PDA

View Full Version : Thrust bearing material?



DICKEYBIRD
09-18-2010, 06:27 PM
I' m piddling a bit with the little Denford CNC'd Sherline Micromill again today and was going to install the helical stepper motor coupler that Evan sent me. (Thanks Evan!:))

I took the Y-axis apart to measure the little roller thrust bearing so I could get another one on the way to repl. the missing one on the X-axis.

Man, I just can't believe how Denford designed this thing. It has a 3/8" dia. raised section on the leadscrew that runs directly against the back side of the support bracket disc. No bearing, no hardened thrust washer...nothing, it runs metal to metal and doesn't even have a way of getting lube on it properly without taking it apart! Also, the bracket is mounted to the end of the table with a single 8-32 screw. It has to be flexing under load. It needs a total re-design but I'm not going to do that right now. I'm going to do the best I can to reduce wear & friction and see how it works.

The leadscrew runs so close to the bottom of the table there's no room for another thrust bearing so I guess one thing I can do is add a thin, case-hardened washer (if I can find a 1/4 x 3/8 x .040" size) and drill a hole through the table to get some oil on it.

Another thought I had was to bore a small pocket in the bracket and install some sort of plastic bushing but the surface area is so small....062 sq. in. if my math is correct. I have some acetal rod in shipment; reckon it would hold up or is there a more durable plastic?

Last thought is an oilite bronze bushing but wouldn't that wear quickly?

Anybody have any thoughts?

http://i57.photobucket.com/albums/g227/DBAviation/LeadscrewBrg.jpg

http://i57.photobucket.com/albums/g227/DBAviation/LeadscrewEnd.jpg

http://i57.photobucket.com/albums/g227/DBAviation/LeadscrewDisc.jpg

RobbieKnobbie
09-18-2010, 06:58 PM
If the thrust load is being taken at the other end of the shaft, say by a set of angular contact bearings, then you would want that end to float. You never (normally) want to constrain both ends of a shaft.

Of course, if the other end of the shaft isn't properly constrained then all bets are off.

Toolguy
09-18-2010, 07:09 PM
I would counterbore the back of the disk 1/8" deep a little bigger than the boss on the leadscrew and make a Delrin bushing to fit into the counterbore.
If you can't go that deep, make it shallower. Oilite is pretty soft. It wouldn't last very long. Also, Aluminum Bronze would work in place of the Delrin. Making a lube hole is a good idea.

DICKEYBIRD
09-18-2010, 07:18 PM
Thanks T/G. I like that method too. I have some acetal coming from Enco so I'll probably give it a try.

I've been looking around a bit and came across Dupont Vespel SP-21. That stuff looks great. A 15% graphite filled polyamide. Haven't found a source yet for small amounts though.:(

Dr Stan
09-18-2010, 08:47 PM
I've been looking around a bit and came across Dupont Vespel SP-21. That stuff looks great. A 15% graphite filled polyamide. Haven't found a source yet for small amounts though.:(

I'll have to keep this material in mind for future projects.

BTW, I've purchased Teflon sheets from Travers Tool in fairly small quantities at a reasonable price. Would that work?

You may want to look at: http://www.travers.com/skulist.asp?RequestData=CA_Search&navPath=All+Products%2F%2F%2F%2FUserSearch1%3Dtefl on&q=block+id+72744+and+class+level3+id+28118&minPrice=$4.79

lazlo
09-18-2010, 08:52 PM
I've been looking around a bit and came across Dupont Vespel SP-21. That stuff looks great. A 15% graphite filled polyamide. Haven't found a source yet for small amounts though.:(

Vespel's neat stuff. I have a couple of drops that are Teflon and graphite filled. There's a MolyD version too.

Check the MSC catalog -- they have various high-tech polymer bearings (Vespel, Rulon, ...) in single quantities:

http://www1.mscdirect.com/CGI/NNSRIT?PMPXNO=1870507&PMT4NO=93955984

Evan
09-18-2010, 09:09 PM
I will send you an envelope with some PTFE sheet you can use. I'll include some thicker material too that can be machined to whatever thickness you need. I think I already have some washers of PTFE made up that will fit that particular spot.

Evan
09-18-2010, 09:37 PM
Never mind. :D

Here is a quicker answer. Pick up some peel and stick furniture foot pads with PTFE soles. I have some on hand so I tried splitting the PTFE from the hard foam pad and it worked just fine. The result is (gasp) .040" thick. I punched a .25 hole in it so you can judge the size. If you want I can send you some pre made 1/4" holes. ;)

http://ixian.ca/pics7/ptfepad.jpg

DICKEYBIRD
09-18-2010, 11:18 PM
Evan, is PTFE is hard enough to resist compression in a washer that small? I've never used the stuff as a thrust washer or bushing so don't have any experience with it but for some reason thought it was somewhat soft. No? If you think it'll work OK I'm all over it!

