MAKING LINEAR NEEDLE BRG. CAGES

this is vertical column of my Mitsui surface grinder it has hardened d2 rails that the cast iron head rides on with it just an oil film scrape surface and I'm going to change that to a needle bearing situation as you can see here this has oil galleys to feed the system and this is in pretty bad shape but this is going to be reasonably easy chuckle chuckle to convert this to a needle roller bearing situation where we're going to be able to preload this and have a much greater stiffness and fine increment down feed here's a view of the completed cages I'm going to show next how where I got the rollers from the individual rollers and then also the design details and aspects that I considered in making these cages what does one do when you need six or 700 needles for a linear needle bearing conversion to your surface grinder vertical feed you buy some appropriate needle gauge needle bearings and rip the cage out and salvage the bearings the reason that we're using caged versus non caged is the non cage to have little tits on the ends that are meant to be caught by the lips of the drawn cup and they're not very conducive to our standard machine cage that I'm going to be making for the linear linear cages that's why I'm doing these you'd actually get more needles per bearing by using the full complement needle bearing as opposed to the caged but like I said you'd end up with the pointy ended rollers which won't won't work as nicely in the case it would work but they just wouldn't work as nicely buying rollers individually is not as easy as you would think even though you would think hey what should be like dime-a-dozen since they don't even have to put in the cage or or in the in the bearing but turns out it's easier said than done unless you're buying you know monumental quantities so back to just buying a bunch of bearings and tearing the cages out here have modelled the column and head of the Mitsui very accurately and this way I can actually work in SolidWorks and determine what I'm going to do here and now I'm looking down the the vertical column and these represent these blocks right here represent the hardened d2 rails and what I've done here is these pockets you see are where I'm going to put hardened steel inserts in they're going to be glued in place and I thought about how I was going to do those by a to get a heat-treated all that and I realized uh by some high speed steel planer blades from grizzly tool relatively inexpensive reasonably high you know high quality steel and that way I can just glue those in and grind them so that provides the hardened surface on this side for the roller and the d2 rail right here represents the other hardened insert whenever you have rolling element bearings you have to have above at least like 58 Rockwell to have something that's going to work well so that enables us to do this this right here insert here gives us the lateral control and then these two will be pre-loaded on against each other and this gap you see right here is will have a fitted spacer that will control exactly the preload of the races against the rollers so this gets sandwiched and small needles are chosen because number one everything I do here takes up space that thins this the material in the head so I don't want to weaken this any more than necessary you also get more damping by having multiple small rollers and you also get Morel a stick averaging of minimizing the roundness effects of the rollers because you have so many of them engaged in one time they tend to average out the rumble of the out of roundness of the rollers when we're talking we're talking micro inches down here that the levels that were we're speaking about so another aspect of this is that the cage in this case I'm making a decision to make this cage just about exactly 2,000 smaller than the roller I have to leave some room because the roller actually gets compressed when it's pre-loaded but what I'm shooting for here is these cages being phenolic would act as an actual bearing service in themselves but I also want to get some squeeze film damping you get a very thin gap you've got a viscous oil or material in the gap and when that gap tries to fluctuate the flow of oil required in the resistance that flow causes a damping effect so all these things are such the things that contribute to this being stiffer but having a less stick slip as far as making very very small incremental down feed moves and yet be stiffer probably than the oil film version because you need some oil film clearance so there is some oil squirm effects so all these are reasons why I said wow this thing's just screaming to have needle bearing rollers there's other reasons why I'm doing this and that is because I'm planning on all kinds of crazy things being mounted to the head on a kind of like a pallet system that uses a kinematic Keystone kinematic mount top and bottom that will allow things like J grinder spindles and all these things like that so the the stability of this head laterally is much more important that it would be for just grinding where in theory surface grinding wise it wouldn't matter if this thing moves side to side as long as the the orientation the spindle was maintained so all those are reasons why we did that now the design of the cage itself I'm showing the assembly here one of the things is when I talked about the oil the cage being able to be just a few thousand smaller than the roller well the cage then these two cage pieces this upper and lower need to be able to float to accommodate that like there's no way I could machine this and have this all work out to where that you know worked out perfectly so by putting a dovetail feature on this cage where these latch into each other it lets this these two side cages the front and back cage actually move to whatever accommodates the position yet holds them positionally both axially and laterally relative to the other cage so that they track properly one of the aspects of a cage is that especially with rollers one when you have them pre-loaded they're going to want to track in a particular orientation wherever they're aimed they're going to want to stay that way so there can be great forces if these slots weren't exactly perpendicular to the travel you could have things where the rollers would actually break the cage from trying to run off