Friday, October 10, 2014

Making the Rotor Turner

Mounting the old fashioned way
When you go to the shop to have brake work done they will sometimes come out, wipe their hands on a rag, look you in the eye and say, "Well, your rotors are looking a little worn.  We need to turn them or replace them."  At that point, get ready to hand over $100+ per rotor to get them to do it, not counting remove and replace.

We have several vehicles needing brake work. We are doing it on the cheap, so we didn't want to buy rotors if we didn't need them and we didn't want to pay to have them turned.  Instead, I decided to turn the rotors myself.

This means that I have to mount the rotors on the lathe, center the rotors and then make the runout as close to zero as possible.  I'm sure Adam Booth could do this with a four jaw chuck in 3 minutes but I'm not Adam, so this is going to take a bit longer.

It took more like 5 minutes to get it centered, then a few more to get the run out to zero. Then, just as I was ready to go, the surface that I was using to register against crumbled and the run out was off so much it all had to be done again. With two rotors still to do on the car and at least two more cars to go, 14 rotors total, it was time to re-think my methods.

Question: How can I quickly and efficiently mount the rotors on my lathe such that the runout is acceptable, and then turn them?

Answer: Build a mounting plate for the rotors.

After a little thought, I decided on a two piece mounting system.  The first was a mandrel that could be quickly mounted in the lathe chuck, 3 or 4 jaw.   If in the 4 jaw, it could be quickly dialed in to have near zero run out.  The second part would be a mounting plate.

The mounting plate would have a threaded center with registration lip to allow for it to be removed and replaced and always go back to the same alignment.  In addition, the mounting plate would have a registration hub to register the rotor against for rotational alignment and the surface outside the hub would faced to bring the surface perpendicular to the axis of rotation.

When the tool is dismounted and remounted, a very light surfacing will bring that face back into perfect alignment again.

1.5" 1018 bar stock - the big pole.

Thus we start with 1.5" 1018 bar stock, or as my lady called it, "The big pole which will become the holder thingie."

0.375 A36 plate, 18.48

 And a piece of 0.375 A36 plate, 18x48.  We'll cut a strip off this and then turn that into our mounting plate.  And to use my lady's words "The plate that becomes the flat part of the holder thingie."

Here is Liri hard at work cutting a 6" wide strip from the plate.

Using a cheap Sawsall we cut a strip off the plate.  While I have a horizontal bandsaw, I do not have a vertical one so no way to cut this except with hand tools.  It cut very easily and I was able to do the first seven inches in about three to five minutes.

Of course it took Liri another 15-20 minutes to cut the remaining 11 inches.

The A36 cut very quickly and easily.

The four jaw was the right tool for the job.
While Liri was working on the plate, I cut a five inch slug off the bar stock, mounted it up in the four jaw, and got ready to spin it up.  Fortunately, the horizontal bandsaw did a quick job of cutting the mandrel blank. While I could have done this with a three jaw chuck, I knew that the next step would be to chuck up a squarish chunk of plate.  That would require a four jaw.  As such, it was faster and easier to just leave in the four jaw and work with that.

With the new Noga mag base, it only took a few moments to get the mandrel centered up and ready to turn.  I do need to make a new chuck key for the four jaw.  The key I have is a little small.  If I do make a chuck key, I'm likely going to make two:  a small and a normal.

When centering using the two key method, you take a measurement on both sides of the chuck.  You figure out where the center should be and then turn both keys without rotating the chuck until the piece is in the middle. 

People using this method report that they can center the work in just about the same amount of time as using a scroll chuck.  Of course, a better option for me would be a nice 6 jaw adjust true chuck.  Yeah, at about $1000-$2000 without a backing plate.

The mandrel is faced.

Here I've faced the mandrel, turned a surface to indicate against, turned a registration ledge.  After that, the end was turned down to the outside diameter for a 1.25-12tpi.  An exit groove was created for the single point threading tool and everything was set up for cutting the threads.

The 6" x 6" plate with center marked.

Having spent 20+ minutes cutting a 6" by 18" strip of 0.375 plate, we dropped it in the horizontal bandsaw, turned on the saw and just a few minutes later a 6" by 6" plate was delivered to us.  This plate was then marked so we had an idea of where the center was.

Plate mounted in the four jaw.

Here the plate has been mounted in the four jaw chuck, centered and we've started by drilling out the center to 3/4 of an inch.  We'll further bore out the center until it is the right size for a 1.25-12tpi internal thread.

Our photographer was sick so we missed all those fun steps.  Part of the process was grinding a threading tool to mount in my 5/8 boring bar.  First time for that, first time using the boring bar.

During the threading process I managed to knock the boring bar out of the tool holder.  At that point it took a few minutes to get everything set back up but I did manage to get everything aligned and was able to finish cutting the threads.

