Machining a bearing spacer: how hard can it be?

[@prbmax]

Your soon going to realize that time goes fast.

[@nbrworks]

I already am! The best I can do is be very present and not wander around too much 🫠

[@humanperson543]

The fastest time ever goes wen working on a project r something u actually care about like working on a car a 30 min job just some how turns into a 5 hour expedition

[@Andrew_Fernie]

I laughed hard when I saw the video title. I have also been down this road. Plenty of suggestions to use a surface grinder but when you don't have one... Nice result.

[@dazaspc]

Indeed a decent one is needed as well. I used to do repairs on Twin Spindle milling machines and it was very important that the tools sat at the same height in either spindle. The spacers once machined were almost impossible to hold in a lathe short of a magnetic chuck and I rarely had access to these. I would just hone each face with a stone by hand and could easily achieve sub 2 micron levels of parrel. it just took 1/2 an hour for each shim.

[@paradiselost9946]

the people suggesting surface grinders have no experience in metrology or mass production and tolerances. that surface grinder is no better than its own accuracy. any wheel bearing kit ive dealt with that uses spacers also includes a set of SHIM WASHERS. far easier to punch out millions of spacers within tolerances of a nominal size, then stamp out shims in various thicknesses. then let the end user deal with the (expensive!) selective assembly.

[@buttcrack7784]

@@paradiselost9946ALL machine tools are only as good as its own accuracy. That’s why any good shop keeps their equipment in good condition. In my 40+ years as a design engineer and working in and with machine shops finish grinding the thickness is the best way to control not only parallelism but also thickness. If you want accuracy in any dimension grinding (lapping, honing, polishing) is the way to go.

[@broheim23]

That is the best explanation of bearing preload I've heard! You've earned yourself a new subscriber.

[@nbrworks]

Welcome!

[@artgoat]

I guarantee those spacers are tighter tolerance than any OEM car manufacturer's spacers. When I was speccing out tapered roller bearings for a trailer, SKF allow for the spindle to be undersized by 0.05mm, meaning the bearings can be offset axially from each other by 0.1mm! (I needed to research this because the spindles I ordered were defective-undersized by more than .2mm)

[@Bobr1997]

.2 is crazy toleraze 💀💀

[@claeswikberg8958]

Surface grinder nicely avoided, though if he needed a matched set of 4 that would be a pain without the surface grinder i guess great video

[@paradiselost9946]

matched set of four is far easier than making ONE. lap them between plates, "nominal dimension", and then punch out some shim washers.

[@FamTech.]

@@paradiselost9946what?

[@Dogfather66227]

I appreciate that you work within your constraints to achieve a [very] acceptable outcome. You use what you have, as most of us would. Nice video as usual.

[@nbrworks]

Thanks, nice to have you here again! 😃

[@darrenconway8117]

So to hit a tight tolerance, I do it with constant depth of cut steps. I do not creep up on the dimension. I dial in a cut depth, then measure the work piece after the cut to measure the difference between dialed depth and actual cut depth. For the next cut, I dial in what I need to achieve the actual depth, rinse and repeat. I can hit tolerances 25x greater than the depth of cut.

[@Ojee2]

That is a beautiful ending my man.

[@nbrworks]

Thanks. Changes everything 😃

[@VitriolicVermillion]

I don't know anything about any of this stuff, but it was hypnotic, and I learned a bit about bearings!

[@nbrworks]

Cool! Welcome!

[@VitriolicVermillion]

@@nbrworks Thanks! Keep up the good work

[@paulkupperman7049]

Congratulations!

[@m.abdullah.y]

i rarely write a comment but this is a great quality video. You walked us through the whole process giving all the detail nicely and presenting everything visually that we can see. I was confused about what bearing preload is but the way you explained it made it crystal clear.

[@nbrworks]

Thanks. Welcome!

