Welcome back to the machine shop for the dramatic conclusion to the rotary table chuck adapter build! Two videos ago I went about the process of reverse engineering the D1-5 cam mechanism of my metal lathe and developed an adapter design using the drafting table. Then in the last video I underwent the emotional rollercoaster that was machining the main body. So in this video I’ll be finishing up the remaining bits and bobs to make the mechanism work. Along the way I’ll be trying a few new techniques including scallop milling with the boring head. Turning an eccentric cam profile with the 4 jaw chuck. And one of the biggest side projects to date, making my own rotary broach to machine square holes!
AI Generated·May contain errors
Running machine tools takes a lot of forethought and planning to ensure that
the part you want to make comes out as you'd hope.
Generally the more complex the part, the more complex the plan.
But sometimes those plans put me in the realm of unfamiliarity.
So it never hurts to have a backup plan. Though it is a slap in the face when you
have to use that backup plan, and it works way better than your initial idea.
And it's a double slap in the face when it takes a fraction of the time.
How's that saying go? Can't see the forest for the trees?
Well I'm about to find myself deep in the woods.
Over the past few weeks I've been working my way towards making an adapter
to mount my various lathe chucks to my rotary table.
I started with a pretty deep dive reverse engineering the D1-5 cam lock
nose on my lathe. And for giggles I drew up the design
using the drafting table. With material in hand, I began machining
the main body of the adapter. And after a roller coaster of bad
decisions and successful repairs, that part is complete.
Which brings me to right now. I still have the cams, retaining screws, and t-nuts
to finish up. So I'll start with the most complex parts
first. The cams. I've been especially looking
forward to making these since there are a few new operations for me to try out.
Namely turning the cam on an eccentric axis and forming the square hole with a
rotary broach. But of course there are a few steps
before I can get to these. I'll start the usual way by facing the
steel and turning down the diameter. These will need to be a clearance fit
for the bores I made in the adapter body last time.
Which, if you've been following along, you'll know are just a little bit
oversized from my original plan. With the diameter set, I'll move on to
preparing the end for the rotary broach. I'm drilling a pilot hole right at the
same diameter as the final square hole flat to flat distance
in hopes of maximizing the size of those flats.
Now the bottom of this hole needs an internal groove for that chip to break
away when I run the broach through here. But of course I don't have a tool to do
this. I guess I'll just have to make one. This should do the trick. I basically
made a tiny little boring bar with a square cutting face.
It has a bit of back rake to help the cutting edge do its thing, and all the
other material removed was to get the necessary clearance.
After mounting the tool in a tool holder and setting the height on centerline,
I'll zero on the face and ID, and then slowly start cutting into the material
at the back of this bore. One last step before broaching the hole
is to give this a lead-end chamfer. This needs to be a bit larger than the
point-to-point distance on the square hole, and will help the broach get
started. Now for the rotary broach. Well, sort of.
Rotary broaches typically have two parts, a cutting tool and a spindle.
Luckily, amongst my grandfather's tooling, there were several identical
live centers with interchangeable ends. They also look suspiciously like a
So I'll save myself a whole lot of work by sacrificing one of these to the cause.
The morse taper end is meant to be held perfectly straight in the lathe tailstock,
but I actually need to hold this at a one-degree angle for it to work.
So instead I'll be mounting this on a one-inch boring bar holder on the tool
post, which of course means I need to turn
the morse taper down to a cylindrical shaft.
After breaking this down to verify I won't be tearing into any of the
internals, and also testing that this isn't hardened
steel, I'll mount this on the four jaw chuck and
indicate the tapered end true, then start cutting away the taper one
pass at a time. I've got a pretty large stick out here,
and no convenient way to add a center support to the end,
so I'll keep the passes light to minimize deflection. I'm not sure what
kind of steel this thing is made from, but that finish is darn sexy.
And the taper's within a few tenths, which is far better than I actually need.
I'll reassemble all the components with some fresh grease,
slide this into the tool holder, and now the fun really begins.
I've got this old battered high-speed steel end mill that's just been begging
to be repurposed, so I'll be grinding this into the
rotary broach on the surface grinder. And if it wasn't already obvious enough,
I'll be mounting the broach in the tapered end of the live center where the
interchangeable bits normally go. I just need to figure out what that
taper actually is. After some quick measurements and
sifting through the standard machine tapers,
it doesn't appear that this is anything close to either the Brown and Sharp,
Jacobs, Jarno, Morse, or even Type B tapers.