and track in the direction they want to go so squareness of the roller axes is very critical to this working and not breaking itself up here's the cage assembly by itself and you can see how these latch into each other and one of the details here is you notice how I made this to be machinable with a sixteenth end mill so that in the same setting rum milling these grooves I'm able to come in here and do this shape and the other edge detail the male piece here will male on this side is also be able to be done with an involution ability and process as I'm designing this so you could design this with a thing here and say oh I've got to come in here with a little dovetail cut or got a grind little dovetail cutter to make this but these reliefs here have no effect on the functionality of this cage all this is doing is holding the cage in position axially laterally but allowing it to float in the other axis so these notches here to be able to do this with a nib and male are irrelevant now we're looking at the actual needle pocket structure and you'll notice that I have these staggered back and forth the reason for that is I have a particular roller length which was chosen based on the most economical needle roller that gave me the most needles for the money but had it with close to what I was after but because of this width of the of the hardened insert that I had I wanted to use more area of it and so I staggered these back and forth wouldn't actually be necessary but basically in theory this spreads the fatigue load out over more of the width of the full insert so can be helpful here the pocket you can see I've got a relief here and what that does is when the roller sits in here this keeps the roller from being able to fall out where it's not covered by the ways so this cage actually hangs out in the air on either end a linear cage moves half the distance that the one of the sliding elements does so there's that there's an exact length of cage that will stay engaged for a full length of travel so that means these are going to be hanging out in space but this will make sure that the roller is held captive and can't fall out just because it's exposed and then these sides right here at each end or what give it the precise guidance but we don't want it to be have more rubbing area than necessary to keep it square because believe it or not the torque required to roll move these rollers to rotate them you don't want it so snug that the rollers are hard to turn because it will end up actually causing quite a bit of drag force in having to spin all those rollers when they're when they're pre-loaded they won't they won't skate I originally had these dovetails up here at the top sharp edge and I was going to grind a cutter and then in thinking about trying to make these they could be completely machined in one operation on my clamping system with the same end mill 1/16 NM I realized just put these pockets in and that's why I mentioned before that I did lubrication wise these channels from one end here to the other end here the other end here basically forms a situation where the oil has to flow through all of them to get through you can't have an oil pocket or an air pocket where oil can just seep out and not get get to where it's going it has to travel the full length of the cage and remember these faces of the each side of this are up against the hardened way surface with only about a thousandth of an inch gap so these this is pretty much going to a kind of a controlled cavity that will allow this oil to flow and stay in there I'm going to be using a vector number four which is relatively thick lay and clings very well so this will get prefilled with that and then the oil galley system that you saw on the beginning there with the oil cups will just maintain a supply that is feeds into the into this pocket area and as these roll by the LEAs get filled as they they go past but shouldn't be your whole lot of flow because of the very very small gap that we have here so here we show the head and the column and the cage the cage is not moving like it would in reality if I move this an inch the cage would move half an inch two inches this would move one inch so that it's it's in mesh if you if you will based on that and here's where I'm showing how those little lips that are on this side keep the rollers captive they can't fall out in this direction because everywhere they're actually held against that central rail and the rail is inside the cage for the full extent of travel no matter how far we go so that's that's how that works and obviously I'm only showing the cage on one side for technical purposes but it's on the other side also and the lateral these inner cages the inner width cage is adjusted by moving one of these rails there's actually screws to move the rails to preload gently on each side to preload it laterally also so that's the general design ideas that were involved in designing the cage the spacing of the rollers was something relatively arbitrary having a decent number of rollers but not so close that the cage strength could be compromised and not so far apart that the number of rollers bearing the load would be too small it's probably overkill in my case here but I'm after real high stiffness with relatively light preload the aspects of doing the modifications to the castings and going the inserts in and will be covered in another video this is strictly relative to the cage design and making the cages of this video I'm sawing the linen phenolic material here with a Harbor Freight tile saw with a regular tile saw blade you can see that this cuts pretty nicely and leaves a decent finish and we're cutting the strip's roughly before we surface grind them here I've got an aluminum vacuum fixture with three steel feet included on the back so that I could be able to hold them on the chuck when I flip it over using the spring hold downs to do that at the moment throw in a vacuum entry hole on the end of the fixture the block in the surface ground with the fill carbide wheel and done and I have my 70 durometer ring cut to length and it will so you stick our piece in please off well line up visually over the area and my son suffering the vacuum pump along and that quick goes right up to full vacuum see the gauge there and the Blair it's definitely don't have too much of that vacuum up there and it's