[@Rustinox]

Chasing microns on a lathe is indeed a real challenge. But as we can see at the end of the video, you have a helping hand :)

[@nbrworks]

Hey Michel. Indeed! 😃

[@jbrownson]

Nice explanation of preload, I've heard that term but hadn't really visualized what was going on

[@peterfitzpatrick7032]

I've done this where I needed 20 spacers... I ended up making an expanding mandrel to do the same, both sides machined without removing... The mandrel was just a piece of 1" 12L14 tapped 3/8 BSPT and slotted using a hacksaw without taking it out of the chuck... 😎👍☘️🍻

[@nbrworks]

That's also a good way to do it. Cheers!

[@TalRohan]

Surprisingly difficult but you got it in tolerance .........However little digits at the end, theyre just perfect.

[@thigtsquare950]

Congratulations on the B.

[@nbrworks]

Thanks, I'm really happy 🫠

[@Soggstermainia]

My first thought is they grind these for a reason, it's just the easiest way to do it. Second thought was hold the stock out face and do the exterior diameter, then past the length if the space put a groove big enough to place your micrometers in, then bring the face back to match he length needed then bore it after. The problem is the inner bore breaks the part free so the inner diameter would be harder to get exact. Just watched to the end, lol. Essentially very similar to what you ended up doing.

[@jimsvideos7201]

You get a like for the double entrendre in the title alone.

[@Gumbatron01]

65 Rockwell maybe? --- I'll see myself out.

[@ls2011schorsch]

nice youtube algorithm , i liked that video a lot !!!

[@Strothy2]

OD/ID is not critical, and the length is, my head: let's get grindin'!

[@aaronhammond7297]

I thought for a second you were making a steel mandrel... why so long?... oh, no. thats the spacer.. "Interesting" approach. Still, I learned something, so thanks.

[@FrancSchiphorst]

Small warning relating to the end. Whenever you get a grinder, they are for objects only ;) And yeah, changes perspective. Oldest goes to uni this year. Time for small fingers is long gone so enjoy!

[@Orakwan]

Even simple parts are difficult. Everything is difficult. But if it was me having to do those spacers, with the tooling I have at work (a nice and rigid turret with 12 stations and good tools and inserts that someone else has to pay for...) I would make the outside, inside, then part off with a perfectly trued parting blade with a brand new insert, right feed and speed and flood coolant, leaving a nice parting face finish. When parting, the OD and ID will distort because it's a violent process, sometimes by more than 0.01mm, and sometimes get tapered or elliptical shaped... But since OD and ID are not precise, what matters is the length and parallelism. The first part will not be good right yet. But it will perfectly parallel. Now, I just need to correct the parting tool length offset to get the length I need, and run two more parts without changing anything. They will be identical and perfectly parallel. In the worst case, some face taper might occur on the parted side, but will still be parallel (just not flat, if it makes sense) but this shouldn't be more than .01mm and it only happens with dull parting inserts or when the parting starts on a cone. In any case it will remain parallel and work as a spacer. Might even chanfer all 4 angles before parting, leaving only a burr in the inner parted side, which can be removed easily. If it's still not good enough, then give me geometry tolerances and surface roughness so that I understand what I'm working with

[@akselbering291]

16:45 THIS! This is how you exactly how much content creators should include their children in social media and no more, while still making for an excellent announcement.

[@quelthalas21]

making life so much easier turning parallel sides: Soft jaws Turn them out, with the same clamping force as you're gonna hold the part with Using soft jaws, i rarely have any issues with parallelism bigger than 0,01mm..and when i do, the error is almost always mine (am a machinist, and the approach is exactly the same for a manual or cnc lathe

[@robmorgan1214]

Use coolant and slower feeds after you get the rough dimensions.

[@chronokoks]

Sorry but how did you mount the spacer on the delrin mandrel? Superglue? Friction? Am I missing here something?

[@nbrworks]

Press fit. It was a bit on the tight side.

[@romualdaskuzborskis]

*Makes precision spacers.. slaps on cheap ball bearings for a cartwheel*.. :D Nice content. Sub'ed.

[@ramonching7772]

Good explanation. But it is wrong to say the length is critical. The hub nut is the one that handle the preload. Actually if you are going to use a spacer, you are just taking chances that all pairs of ball bearing have the same clearances.