Then it occurred to me that this tooling was made overseas, and they likely used
metric wherever they could in the design. And sure enough, the taper is exactly a
30 to 1, with a 15 millimeter large end and 30 millimeters long.
So this is what I'll have to replicate on the high-speed steel. To start, I'll
chop the ends off this end mill, since I really only need the middle section.
I need a way of holding this at a precise 30 to 1 angle,
which is a perfect job for the magnetic sign plate I made in a previous series.
To hit the 30 to 1 angle on an 8 inch center distance of this plate,
I need a gauge block stack of .2667 inches. I'll place these on the sign plate,
lock it down, and then comes the next piece of the puzzle,
the collet indexer. This will give me a convenient way to hold the high-speed
steel and also rotate it as I grind the taper.
I know from some previous projects that this indexer has a tiny amount of runout,
but I'm just going to cross my fingers that this won't cause any issues down the road.
Setting this up on the surface grinder reminds me of just how unnecessarily large
this sign plate actually is. With the wheel all the way up, it does fit, but just barely.
The last bit of setup is to center the axis of the wheel with the collet indexer.
To do this, I'll bring the wheel down a tenth at a time and pass the part by the wheel.
As soon as I see a hint of sparks, I know I'm reasonably close to aligned,
and I can lock the table at this position and finally start grinding.
The process here is pretty straightforward. I'll start each pass beyond the end of the material,
then rotate the indexer at a consistent speed as I slowly feed the part under the wheel.
Once at depth, I'll shuttle back off the part, lower the wheel a few thou, and start again,
then repeat this step about 50 more times.
All those passes have done a number on the wheel, so before taking the final cuts,
I'll give the wheel a quick dressing to square it back up,
then remount the indexer and take the last few passes.
All right, time to see if that actually worked.
If I did this right, the taper should lock into the spindle.
Yep, I'd say that's locked all right.
With that part of the tool out of the way, it's time to tackle the business end.
Because I'll be running the broach at a 1 degree angle of attack,
I'll need to form the square end with a clearance angle greater than that.
I'll set the sine plate up at 1.5 degrees,
then mount the collet indexer the opposite way on the mag chuck,
as I'll need the opposite taper on the cutting end.
Once set up square on the mag chuck, I can start an even more tedious grinding process than before,
this time around using the collet indexer as an actual indexer.
For each wheel height setting, I'll take 4 passes, indexing 90 degrees to each face,
then bring the wheel down another 5 thou for the next 4 passes.
It's important to work in sequence like this rather than one side at a time,
so that I guarantee that the square end stays as concentric as possible to the rest of the tool.
The consequence of this, though, is that I have to be very careful not to rotate to the
wrong marking and possibly scrap the whole part.
I'm not exaggerating when I say this one operation took close to 2 hours to complete.
Just like last time, as I got within the last few passes,
I stopped to redress the wheel before taking the tool end to the final dimension.
Now that's looking pretty nice if I do say so myself.
But I'm not quite finished.
I still need to put a rake angle on the front face of this tool,
or else I'll just end up bulldozing my way through the material,
likely breaking something in the process.
So for this very precise and delicate operation,
I'll be using one of the finest tools in my shop,
a Dremel hose clamp to a tool holder.
Now I know you might be wondering where I came upon such a rare tool.
Believe it or not, I found it under a Christmas tree the winter of 05.
And it's been my trusty tinker companion ever since.
Did I seriously just jinx myself?
Unbelievable.
Let's see what's going on here.
Nothing looks melted.
Hmm...
One of the power wires is a little loose.
I bet that's what the problem is.
My trusty companion lives.
You might have picked up on this,
but I'm kind of just winging the geometry here
and going solely based on looks
and comparing to other rotary brooches I've seen online.
Once I clear the face past all four corners, that's it.
The brooch is complete.
Talk about a side project.
Probably didn't look like much to you,
but that was about two days of work.
Time to see if those efforts will pay off.
I'll remount the cam I started,
then set the tool post to a one degree angle
by locking the tool post square with the compound set to zero,
and then adjusting the compound to one degree,
which should be close enough.
The far more important part of this setup
is to get the end of the brooch
as concentric to the lathe spindle as possible.
If this is too far off center in either direction,
the side loads will be severe enough
that I risk ruining the part
or breaking the brooch altogether.
To do this, I'll mount an indicator on a mag base
right to the spindle itself
so I can sweep it around the sides of the square brooch.
It's important that the indicator is reading
right at the end of the brooch,
since the one degree angle it's set at
could throw this off.
Once I get within one thou of the same reading
on all four positions, I'm finally ready to cut.
I'll get the spindle going,
drop a little oil on here,
and we're rocking.
Well, we were rocking,
but it seems to have come to a halt.
You can see where it's clearly started,
but it looks like there's just too much material to remove.
Turns out the recommendation I found online
is to oversize the pilot hole for just this reason,
so I'll open it up another 50 thou or so.
Since I need to pick up where I left off,
I'll manually align the brooch with the first attempt
before turning the spindle back on.
All right, I think I can feel it moving this time.
Hang on.
That's not the brooch that's moving,
it's the material.
This is just a little too much force for the collet chuck.
I'll try this again with the 3 jaw,
which should solve that problem.
I'll also run this at a much higher speed than before.
Hey, hey, it's going!
It's doing the thing!
All right, that turned out awesome.
And the wrench seems to fit perfectly.
But how did the tool fare?
Now that, I'm much, much less stoked about.
It looks like all four corners of the brooch chipped off slightly,
and I have five more cams to make,
so this could be a problem.
I'll definitely need to regrind the face of this before the next one,
but while I work out a better approach,
I'll keep chugging along on this first cam.
The last concentric operation on this part is the retaining screw groove.
Rather than grind another grooving tool,
I'll just cut this away in three passes using a small parting blade.
Now I can cut this off long enough to potentially make another cam on the other end.
For the next operation, I need to set the part up off-center
in order to form the eccentric cam profile.
So on goes the four jaw chuck once again.
All right, I can't stand it anymore.
Hey, Brandon.
That's a really nice shirt you've got on there.
I'd sure like to have a cool shirt like that.
Do you know where I could find one?
Oh, there's a link in the description to all new Inheritance Machining merch.
Awesome, thanks buddy.
I'll let you get back to it.
Now this eccentric cut can't just be in any old spot.
It needs to be at a specific angle to the square hole we just broached.
To get the angle set relative to the chuck,
I'll clamp a 1-2-3 block to the jaw faces
and use a protractor head to align the square hole.
This isn't a precision technique by any means,
but it should get me plus or minus a degree or so, which should be all right.
Now I can start dialing in the offset for the eccentric cam profile.
It may seem a little odd,
but the first step is actually to indicate the material concentric.
Once centered, I can pick two opposing jaws to make the offset for the cam profile
without touching the other two jaws.
This, in effect, lets me maintain the angular position of the cam profile
relative to the square hole.
All right, let's do this.
As you'd expect, this starts as an interrupted cut,
which I've heard is not the best for insert tooling.
But I think because the interruptions have a gradual lead-in,
it's not too hard on the insert,
despite the racket it's making.
The further I go, the less noise it makes
until I'm past the outside diameter
and just turning a continuous shaft like normal.
After verifying I'm at the right diameter,
I'll give this a quick polish
before breaking it down for the milling operations.
Of course, all the other features need to be aligned at right angles
to the square hole and cam profile I just cut.
So I'll set this up in a collet block
and use the square to align the drive end with the block faces
before cinching everything tight.
With the collet block clamped horizontally in the mill vise,
I'll use an edge finder to locate the cam end and center line.
To cut the clearance scallop on the side of the cam,
I'll be using the boring head.
Because I won't have any good way to actually measure
the radius of the scallop once I start,
I'll have to set it up before I start cutting.
Clamping a scrap piece of aluminum off to the side,
I'll drill a slightly undersized hole.
Then switch back to the boring head and start making passes,
making adjustments until I hit the desired diameter.
Now as long as I don't change any of the settings on the head,
I can be reasonably confident that the scallops on all six cams
will come out the same.
Back at the cam, I'll take successively deeper passes
until I reach the correct depth.
Yep, that should do it.
Switching to a 316's end mill,
I'll cut a slight detent on the retaining screw groove.
This wasn't part of the original plan,
but will be a nice feature that keeps the cam in a neutral position
when it's not engaged with the chuck.
Then I can stand the collet block up on its end
and rotate it 90 degrees to cut the alignment mark
on what will be the exposed face.
And then the cam is nearly complete.
I just have to part this off on the lathe,
face the last end to length,
and drill a hole for the spring.
And that's it.
Of course I still have five more of these to make,
but it seems like a really good idea to make sure
this is going to work before I spend the time doing that.
To test it, I'll drop the cam in one of the holes of the body.
Place it on a test chuck and...
