Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Self-lubricating thrust washers and 4500 rpm shaft shaft

  1. Jul 11, 2012 #1
    I hope you all will permit a question from a non-scientist seeking some practical guidance on a vintage table saw restoration.

    I have a 1930s era table saw whose 5/8" shaft spins at 4500 rpm inside a porous bronze bearing that is lubricated with SAE 20 oil.

    On the motor pulley side of the shaft, there is a shaft collar with set screw between the pulley and the bearing. This shaft collar prevents the shaft from making excessive lateral movements. But it seems to me to have drawbacks:

    1) if the face of the shaft collar is not touching the bronze bearing, there is opportunity for some lateral play in the shaft, and hence some play in the blade; or
    2) if the face of the shaft collar is touching the bronze bearing, there is considerable friction

    Therefore, I thought that a self-lubricating thrust-washer could be inserted between the shaft-collar and the bronze bearing. The shaft spins inside the thrust washer, if I understand correctly. Am I going down the wrong path? If this makes practical sense, is there one type of thrust washer that would be better suited than another for this application? E.g. oil lubricated bronze:


    or PTFE/oil impregnated:


    or bronze or graphite?


    Here's a crude sketch to orient you. And a picture.

    Thanks for sharing your engineering knowledge with a guy who studied ancient languages in college!

    Attached Files:

    Last edited: Jul 11, 2012
  2. jcsd
  3. Jul 11, 2012 #2


    User Avatar
    Gold Member

    Welcome to PF, Tr888.
    I'm the least educated member here, but I build a lot of stuff. To start with, I'll say that some of that old stuff is superior to what you can buy today. As for your immediate question, though... I'd chop a circle out of the lid of a margarine tub and use it as a spacer. That probably won't approach anything like the efficiency that you want, but at least it's free.
  4. Jul 12, 2012 #3
    I don't need free but inexpensive would be good. :-) I don't want to have to keep replacing this spacer. What you you suggest if you had a few dollars to spend?

    P.S. I agree that much of the old equipment was built to last for several generations rather than several years. Loss of enduring quality is an inevitable by-product of buying with borrowed money and easy credit rather than with saved-up money, as we often do today in our throwaway culture: if it falls apart, you buy another. If, back in the the day, you saved up for a year or two to buy a new fridge or washing machine, you expected it to last a lifetime.
  5. Jul 12, 2012 #4


    User Avatar
    Gold Member

    Well, you could replace your bronze bushing with a double tapered roller bearing set. That would eliminate any end play without face contact, but it would be expensive and complicated. Perhaps you could start a thread in Materials and Chemical to ask for a suitable plastic to use as a spacer; I'm really not qualified to suggest one. The only thing that comes to mind is Delrin, but it has some problems of its own.
    I think that you hit on the secret of old-time reliability with your mention of "generations". Way back when, a tradesman's descendants were expected to go into the family business. It was also before mass-production caught on in a big way. A cobbler or blacksmith could use the tools passed down by his great-grandfather.
    One of my friends was a saddle maker, who also produced authentic holsters for movie westerns. His pride and joy was a sewing machine that was built sometime around the late 1800's or early 1900's. That thing would stitch 2 or 3 slabs of 5mm thick cowhide together like it was punching through marshmallows. Try that with a chip-controlled unit from Walmart and all you'll get is a bunch of bent needles and a burned-out motor.
  6. Jul 12, 2012 #5
    You're thinking plastic but I'm wondering if metal-to-metal would be ok. Here's my thinking, but it's totally based on assumption since I have not even had this saw hooked up to a motor yet.

    The porous bronze bearing is automatically lubricated with SAE20 oil from two tiny reservoirs above it, one near each end. They're about the size of a pencil eraser. The manual reads:

    This tool is equipped with the new type bronze bearings which meter filtered oil to the moving parts. These bearings have invisible pores that become saturated with oil. The heat generated by the turning of the shaft causes the lubricant in the reservoir to automatically flow through these bronze bearings, thus lubricating the moving shaft. When the shaft is at rest, the oil is absorbed by the bearing.

    I assume the oil moves out to the face of the bearing too, because on the blade side of the shaft, there's a chamfered metal collar that seems as though it could rest against the face of the bearing. See picture. I have it pulled away from the bearing so it can be seen better.

    I'm wondering if something comparable to this chamfered metal collar should be used on the pulley side. Would the oil in the porous bronze bearing and a self-lubricating shaft washer be able to have face-to-face contact at 4500 rpm?

    I do not want the shaft to have 1/16" of lateral play, 1/32" on each side. But maybe that was an acceptable amount of slop for small table saws back in the 1930s, and I'm expecting narrower tolerances than the design was capable of?

    Attached Files:

  7. Jul 12, 2012 #6


    User Avatar
    Science Advisor
    Homework Helper

    Maybe you do want that much play, if what you are sawing doesn't move in exactly the direction you intended - unless you want to break lots of saw blades.