[@evanlacava9213]

Use a part off tool next time and do it all in one op. Face, turn, drill / bore and partoff. Slow feed on parting. Then grind for final length if needed, or lap flat

[@Orakwan]

What exactly is that machine at 16:02 ? Does it work on all shapes and sizes? We currently dont have any professional-looking means of engraving parts numbers and its bad

[@nbrworks]

Hi. That's a diode laser. I have a dedicated video about it, in case you want to see it. It's not the right machine for metal engraving, but it's powerful enough to do it in steel and stainless. With an etching spray it can even mark brass and aluminum. If you're interested in a better machine to engrave metals, watch out for my upcoming videos 😉

[@Lukey-ts2xv]

Do you know how flat your stand though? If you're aiming for micron precision i highly recommend getting one of those fancy cast iron based Mitutoyo comparator stands with the serrated anvil, those things are awesome, and apparently have tighter tolerances than their granite based ones. Don't ask how much they cost though, i only use them at TAFE (trade school in AUS), but they look and feel like i cant afford one lol.

[@Gnomebitten]

Well done, of course its a job that a surface grinder would make easier, but who has the space in the shop?

[@chrisstephens6673]

I have no space, which is why mine sits in the front garden under a cover.🙂

[@paradiselost9946]

you take a note from the book they use in the bearing manufacturing plant... lapping between parallel plates that can be made from scrap, using rudimentary tools... surface grinding is only as good as the machine itself.

[@aceroadholder2185]

I have a Sanford bench top grinder on a two-foot square table.

[@chrisstephens6673]

@@paradiselost9946 even a worn out one should be able to hold parallel to a micron or two on a ring of that size, after the mag chuck has been reground!

[@nbrworks]

@@aceroadholder2185 I have looked for one of those in the past, but where to find one?

[@ОлександрСавченко-т6й]

Hello! Is the animation in the video made in Inventor?

[@nbrworks]

No, it was in fusion 360.

[@ja-no6fx]

thats a lot of decimal places lol

[@malbirrell]

You could maybe make some that are threaded and adjustable for length

[@paradiselost9946]

or make them undersize to a standard nominal, then make a punch and die, and stamp some washers out from shim stock? which is more often than not the approach that all manufacturers take... far easier to make identical parts within tolerancing limits, then select a shim or three to get the correct preload, than deal with a dozen different length spacers and use selective assembly... threads undo. threads are not a locating feature... threads can be wonky in regards to the end faces...

[@kleini3]

I would just turn everything in one setup and part it off to length

[@c4t4l4n4]

Maybe a tool post mounted grinder might be an affordable solution. Just need to be able to tram it parallel to the surface of the part and make sure you cover the ways nicely.

[@bobweiram6321]

Muito bom!

[@Joe-bm4wx]

I’m gonna be honest, for an automotive application, I think 5 hundredths of a mm is more than good enough.

[@nbrworks]

I thought the same at the begining, but after measuring the 2 wheel bearings in the video (tapered rollers), they're both within 0.01mm/~0.0005''... so I thought, at least would be nice to match that.

[@imyaaniggua]

can you grind them on the magentic table ? i would do it like that i wouldnt bother about making jigs

[@nbrworks]

I don't own a surface grinder.

[@imyaaniggua]

​@@nbrworksdifferent workshops demand different solutions

[@ramonching7772]

Length error of 0.010 or 10 microns is not for lathe work. It is already in the realm of grinding.

[@aaronbrown1344]

You can get a spacer that size with parallelism within 0.0001". You can put the finger on your indicator in the bore and clock the backface of the spacer in the chuck. If that's not possible, you can clock the front face where you parted it off and that will normally be very good. You can also turn a reference diameter that has a shoulder before parting that you can use to clock it parallel when you turn it around.

[@barryshrives]

Hi little B come on in❤

[@klazzera]

I hate bearing spacers. They seem like a novice engineering design fault. I truly see no reason why we shouldn't use spring washers for preloading. Please someone knowledgeable tell me.