Well, sh**.
That is definitely not supposed to go all the way around like that.
I can feel it starting to get a little snug here,
but then the cam just keeps spinning instead of getting tighter.
Something isn't right here.
I started to recheck my measurements
when I remembered that I took all the studs
out of this drive plate in the last video.
I bet I know what's going on here.
Yep, false alarm.
I just didn't have the studs adjusted to the correct positions.
With the stud locations corrected, the cam now locks perfectly.
Man, that's a relief.
My heart couldn't have taken that.
With renewed confidence that the first cam is right,
I'll start the process of making the remaining five.
This time around, I'll do this in more of a batch style process
to try and economize my setups.
I'll start by doing all the turning operations first.
And also to save on scraps,
I'll make them two at a time with a center section
I can grab with the chuck on the following operations.
While I was at it, I went ahead and prepared six.
You know, just in case.
I'll quickly regrind the face of the rotary broach
to clear up the chipped edges from the first attempt.
Then begin the process of realigning the broach tip
with the spindle center
and working my way through the remaining cams,
broaching them one by...
Oh no.
If you're not sitting down already, you might want to.
A glorious broach that I spent a good six hours making
is a broach no more.
It seems the cutting tip had enough of my abuse
and decided to seek refuge deep inside the end of the third cam.
I just love it when two days of work gets flushed down the drain.
If I had to guess what happened,
the corners probably started chipping again mid-cut.
Then the uneven side load was just too much for the broach to handle,
so it snapped off.
I don't have time to make a new one,
so I'll have to unleash the backup plan.
A mill and a file.
I'll get a cam set up vertically in the mill
and locate it on the spindle center with a coaxial indicator.
Then I can take plunge cuts at each of the four corners
with an 1⁄8 inch end mill.
And also a quick pass around the perimeter to square up the sides.
From here, the remaining material on the corners
will need to be filed away for the wrench to actually fit.
Or not.
You gotta love it when a five minute operation can replace hours of work.
On the bright side, at least I won't have to spend any time filing these.
With a new plan of attack,
I'll knock out the remaining square hole milling operations,
turn the eccentric cam profiles on the lathe,
jump back to the mill to cut the scallops, detents, and alignment marks,
and finally part, face, and drill the ends.
Then voila!
All six cams are complete!
But don't get too excited,
there are still a few more parts I need to whip up real quick.
The retaining screws are just standard socket head cap screws
with a cylindrical feature on the end.
To cut these, I've machined a counterbored holding block
that I can thread the screws into.
This gives me something substantial to hold onto in the lathe.
I've turned this to a specific geometry
so I can use the end of the block as my zero reference
to face all the screws to the same length before turning the pin features.
Alright, that's six cams and six screws.
Let's assemble this to see if everything works out as I hoped.
Here you can see the detents I added have a good positive engagement.
This will help keep the cams in the neutral position,
allowing me to more easily mount the chuck.
Oh yes, very nice.
I'm pretty consistently getting a quarter to three-eighths of a rotation
on each of the cams before they lock down.
This is exactly the same engagement range I get on the lathe cam
lock I copied this from,
so that's about as good as I could have hoped for.
Before I can mount this to the rotary table as it's intended,
I have a few more pieces of hardware to knock out.
Those being the special rounded T-nuts.
I'll drill and tap holes for the three nuts I'll need,
then move over to the rotary table to cut the T-profiles and rounded ends all in one setup.
Now that's literally everything made.
I still need to harden the T-nuts, and especially the cams,
to prevent them from deforming and wearing after repeated use.
But I'll have to save that for another time.
I don't think it'll hurt anything, though,
to mount this up on the table real quick and take it for a test spin.
Pun intended.
While that looks about like I expected it to,
albeit the whole thing is quite large,
I seem to have a habit of doing that.
It also feels quite stable,
but the true test is to see how much runout there is.
That's actually not too bad,
considering I'm running this on the outside of the chuck,
which isn't exactly a reference surface.
Overall, I'd call this project a success.
I've added a whole other level of functionality to the shop that I didn't have before,
and even if I don't end up using this as much as I thought,
the lessons I learned along the way were worth the trouble all on their own.
But I am pretty confident that this tool will be seeing some use over the next few months,
so keep an eye out for that.
As always, thanks for watching, and see you next time.
I just have to say that is a mighty nice shirt you got there.
I distracted myself with that.
I just have to say this.
Hey Brandon.
F**k it up again.
You think we got it?
I think we got it.