    Think about what is the width of the slot cut by the saw blade, before you try to nail down the tolerances too tight.

    You may find the belt drive tends to stabilize the shaft in a fixed axial position when it's running (assuming there is no axial float at the motor end of the belt), even if the tolerance in the bearing appears to be sloppy.
  8. Jul 12, 2012 #7
    The width of the slot cut by the blade, i.e. the 'kerf', is 0.125" normally for ripping wood lengthwise along the grain and 0.075" for a thin-kerf blade for making fine cuts across the grain.

    The goal is to have repeatable cuts -- e.g. you might want to make a number of identical pieces--as close to identical as you can. If the shaft can wander ~1/32" between cuts then repeatable accuracy might not be possible.

    Since I can move the shaft with only finger pressure, I'm expecting it to move.

    Would a needle roller thrust bearing placed between the shaft lock collar and the bronze shaft bearing work?

  9. Jul 12, 2012 #8


    User Avatar
    Science Advisor
    Homework Helper

    You would probably need to spring-load the shaft somehow to maintain an end load on the thrust bearing. Trying to keep it loaded with a set screw adjustment doesn't sound very good. You need to maintain a minimum thrust load on the bearing as well as staying below the maximum load, otherwise the needles or balls can start to skid rather than rotate, leading to wear and/or overheating.

    The fact that you can move the shaft with finger pressure when it is not rotating may not be the same situation as when it is being driven under load, with the belt tension providing a restoring force if pulleys try to move out of line. (But take care if you try to check this out when the machine is running, of course!!)
  10. Jul 12, 2012 #9


    User Avatar
    Gold Member

    You really don't know how your table saw will perform until you have it operating. when running whatever axial tolerance you think you have will go out of whack as the parts begin to heat up, and you could end up binding the collars against the bearing. Try some test runs before commiting to how you will tweak your machine.

    A certain wobble on the blade might have been preferable way back then in the 1930's. Compare what blade you can buy now to the hardened teeth variety of years ago. As the teeth wore the cut had more of a tendancy to drift to one side bind the blade, and with the little bit of shaft play the sharpening of the teeth could be less frequent.
  11. Jul 13, 2012 #10
    Thanks for that caveat.

    Would these Belleville bearing springs or finger disc springs keep the roller bearing sandwich pressed together?

  12. Jul 16, 2012 #11
    FWIW, here is a picture from a similar saw showing the original shaft collar between the pulley and the sleeve bearing.

    Attached Files:

  13. Jul 16, 2012 #12

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    If it were mine

    i'd clean up both the collar and the pulley that's on other end
    and set them for about 0.010 side play in the shaft.
    Thin bronze washers sound like a good idea.

    Here's my non mechanical's take on it -
    Since the shaft is free to float left-right,
    any force against the thrust bearing you propose to put on blade end of shaft would have to be balanced by a similar bearing on pulley end .

    I think that's what mechanical engineers do in machinery design. But i am not formally educated in that subject, just have observed how Ford built the engines and transmissions i tinkered with.
    The flat thrust bearings i've encountered always have a few mils clearance.

    surely 1/100 inch is good enough for woodworking?

    I'm glad you posted this, for i have a similar Ward's saw agout same vintage. I'll be restoring it this winter. If i have same issue, might experiment with leather for thrust washers.

    edit hmmmmmm I just felt my table saw, modern Craftsman, and can't feel any play. Are you sure there's not a ball bearing inside that fat housing on pulley end, see picture in immediately prior post ?

    old jim
    Last edited: Jul 16, 2012
  14. Jul 16, 2012 #13
    Hey, Jim.

    Ignore that picture of the shaft collar with the set screws. I hadn't seen that saw in person. Turns out, it is a ball-bearing model, as you had surmised, which was offered in the same catalog year. The model I have has sleeve bearings only.

    Here's a picture of the quasi arbor bearing that is mounted blade-side. You can see where it is worn where it has made contact with the face of the blade-side sleeve bearing.

    My thinking is that even if I leave clearance between the shaft collar and the sleeve bearing, there will eventually be contact with the face of the sleeve bearing; so the root of my question is whether it would be better to have that contact be against something with inherent lubricity.

    Attached Files:

  15. Jul 16, 2012 #14


    User Avatar
    Science Advisor
    Homework Helper

    That makes sense. You don't really need a thrust bearing here, because if you are sawing straight, there isn't any signifcant thrist force. The thrust bearings in your links can handle end loads of up to thosands of pounds which is complete overkill.

    What you want is a location bearing, to stop the shaft moving in either direction. A ball bearing is a simple way to do that. (Ball bearings can also take thrust loads if the ball races are cone shaped rather than cylindrical, but that's irrelevant for your saw).
  16. Jul 16, 2012 #15

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    Last edited by a moderator: May 6, 2017
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook