Want to learn how to make better pullies? ETA, Whoops, I made a hockey puck!

[RaisedByWolves]

I’ve been here a while now, and have seen the subject of making pullies discussed both here and on the net a bunch of times.

The conventional thinking is to turn the bore and the sheave portions in one setup for the ultimate in concentricity, and you may agree with this and its ok if you do.

But there is a better way, let me explain.

When you turn out a pulley from stock, boring the center and cutting the v for the belt, it’s true that you make the most concentric part possible.

This part stays concentric right up until you tighten the set screw down on the key in the keyway. Once you do this the pulley is thrown off center by half of whatever clearance you left for it to slide over the motor shaft/drive shaft.

This may only be a couple thousandth or so and if in you’re use case this is acceptable, you can stop reading here.

For the pulley I need and in certain other similar cases you might want something better, if this interests you read on.

The pulley I’m making is for a nice little mini lathe I’m working on. This little lathe is accurate and mostly very nice right out of the box, but like anything else there were some corners cut either as a cost saving measure, a production requirement or worst case, both.

So what we have here is a Soviet bloc Bakelite motor pulley. Now the communists love their Bakelite. It’s cheap and easy to manufacture parts out of, and it has the added bonus of being dirt cheap when you stuff the mold with whatever random garbage you have laying around.


Here is an AK bayonet I bought due to being able to see the nylon string inside the molded surface.







This is a new 25yr old pully from my lathe that I'm replacing that is flaking away at the edges.








In the above pics you can see what looks like felt or old mattress ticking imbedded in the mix but I have seen nylon string, saw dust and wood chips etc.

This yields a part that not only is brittle, but also Has the benefit of being’s dimensionally unstable.

This means that if taken out of the mold too soon or if there is too much crap thrown in the part will deform over time.

For a clock radio or a knife handle this will not matter, but for a motor pulley on a high speed precision lathe it matters a lot. Considering the motor I'm using is turning 2500rpm, almost double what the factory motor turned, I'm not sure this pully will even hold up long term.

This pulley was so bad it shook the whole lathe enough that indicating a part under power was all but impossible, the indicator needle was just a blur before ever touching it to the part.

So I decided to make my own pulley that was not concentric with it’s self, but more importantly concentric with the motor shaft it’s running on.

This will result is a pulley that runs much truer when mounted on the motor and is accomplished by doing the final turning on a mandrel.

Here I have already bored the hole for the motor shaft, broached it for the key and added a set screw that will engage the key seated in the motor shaft. Along side this is the piece of steel I will make the mandrel out of.






Now I’m turning down this large piece of steel rather than using something closer to the 15mm mounting hole/motor shaft, as I will machine in a register for the pulley to…., we’ll, register against to eliminate wobble and vibration during the cutting process.


Here I’m checking clearance where my final depth of cut will be in regards to the tooling and the chuck and setting the carriage stop to this distance.





What I didn’t get a shot of is that I retracted the compound so there was no overhang and also centered the cutter tip over the compound's ways so there is little risk of deflection as I will be taking HEAVY cuts to hog out the material as fast as possible.

And for this you may have noticed I’m using a round carbide insert tool holder. These things are awesome for removing material fast and efficiently without creating a lot of heat.


Beauty shot. I think these inserts are ground and would be considered “aluminum” inserts, but they cut steel just fine and put up with all of my abuse.







Here I’m just starting out by facing the part and you can see another advantage of the round insert, I can face and turn in one setup.








I love these inserts as it only took me 6-8 min to turn this 3” piece of barstock down to 15mm at 1.5” length. Wouldn’t even have taken that long but I had to creep up on the feed to figure the best feed setting as I was taking .200 D cuts. This really isn't even a heavy cut for this setup, but I will be using this same insert for turning the pully so I wanted to preserve the edge as much as possible.

You can see in the video I'm interrupting the cut every now and then also. And with cuts like that you can turn out nice lil purple catapillars that just drop right down into the chip tray.






So when the smoke cleared I had my mandrel and it looked like this. The part with the bluing remaining and the shiny spot near the mandrel corner? Are undercut to ensure that only the mandrel body and the register come into contact with the part. You can also see the shaft has been drilled and tapped for a 5/16-18” cap screw.





Now, this is where the rubber hits the road as far as making this an accurate turning process given the fact that this is a keyed pulley.

In the below pic you can see a piece of shim stock fitted into the key way. The sole purpose of this is to give the set screw tip something to push against and not mar the shaft which would leave a nasty burr on the precisely bored …..umm….. bore. This piece gets pushed into the keyway fully so the cap screw and washer can seat against the face of the part.





The set screw gets tightened down against the shim and then the 5/16” locking bolt gets tightened down.






Now I have the part trapped both radially as it will sit on the motor shaft and also axially and I'm ready to go.

 

[RaisedByWolves]

At this point I was ready to begin hogging material out of the blank I had made from aluminum bar stock. I had to start with bar stock as I did not have aluminum round stock the diameter of the large side of the pully.

Here I'm taking .250 cuts with the same insert I used on the steel and on several other projects and man, look at that finish! Love these inserts.





After a bit I had my blank roughed out and brought to size and I was ready to get to work on the important details.






At this point I realized I had made somewhat of a boo boo, I left the pulley I’m copying at home.

Well, it was crap anyway and as they say, lemons/lemonade.

This is the type of situation that separates the men from the… ah… other men. You goof, blunder, and have those few moments of despair (Why was I born like this!!!?!1!!!) and you need to find a work around.


Turns out what I have on hand to copy is much better than the plastic fantastic pulley I originally copied anyway.

I brought in the middle pulley (first pulley after motor pulley) to check my form tool angles. This was done as I took my form tool home to do this check and found my initial grind to be off. I looked up v belt sheave angles to grind the tool initially and came up with 40* for the form tool.

This was close, but when heald up to the known good aluminum sheave it was noticeably off. I reground it to 34* and it was perfect.





Fun fact, belt and sheave angles change in relation to the size of the belt. Learn something new every day. I knew there were different sized and classes of belts, but I did not expect this.

Now, I still had the problem of having measured one crappy deformed sheave and not really trusting it, and having to find a zero point to begin cutting the new grooves.


I had this all drawn out on paper, but not having the bad sheave in hand to reference I decided to change plans and go with what I had.


I decided to begin cutting the larger groove first and using this as my zero, as I already knew the set screw hole was mapped out to be centered in the large groove, as this was set central to the width of the entirety of the part.


So I picked this up with the form tool and decided to just “work it in” as I cut. While I knew my tool width to zero off of the edge like normal, I couldn’t get my form tool to touch off due to the cap screw and washer holding the pulley to the mandrel being in the way.

Now overcoming this series of misadventures is the thing people call experience. It’s not that the guys in the trades don’t make mistakes, it’s that they gain enough confidence over time that they trust themselves to overcome these stumbling blocks.

Ie they have learned to deal with the problem as presented and not get head cramps.

The key is to trust yourself. You will either get it right or you will learn from your mistakes. It’s the only way.

At this point I didn’t have a firm reference for my “zero” but it really doesn’t matter as I’m making the part and therefore I get to say what’s critical and what’s not.

The only thing that really matters is that the v groove centers are the correct distance apart in order for the belt to track true, and once that is established I can take artistic license (read fudge) the rest of the part.


Here you can see I’m using the drilled clearance hole for the set screw as my guide for where the zero point is, and I’ll just adjust this cut as I progress until it is centered.






And to keep track of where I am, I have a digital indicator on the lathe saddle to set my zero once I find it, but also to give me my center distance between the two grooves once I measure that from the “good” sheave.





When this groove was finished there was still some of the set screw clearance hole left exposed, but I’m not really worried about as it is on the “high speed” side of the pulley and will not be in play for anything critical like threading or indicating a part under power.






To get my distance between centers I put the good sheave in the vice of a B-port and used a dowel pin to pick up the center of the V. Now you must think a dowel pin in an r8 collet will have a bunch of runout and make my measurements erratic.

You would be right and you would be wrong.

The dowel and collet in combination do indeed have a noticeable amount of runout to even the naked eye. Thing is I don’t care about that as I’m not rotating the pin, so the tip stays put in the same place regardless of this. (Lightbulbs going off in heads everywhere)






Picked that dimension up and checked it twice and it came out to .512 which is close enough to 13mm for my liking. Told you the runout of the dowel pin didn’t amount to much.


Using the aforementioned and zeroed indicator I moved the saddle and began cutting the second groove. Had to increase the spindle speed a bunch for this as the diameter I’m cutting went from 3.350” down to nearly 1”.







When I was getting close I became concerned the remaining wall tjickness in the bottom of the groove was becoming too thin and measuring it showed this to be true.