Robert, thanks for the link to the Vespel bushings. That stuff must have a bit of Unobtanium in it though. MSC wants 40 bucks for ONE 1/4 x 3/8 x 3/8 bushing. Ouch! I looked around on ebay a bit and man, that stuff's outta my league for sure!

Evan
09-18-2010, 11:28 PM
The stuff on those furniture glides is harder than regular PTFE. I am sure it will hold up just fine. It must be fortified with something. Even if not PTFE can take about 1000 to 2000 psi without cold flow at low surface velocities. On a screw that size the surface velocity will be negligible.

DICKEYBIRD
09-19-2010, 10:24 AM
OK, after a bit of Google-ing this morning, I found these bearings. I never knew there were so many choices in a bearing so small. Page after page after page...same size, different materials, seals, flanges...yada, yada ad infinitum!:eek:

http://www.bocabearings.com/search.aspx?SearchType=StartDimension&radix=inch&Outer=0.3750&InnerOperator=%3c%3d&OuterOperator=%3d&WidthOperator=%3d

The ones I'm looking at are R168-ZZ which are radial bearings and not thrust bearings but shouldn't they do OK in this application? I guess I could order a few and set one up in my screw press to see how much force it takes to pop one apart. I think the Vexta PH266 E1.2 stepper motors only have around 80-90 oz/in holding power so the forces can't be very high.

Toolguy
09-19-2010, 03:23 PM
Someone asked if Teflon would work. Teflon is very soft and would be a poor choice for this application.

Peter.
09-19-2010, 03:51 PM
Where did that thrust bearing come from? Surely it should be between the end-support and the boss on the screw?

Evan
09-19-2010, 04:28 PM
Someone asked if Teflon would work. Teflon is very soft and would be a poor choice for this application.

I disagree based on my practical experience. Virgin PTFE will withstand around 2000 psi without cold flow and more if it is trapped between hard surfaces. I use PTFE for thrust bearing applications a lot and it beats rolling element bearings because it doesn't need lubrication. In the application above if the thrust plate is recessed and a PTFE washer is inset it will last forever because the PTFE is trapped and supported on all sides. The only requirement is that the contact surfaces be smooth so they don't grind it up.

Toolguy
09-19-2010, 04:41 PM
I was thinking in terms of a thin washer. Even a small screw can exert quite a bit of pressure. I agree that encased on all sides it would be ok.

DICKEYBIRD
09-19-2010, 04:55 PM
Where did that thrust bearing come from? Surely it should be between the end-support and the boss on the screw?If you're referring to the roller thrust bearing shown in the 1st picture in this thread, it's located between the face of the helical coupler and the end support, the disc next to it to the left in the picture. The boss on the screw runs directly against the bare metal on the other side of the support, hence my concern.

Unfortunately, the screw is located so close to the table, there's only room for a .375" bearing or thrust washer.

Anyone have any experience with the tiny bearings I linked to? By the way, the R-168's are on page 12 of the link. I realize they're designed for high speeds rather than load but could they work?

Peter.
09-19-2010, 05:32 PM
If you're referring to the roller thrust bearing shown in the 1st picture in this thread, it's located between the face of the helical coupler and the end support, the disc next to it to the left in the picture. The boss on the screw runs directly against the bare metal on the other side of the support, hence my concern.

Unfortunately, the screw is located so close to the table, there's only room for a .375" bearing or thrust washer.

Anyone have any experience with the tiny bearings I linked to? By the way, the R-168's are on page 12 of the link. I realize they're designed for high speeds rather than load but could they work?

I suppose there must be a similar thrust bearing in a similar position the other end? If that is the case, the boss on the screw should never come into contact with that end-cap, unless there was a problem at the other end of the table.

DICKEYBIRD
09-19-2010, 06:00 PM
I suppose there must be a similar thrust bearing in a similar position the other end?Nope, nothing at the other end at all. Not a very clever design. Nice, eh?

Peter.
09-20-2010, 01:33 PM
No, I guess not. Seems a strange way of going about things.

whitis
09-20-2010, 02:33 PM
That is the original sherline manual mill thrust bushing setup. It works a little better than you might expect but is not a good design. Originally, the hand wheel has to be pressed tight against one side of the bushing to minimize backlash. Sherline tosses that part for the CNC kits which have real bearings. Don't lose that socket head cap screw, it is probably an obscure #5. The two additional screws, and the holes to support them, are add on.