it's on there so it's definitely down fast silicon carbide wheel is a good deal for grinding plastic Christi plastic like this phenolic and hopefully I'll be able to park off that whatever 45,000 so you have to take off just in general what I do is going to this has some fabric texture on this side which I don't want so I'm going to just skin this side to clean up you get a nice brown finish on this side then I will flip it and take the rest off we get down to the air putting the ring back in putting a piece down again and I'm just taking a plunge cut on this full wheel width as I go across these and you'll notice all the color and the coolant there that's all of the phenolic material that's coming off and here I'm just getting the fur off of the or the fabric texture off of the phenolic the as molded fabric texture kind of shows through that glossy surface and I want to get rid of that that glossy surface is also not the best for lubrication wise mm-hmm you can see there where it's ground down flipped over now and now we're going to take substantial cuts where I'm removing the I'm actually creep feeding it almost full depth just leaving enough for a single finish pass once I get the most material removed because this cuts very easily with a silicon carbide wheel and here I'm doing my my pass where I'm just taking a traverse pass across there for the finish cut on that now when you see me peel this off you'll see how deep that cut was going from 1/8 inch down there like it was 74 and almost 1 cut flipping around and friend to do the other side you can see all the concentration of that phenolic dust in the coolant there it'll clog a filter in a heartbeat so beware if you grind this stuff but it grinds very nicely finish pass again [Music] now I'm doing the edges I've got a clamp to a long bar so I could maintain the squareness all four pieces are or six pieces are actually clamped together and now I just slide my whole block down on the chuck to maintain my position and Heights and everything and just regrab the bar and I'm able to do that one long length which is longer than the grinder more than accurate enough for what we're doing here so I'm creep feeding the majority of the material off and then coming down doing a you know finish pass on on these now that that one side is done I flip them over into the other and here I'm squaring the ends they have it in the vise crossways and I'm actually just feeding through full depth of cut all one cut right through measuring the overall length there now I've got a sanding block there with some diamond film on it and I'm just using that to deburr get any burrs off of there and get it ready for the rest of the milling operations you can see the Leonard phenolic in there cleaning off my fixture plate that I use with my clamps that you've seen in my previous ball ball cage video cleaning out where this is going to sit in the bottom of the vise flipping this over letting that sit down tight I have my jack screws backed off a little bit so they're not touching the top of the vices here and I tighten each vise up independently as I'm pressing down and then once I've done that I will tap down he's in the hardened a tube block they used for tapping finished parts works very well then I'll come in and actually put an indicator on here at the at rest position that's zeroed there and then I adjust that's a set screw on top working against the screw at the bottom and I'm actually intentionally putting a thousandth preload on there I'm bowing the thing up got a mirror finish on my fly cutter there diamond lapped and that piece of ABS is double sticky taped down and I'm using that as my flattened surface now I'm doing the actual dovetail edge treatment with a bow flute a sixteenth of flute cutter and here we're doing the pocket mill along with the oil grooves you saw and this is the same procedure where we have these clamps that can be moved at will I pause the program just slide these along till the next pocket just moving our way down through and that's just a actual subroutine that's moved down incrementally on the whole part move the clamps again carry on down through and this works very effectively for this style of part you can also see why I was trying to design with a single cutter since I don't have a tool change or a single cutter and they able to do the entire part contour in one shot with one cutter so that's makes life a lot easier and keeps everything registered to itself well just using dry air here to blow the chips away and I have my vacuum system it's always on the mill there that's something away all the dust do a lot of cast iron work so that's a must-have for getting rid of the dust there so it turns out that the position difference of the roller notches between this one with this style edge and this one that has the other mating part of the dovetail is a sixteenth of an inch from nine thirty Seconds to seven thirty Seconds so I'm just going to use rollers to space this off from the same setup so I don't even have to move the program I'm putting my eighth inch diameter pins in just happened to be burrs in this case carbide verse and then I am taking the strip in place but I'm putting my 1/16 dowels between here and the pin to give me my spacing and this phase is this other part in with the exact same program lines it up so I can just run don't have to have a different program or you know shift the whole program by a certain amount I can just do it directly so I've got that in the pocket there put my clamp on press firmly and I just work my way down the down line here pressing these tight against the pins I'm tight against my end stop pin here and clamp so on and run the program just like I did with the other ones and the edge dovetail on these will get machines and a different set up here I'm using a diamond sanding disc which is basically nickel plated diamond the diamond is plated down to those little spots with nickel so there's no free abrasives loose I'm using that for deburring getting rid of the fuzz on the edges of the all the cuts I have two pieces of tool steel together that I'm milling the edges on just to square them up to use as gel liners for doing thin work like this here I'm surface grinding them individually I'm doing it half at a time diagonally on the on the Chuck so that I can actually grind the whole length of these and then