The bottom of any sheave does not contact the belt so I had a bit of wiggle room, so wiggle I did. Having that indicator in play was a godsend at this point as I just zeroed it and literally wiggled the form tool back and forth until a gauge pin told me I was at the proper width. This left me with 2.5mm wall thickness at the bottom of this groove, plenty of meat left there.





Having the important parts done everything else was just cleanup and clearance cuts. Here I’m clearancing the back of the pulley so it will clear the motor mounting screws on the lathe . I purposely made the large grooves walls(?) thin on top so as to match how thin I would need to make the back for said clearance.

The shoulder left is the amount of stick out I need from the motor shaft bearing to align all the V grooves.









Having done that I now needed to clean up the thicker wall of the small sheave to match the rest. This is also needed to give the retaining nut a little better purchase in the motor shaft
threads.

This was done with the set screw only and taking light cuts.







And all of a sudden I had a finished part!






We’ll mostly finished. I decided to take a riffler file and clean up the edges of the set screw hole so as to make life easier on the belt. Probably not necessary, but eh, only took a minute.

 

[RaisedByWolves]

Got the pully home and took what Im hoping will be the final measurement. The collar I turned on the back was meant to be left full in order to bring it to its final length, and boy howdy was it ever full.

I need to remove .150 from the length to get the sheaves lined up correctly.






No big deal really, some of this was planned, some of it just..........appeared? I think in the initial planning stages I had this worked out to the pully needing to be 1.370" thick based on the measurements from the factory pully and the motor shaft shim that came with the motor and drive kit from custom crafter. Given that and the fact that we had 1.5"x4" stock on hand would bear this out.

Now for the good part, I came in with only .001 runout both radially and axially! And I think I can tune much of that out once everything is brought to size and finalized. I also took a (very) crude vibration assessment and was pleased with that result also.

My setup was less than stable using a surface gauge sat on the headstock to hold the indicator, but there was only a tiny bit more vibration with the pully, key and set screw than there was with just the motor. Pretty impressive considering this motor turns 4500rpm. The neat part is I can tune out the last bit of resonance Im seeing around 2500rpm by shortening or lengthening the grub screw.

I even went and learned how to stitch all of my dumb short clips together to make a decent video, however Scorsese I am not.

ETA: Updated video to show extreme vibration with factory pully.

 

[RaisedByWolves]

I made a hockey puck, or a paperweight is you prefer.

I was going to leave this thread be as I showed the necessary, but I made a lil boo boo and compensated for it and thought some might benefit by seeing that, then in the end the whole thing turned to poop.

I was attempting to make a second pully to replace the horrible spindle pully, so I started recutting the mandrel. If I had known I was going to make two pullies rather than just the one I would have started with this one as it would have saved me time hogging the same material twice to get the two different sized (15mm and 22mm) mandrals I needed.

So I sawed off the old small mandrel and cut a new one to 20mm, but I overcut by .001 and I wasn't happy with that. Rather than re cut all that material to get it right, I used a cheat code and knurled the shaft raising the surface, then cut this to the proper size. This was easy being I now knew what the wrong size was.


Just a fine knurl raised the surface .0025 for an increase of .005 in diameter.






Then I blued it so I could see when I touched off on the peaks of the knurl and knocked that down with the radius tool and it looked like this.


Exactly 20mm, same size as the spindle.




You'll notice that I couldn't knurl the entire length of the mandrel as the knurl tool we have is a bit bulky. That's no problem though as I only need to center the pully on this and the register lip the piece backs up on will do the rest.


So I did all the same steps I did on the first pully and this was easier this time as I already have things figured out.

Only I outsmarted myself.

Since these two pullies work in conjunction with a 3rd central pully I was sure the V grove spacing would be the same, I mean, why would you vary this from one part to the next and overcomplicate that center pully, and make things more complicated than they need to be, right?

Why would you do this? COMMUNISIM!

So my second perfect looking pully has the V grooves off something like .125 or 3mm or some ****. ****!




The lighting was bad today and the flash....well its hitting a shiny round surface, but you can still see how bad this came out. Everything seemed to look right, up until I cut the last V, but when that was done things looked a bit chunky.

When I got home I looked at the intermediate pully and damn if there isn't a different groove spacing from one set of grooves to the other.





You can see where the large pully has about a 4mm step to the next smaller one, then another 4mm step between that one and the third, them only something like 1mm between that and the smallest pully.

I can only surmise that the bakelite pullies were a purchased part, probably off the shelf, and Prazi had to make the intermediate pully to suit these. That or there was some machining limitation in making the intermediate pully.

Anyway, the third pully should be a breeze now that I have already made all the mistakes.