The Fixer
09-20-2010, 03:21 PM
I'm not very knowledgeable about the various plastic/nylon materials but I have found the sliders from snowmobile track rails works for a whole lot of different applications I've used it in.
al

DICKEYBIRD
09-20-2010, 04:04 PM
I'm not very knowledgeable about the various plastic/nylon materials but I have found the sliders from snowmobile track rails works for a whole lot of different applications I've used it in.
alCool! I'll run down to the snowmobile shop and see if they can spare a few scraps.

Oh, wait....I'm in Memphis, TN. We ain't got no steenkin' snowmobile shops in Memphis.;)

Seriously though, I wonder what type of plastic they use. It must be some tough stuff.

I've got a few of the little midget bearings on the way. I'm hoping they will stand up to forces involved.

Evan
09-20-2010, 04:12 PM
I looked up the specs on the ball bearing. I don't think it will hold up in a thrust application. The rated static load is 18 kilos in the radial direction. For deep groove bearings the usual axial load factor is no more than half the remaining unused radial load factor. What is worse is that the static load isn't the limiting case for ball bearings even though they don't give the real limiting load since it is hard to define. The worst case load for any ball bearing is small low rpm repeated angular displacements where the bearing doesn't make full rotations.

Unfortunately that describes CNC operations perfectly. So, that bearing will have a limit value static load capability of maybe 20 lbs in the thrust direction but that will be downgraded by the radial load and much further by the repeating small angle nature of the axial load.

Short story is that it won't last long.

I would have replied to this question sooner but I couldn't access the site earlier.

whitis
09-20-2010, 04:53 PM
I would counterbore the back of the disk 1/8" deep a little bigger than the boss on the leadscrew and make a Delrin bushing to fit into the counterbore.
If you can't go that deep, make it shallower. Oilite is pretty soft. It wouldn't last very long. Also, Aluminum Bronze would work in place of the Delrin. Making a lube hole is a good idea.

You would need bushings on both sides and the piece is only 0.182 thick so your two counterbores would go clean through the part.

If you add much thickness, the set screw (on the original hand wheels, at least) will no longer bear on the turned down portion of the shaft and will mess up the part which is supposed to be smooth for mating and sliding the wheels on/off. A coupling will have similar problems, though it may be in the form of falling off the 1/4" portion of the shaft. For the new coupling you need to adjust so the coupling grabs the part that isn't turned down if it is compression type or the set screw bears where it is supposed to if set screw type.

If you aren't going to buy the sherline CNC upgrade mounts or build you own mount with bearings right now (where you may run out of shaft length), just leave it be and make sure you snug it up good (but don't overdo it, either) when reassembling. They have been selling machines for 30 years with that bushing design.

You could pick up flexible (HDPE) cutting board sheets at a dollar store and cut a pair of washers out of it; that would probably be thin enough. Put one on the lead screw side and one on the motor side of the bushing.

If using a compression type coupling, make sure the clamping portion rides on sufficient material on both sides of the goove. With the coupling you got from Evan, that might be hard to achieve.

DICKEYBIRD
09-20-2010, 04:55 PM
Ok, thanks Evan. I saw the "18" rating under the specs section but didn't have the smarts to work out what that works out to in real world usage.

When you think about it, the balls must be very small when the O.D. of the bearing is .375" and the I.D. is .250"! Ain't much room left when you have and inner & outer race to cram in there too. It's amazing that bearings can be made so cheap & so small. Those are not the smallest ones either...by far.:eek:

DICKEYBIRD
09-20-2010, 05:09 PM
You would need bushings on both sides and the piece is only 0.182 thick so your two counterbores would go clean through the part.

If you add much thickness, the set screw (on the original hand wheels, at least) will no longer bear on the turned down portion of the shaft and will mess up the part which is supposed to be smooth for mating and sliding the wheels on/off. A coupling will have similar problems, though it may be in the form of falling off the 1/4" portion of the shaft. For the new coupling you need to adjust so the coupling grabs the part that isn't turned down if it is compression type or the set screw bears where it is supposed to if set screw type.

If you aren't going to buy the sherline CNC upgrade mounts or build you own mount with bearings right now (where you may run out of shaft length), just leave it be and make sure you snug it up good (but don't overdo it, either) when reassembling. They have been selling machines for 30 years with that bushing design.

You could pick up flexible (HDPE) cutting board sheets at a dollar store and cut a pair of washers out of it; that would probably be thin enough. Put one on the lead screw side and one on the motor side of the bushing.