you'll see there where I've spun it done the other side and blended in just seeing how it matched up now I'm using that to do the edge dovetail treatment on the one cage that supports the other two cages and you can see I'm just going around using an end mill to circle around do the actual dovetails on there obviously sped up flipping it doing the other side so you think bozo doesn't visit these parts huh well very first move of the very first part just after getting done grinding these parallels nice and parallel I noticed this deal going deeper than I'd like and I'm supposed to be programming ninety thousand steep 0.09 and on the very first move I programmed point nine miss zero so nice raspberry there on my parallels so I'll have to make sure I use them inverted all the time let's see that so I have two bars that I ground to be basically linear JA facings a full-length look this I just super glued a piece of ten thousand spray steel on here and this is just held by the friction of the of the jaw in the back because I'm using the wavy parallels here to preload my other parallel in the center and I'll show you that when I expose so I'm just peeling that off that was just to control the the end position so that once I edge found this the position would be good I was just trusting it's not critical the the the actual axial position of this is not critical so I wasn't really worried about not moving and then we're gonna back up the jaws here and show you what was going on inside so this jaw is sitting on parallel or on the del pin to bring it up to the top it's only an inch and half high roughly and so that's what they were sitting there for then I had the actual parallel that I was using in the back here and then I was using the wavy parallels to preload all this together so this was actually pretty stiff because this was squeezed down almost completely flat and the other doll sitting on the down pins also so that was the that was the setup on how I did that I had these all three vise locations I had this set so when I push down I wasn't pushing down the part over air or out there was a parallel in heard everywhere and then I just grabbed it nice and solid and like I said the friction of this wavy parallel compressing this very tightly just frictional II held this parallel in place so that my X location stayed good for running all these on the right has been deburred on the left as before being deburred you can see all the fuzzy edges from the linen material in this phenolic and that all has to be removed so I'm using diamond plated mandrels and things to abrade this and the advantage there is that there's no free abrasives to get stuck into the phenolic you don't want to charge this thing with the material that could end up chewing up your rollers over time and so that that's a bad thing so using various shapes you'll see here in the next shot I'm doing the backsides of these just relatively low speed just letting the thing ride actually in the groove to chamfer the edges using the sanding or the diamond sanding block again to knock off the the burrs and get those things nice and flat those sanding things work real great I guess you can see there it's 800 this is 800 mesh which is a really nice sizes it removes material reasonably well but doesn't leave coarse finish now I'm using a little steam cone to get in here and to get these these transitions where there's that fuzz from the breakout there on the edges and going in and manually doing all these yeah to do this to every single spot on all the sides but yeah just necessary and using the flexible shaft which is nice and controllable and quiet compared to an air tool doing the same thing just riding around the part to bear the edges of the dovetail milled section the nice little small diameter cutter and the small diameter in the body work well here I'm just using it manually to go in and just drag and basically sand the edges of the little lip that's in there that captivates the roller from being able to slip through this is a very very fine abrasive nylon brush and this is basically just removing the hair off of anything that's left getting rid of that and in theory there could be some free braces but this is super super fine and it really does a nice job of just getting that last little bit of fuzz off of there so that these have a decent finish at the head from a lot of different angles you know over and over but works works really well to clean these up because the the hairy nature of phenolic is is people have who've worked with it know exactly what I'm talking about it's it's one of those difficult aspects of it is getting a nice smooth finish when you've got a lot of operations because the fibers of the of the fabric that it was made from you know don't cut as easily as the crispy resin that it's been molded in I've washed these with a strong degreaser scrub them with a toothbrush basically the whole idea being to get rid of any loose particles abrasive particles even though I used all bonded Nicolle bonded abrasives on this shouldn't be any free of races I did use a abrasive nylon bristle brush to divert some of the edges so in theory there could be some minor contaminate in there so that's the main reason for doing a really severe thorough scrub of these and rinsed to get rid of all that so I have my PVC heat sealable material here tubular and I'm putting these in to keep them clean until it's assembly time and we're I'm going to be putting these together and putting them on the on the machine so these are going to get all inserted into the bag I'm just gonna grab them all get them in move up here to my heat sealer that's hanging on the wall and I do take it down if I need to but right now he's gonna eat seal this have you found that interesting or informative or both and like I said I'll be covering the other aspects of the needle bearing conversion meaning the hardened ways and lapping the deep two rails to make sure they're good in parallel and the preload system and a subsequent video on the rest of the Mitsui rebuild there'll be quite a few items on that so hopefully I'll get to them a little quicker than I have been

Channel: ROBRENZ

Views: 53,896

Keywords: surface grinding, surface grinder, needle bearings, vacuum chucking, phenolic, preloaded

Id: fsqylnTC64Y

Length: 31min 52sec (1912 seconds)

Published: Mon Jan 21 2019