[@leonordin3052]

I am just guessing here. Maybe it is when id is press fitted. (Which party spindles are because thermal expansion inconsistensies are more on od) You need more pressure to push the bearings to remove the axial play. The bearing spacer will make sure you dont overtighten. Otherwise I am thinking if the id is not press fitted then the shaft could move axially in the bearings. In this case the shaft is bolted between the id of the bearings so it is firmly in place. Also using bearing spacers will provide precise control over the position. That is you can make sure, yes I have now preloaded this bearing by x amount of microns, or with x amount of microns of axial play.

[@kyriacosvasiliou8342]

How about parting the spacer off?

[@K0balts]

exactly. No moving of part needed, all operations can be done in a single setup. Except chamfering / deburring the chuck side inner edge.

[@JaakkoF]

Harder to keep te tolerance and to get a straight parallel cut and maintain surface finish.

[@oddspaghetti4287]

- turn OD - face with a parting blade - drill hole with a nominal drill resulting in a clearance hole - part off at final dimension with the parting blade you used to face and know the width of accurately - deburr with hand tools and a file There, could have saved you so much time. Remounting the work piece just so you can get a shiny surface is one of the most common reasons I see people make bad out of spec parts.

[@gyrogearloose1345]

Not going to cut it my friend!

[@Beagle36]

Brother at this point in precision, just invest in a surface grinder. I can understand not wanting to use it though.

[@nbrworks]

I'd love to find a small bench top one, but I don't even know where to start looking (in Europe)!

[@Beagle36]

@@nbrworks ah ok. Yeah that makes sense. Hell, if you really wanted one you could probably source the materials and DIY it yourself

[@mslucass]

Question: at 11:03 you can see lines inside the freshly turned surface. Is this an artifact from the lead screw?

[@nbrworks]

It's from the nose radii of the tool (small) plus fast feed.

[@neffk]

consider lapping to get the final few microns

[@paradiselost9946]

nearly there! you make the spacers to a nominal size. 12mm. you lap them all together, in the same set of lapping plates, easily made, easily rigged up on a drill press. you then stamp out a few shim washer with an easily made punch and die. you then let the end user assemble as required. excess shims are then tossed in a drawer and forgotten about...

[@nbrworks]

After doing the first 3, which were not so good, I considered this option. But then I also wanted to try the method shown in the video - just to see if it would work. To some level it did (I understand exact copies would be tricky)... and I thought it was nice enough to share. I don't own a surface grinder, otherwise I would have used it. I guess that if the required spacer turns out to be in the middle of the 0.05mm, the person fitting the bearings can always use one of the shorter spacers and order some shims. But I think he will be covered this time.

[@martindietrich2011]

In my shop this is a task for the surface grinder .

[@pierre-alexislachance9718]

Why not grind them? Much easier to get to them all to the target dimension all at once with more precision and have a very good parallelism.

[@nbrworks]

Because I don't have a grinder?

[@Swedishchef11]

7:00 or a grinder with magnetic table.

[@ibrahimdeniz7308]

Im sorry but what is this? "Im surprised by the precision", its almost like you put round bar in a lathe and machined it.

[@GraphicManInnovations]

that is why surface grinders are made, your method is the pure definition of the "hard way", not only time consuming but a lot of material waste as well with really no good reason, also think about reducing the length if needed, you cant simply put it back to the lathe, that will defeat the initial purpose of making it in one setup, a surface grinder will gladly take the part over and over again

[@wtchr6883]

I have put dozens of bearing spacers back on my lathe. Big deal it's not. They are one of the simplest parts you could make.

[@paradiselost9946]

@@wtchr6883 i can go one better... ill make you a dozen all within a micron. with a drill press. a cheap nasty drill press... i will also make them all undersize. then include a stack of shim washers.

[@wtchr6883]

@@paradiselost9946 bearing spacers are not critical dimension parts. That's my point.

[@paradiselost9946]

@@wtchr6883 you obviously dont work with many bearings, or just like replacing them a lot...

[@wtchr6883]

@@paradiselost9946 you obviously know not what you speak of. I made bearing spacers for years for offshore cranes and winches. Never had a single bearing fail because of a spacer.

[@lUnderdogl]

Buy cheap car parts.

[@mylittleworkshop7219]

Nope