(Please note the doubleboost light there in the foreground.)

I had machined the mandrel first so I was able to test fit the plate to the mandrel to check the threads.  Once the threads were correct, I cut the registration ledge.  At this point, I know that I should have made a test plug for the registration ledge and cut the registration ledge first, before threading.  Doing it in the opposite order allowed me to avoid cutting the test plug but did cause damage to the threads that had to be repaired for each test fitting.

Once the mounting hole was completed in the plate, the plate was removed from the four jaw chuck and the mandrel mounted.  I could have switched to a three jaw chuck but I decided to continue with my four jaw learning curve.

With the plate mounted and spinning I started the process of rounding the square.  This is an interrupted cut.  I'm still not sure what the best way of doing such a cut is but that A-36 (machinability of 72%) cut easily even if it was throwing very hot chips everywhere. Each chip left the cutting tool at around 600°.  I had them landing on my head, bouncing off my safety glasses, landing on my arms, and more than a few going down my shirt collar.

Finishing the plate.
Once the plate was rounded, it was time to finish it.  First step was a light facing to bring the face true.  After that the inner hub was created to register the rotor when mounted.  Then a relief groove was cut just outside the hub to hold a raised ridge on the rotor.

With all of those operations completed, the plate was dismounted and layout dye applied.  The rotor was put in place and a transfer punch used to mark the center of the lug bolt hole in the rotor.  With that mark in place, it was a simple mater of drilling that hole on the drill press.  I could have used the milling machine but it was faster/easier to just use the drill press.  Once the first hole was drilled the rotor was fastened to the mounting plate and center punched for the remaining holes.  It turned out that the layout dye wasn't needed as the transfer punch left a very clear mark.

After all five holes had been drilled, the mounting plate was screwed back on the mandrel.  Because the mandrel had not been moved there was no need to reface the mounting plate.

A mounted rotor, ready for turning.
Now we have the first rotor mounted up and ready to be "turned".   You can see how the registration hub is holding the rotor centered while the back of the rotor is firm against the mounting plate.  There is no wiggle or wobble in the system and after starting up the lathe, while standing well out of the way in case it all came apart, it was time to turn the rotor.

Rotor, turning at speed.

Here is the rotor turning at speed.  There is cast iron dust going everywhere.  I did place a rag over the cross feed screw to help protect it a little bit but clean up is a pain.

This is one of the first faces cut.  Later faces looked much better as I changed out my tooling. 

Here we can see the huge piles of iron dust generated by the facing operations.

This was both a useful project but also a great learning project.  I've been spoiled by using 12L14 (machinability of 170%).  The 1018 of the bar stock and the A-36 plate do not machine as nicely.  This shows in the quality of the surfaces.  In addition the chips on the A-36 were ugly, nasty, stringy, evil things.  They remind me of the time I got to play with razor wire.  They feel like a sliced finger waiting to happen.

Grinding tools shouldn't scare me but I do need to get a wheel dresser.  This is a tool that takes the gunk out of a grinding wheel, it makes it square and exposes new grains of the wheel to cut faster and cleaner.  At this point I'm not going to "fear" making my own cutting tools.  I've made the threading tool, a round faced forming tool, a boring bar and a few other tools.  My big need is to get out my stones and hone my tools to clean up the cutting edge.  Not that big of an issue with this project, but one that I have to address.

Cutting threads is fun but I've got to be more careful in my prep work and in measuring.  I'm pretty sure that any real machinist would have thrown the work out as "ruined" for the slop in my measurements.  My only saving grace is that these are matched parts.  As long as all the rest of the parts are made to fit, it doesn't matter.

The Machinery's Handbook has everything you will ever need.  Finding what you need in that 2800 page work is a different story.  I had actually spent the time using the formulas to figure out the specifications for the 1.25-20TPI threads.  Later I was looking through tables and found that for 1.25, "fine" was defined as 12TPI and that was in the book.  There was also a row for 1.25-20TPI for "Ultra fine" but in looking at the book and a bit more investigation in the book I found that 12TPI was a better choice.

While the total amount of time spent on the project exceeds 6 hours, I feel that it was worth it.  The first two rotors I turned were "ok enough to pass inspection" but they did not turn true.  With my mounting tool, it is possible to take the rotor off the car, mount it up, turn both sides and have it back on the car in under 30 minutes.  Without the mounting tool it takes closer to an hour.  With 12 more rotors to turn, the tool will save me time in the long run.

While a two part tool with threads is a bit complex, the ability to make other plates is a great thing.  In all likelihood, I'll turn two more plates to have the mounting hole in the center and then rounded ready to use in the future.  With that work done, it really is only a matter of a few minutes to make a custom mounting plate for any set of rotors.