape projectile at Mach 5.

Maybe the gears are that fragile, but if they are how do you drill with it?

If it needs a pipe extension that was a choice of engagement. Broaching a square corner on a mill is done by nibbling. Not firing a full size net shape projectile at Mach 5.

Maybe the gears are that fragile, but if they are how do you drill with it?

-- Bob La Londe CNC Molds N Stuff

---------------------- I base how hard I push it on a deflection test with an indicator.

I start with a center drill, then 1/8", then increase in steps it can handle, perhaps 1/4->1/2->3/4->1" or hole saws up to 2" or slitting saws to

4" in mild steel. It makes chattering and other noises if pushed too hard. At ~700 Lbs it's not a very rigid milling machine, less so than an RF-30, and I don't risk breaking unavailable parts.

I like the ergonomics of my Clausing better than a Bridgeport's for making one small precision part but it's definitely not for rapid volume production, the same as my lathe.

I am just chasing dust devils here. I respect that you probably know best how to use your machines.

I am just chasing dust devils here. I respect that you probably know best how to use your machines.

-- Bob La Londe CNC Molds N Stuff

------------------------------ If my old machine tools were still competitive for making money they wouldn't have been available for the price.

They are a good match to my needs, which are often fitting a new part to an existing one that may be worn or damaged. For paying jobs once I've worked out the prototype design by cut and try I draw up the result and send it to a production shop, then return to my main tasks of electronic design and assembly.

Segway's ProtoTrak Bridgeport was no more useful on ill-defined custom work than my Clausing, and more awkward to operate when I could get time on it; I never got any on the CNC lathe. Their manual lathe was a Smithy Granite with broken milling gears, no competition to my toolmaker-equipped South Bend.

For serious broaching I make a guide and use the arbor press, for examples pulley hub key slots and ANSI splines to fit a hydraulic motor shaft. The broach for that was more work than it was worth, I should have bought a new keyed shaft motor of lower flow rate and torque demand immediately instead of later. That was the belt driven hydraulic pump for my tractor bucket loader and I had to overload it once by turning up the pressure to get unstuck, then fix what broke.

I've got a set of keyway broaches for pulleys and wheels. The trick seems to be a bunch of oil, and taking multiple passes. Start with zero shims, and just add one shim each pass. Its not to bad on any wheel's I've keyed.

Now for splined shafts and bores that's a lot more work. Just more work. Just a lot more temporal volume of work. That's where an indexing plate (could just be a gear with the right number of teeth or a multiple thereof) on the spindle of the lathe to lock it and index it, disconnecting the drive from the spindle, and letting the carriage do all the work might help if your lathe could handle it.

Of course you can make specialized holding and indexing tools for the press, but I'm not sure I would want to do that much extra work.

Or you could just get a giant vertical slotter. LOl.

I've got a set of keyway broaches for pulleys and wheels. The trick seems to be a bunch of oil, and taking multiple passes. Start with zero shims, and just add one shim each pass. Its not to bad on any wheel's I've keyed.

Now for splined shafts and bores that's a lot more work. Just more work. Just a lot more temporal volume of work. That's where an indexing plate (could just be a gear with the right number of teeth or a multiple thereof) on the spindle of the lathe to lock it and index it, disconnecting the drive from the spindle, and letting the carriage do all the work might help if your lathe could handle it.

Of course you can make specialized holding and indexing tools for the press, but I'm not sure I would want to do that much extra work.

Or you could just get a giant vertical slotter. LOl.

-- Bob La Londe CNC Molds N Stuff

------------------------------

That is good advice and I've done them. I bought a few keyway broaches and make the bore guides as needed, used a 54 tooth change gear to index the shaft to mill for the 13 spline motorcycle sprocket, and the lathe chuck jaws IIRC to index the brass carb jet hex sockets, broached with the carriage feed one point at a time with an undersized Allen wrench cutter. The back gears locked the spindle. Afterwards I bought an indexing wheel and bored it 2-1/4" to fit the lathe's threaded spindle, clamped under the chuck or 5C spindle thread protector.

Although I could buy a full broach set I consider making parts like bore guides to be valuable practice in turning a plug to accurately fit an inch or metric hole, plus some are small enough for the job interview portfolio. These old worn machines need understanding and gentle persuasion to make them do what I want, and I need practice to achieve 0.001" or better accuracy.

Milling the ANSI spline broach was an example of taking on a project I shouldn't have, although the broach worked, but the die cast pulley didn't hold up. I've often been on the border between buy or make which can be difficult to decide without the experience of making it once.

I've only rarely worked production where speed mattered, in R&D usually I did something new, difficult and demanding once or occasionally. This is typical in my largely self-taught electrical engineering.

formatting link

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required