If using a compression type coupling, make sure the clamping portion rides on sufficient material on both sides of the goove. With the coupling you got from Evan, that might be hard to achieve.Thanks, you are very perceptive sir!:) Sounds like you have a lot of experience with the little critter.

Using the little 3/8" bearing, I had planned to turn a bit off of the boss on the leadscrew and counterbore the support piece but have to admit I didn't investigate it thoroughly enough. You are very right, there's precious little space to work with.

The support piece and the boss materials are both somewhat hard and have polished themselves where they contact each other. If I don't come up with anything else, I might just rig some type of drip feed oiler and see how that goes. There is a roller bearing on the other side.

DICKEYBIRD
09-23-2010, 11:00 PM
OK, I have some crude testing results and seek some math help from yuse engineer types.

My black acetal rod came in from Enco so I rigged up a quick & dirty test. I chucked a piece in the lathe, faced it and drilled a 1/4" hole. I chucked up the leadscrew in the tailstock, inserted it into the hole with the 3/8" dia. thrust face against the acetal, spun up the lathe at 450 rpm and applied about 10 lbs force for several minutes to get some break-in time. I did this dry just to see what would happen. No smoke or anything but it did feel pretty warm to the touch. I'm guessing over 110 deg F.

I applied some way oil and kept going. Maybe a little less heat was generated. I then put on some *moly grease and voila, the heating dropped drastically.

I decided to lower the rpm and add some pressure to simulate real world conditions on the mighty Micromill.:rolleyes: I ran the lathe at ***220 rpm (Math check please...20 TPI leadscrew @ 220 rpm is 11 IPM...yes?) and applied 50 or 60 lbs. pressure (as far as I can tell) for 4 or 5 minutes. I was impressed that there was no wear other than the factory lathe turning marks on the leadscrew thrust face being impressed into the acetal with little heat buildup.

This is where I need some Jethro Bodine cypherin' and goesintas. 20 TPI lead screw, about **80 oz/in holding torque stepper motor, 3/8" OD-1/4" ID thrust surface and 60 lbs pressure applied with no wear. I think it would stand more pressure but SWMBO called me to supper and now I'm full & lazy. Does the 20 TPI screw equate to a ratio of 1:20 so an 80 oz/in stepper theoretically can generate 133 lbs of force? I know steppers of course don't produce near the same torque at speed as they do at rest but this is just a rough guesstimate.

Let's say the maximum force was indeed 133 lbs; I guess I would need to do the test again with twice as much pressure applied to see if the acetal thrust washers can survive in a worse case scenario?

Oh yeah, I saw on a spec. chart that copolymer acetal has a Rockwell hardness of 80 to 117. Personally I think I'm good to go with a pair of .050" acetal washers, broken-in on the lathe, lubed with moly grease and let 'em rip for a while, checking occasionally for signs of wear. That tiny little area was absorbing what felt like a lot of pressure without blinking.

What say ye of properly trained noggins?


*Moly should really help acetal leadscrew performance.

**Vectra PH66-E1.2, 1.2Amp, 5 ohm, 1.8 deg/step motor. Only found 1 mention on the web and it said about 80 oz/in

***Probably a bit faster than the top speed I reckon this little critter will do under a usable load.

whitis
09-23-2010, 11:49 PM
Machining forces on the sherline will probably be around 50lbs (above that the spindle will poop out, maybe a little higher if you have the DC motor), plus a bit more to overcome friction.

80oz-in is 5lbs at 1 inch or 40lbs at the 1/8" shaft radius. Multiplied by the 20TPI, that gives 800lbs but most of that is lost in friction at the nut and also some in the bushing.

At 1 inch per second, the leadscrew will turn at 1200RPM and at 2 inch per second 2400RPM. If you have good microstepping drives with a supply voltage of 24 to 36V, you will probably get somewhere in that range with ball bearings for the thrust bearings.

DICKEYBIRD
09-24-2010, 09:34 AM
Machining forces on the sherline will probably be around 50lbs (above that the spindle will poop out, maybe a little higher if you have the DC motor), plus a bit more to overcome friction.

80oz-in is 5lbs at 1 inch or 40lbs at the 1/8" shaft radius. Multiplied by the 20TPI, that gives 800lbs but most of that is lost in friction at the nut and also some in the bushing.

At 1 inch per second, the leadscrew will turn at 1200RPM and at 2 inch per second 2400RPM. If you have good microstepping drives with a supply voltage of 24 to 36V, you will probably get somewhere in that range with ball bearings for the thrust bearings.Ouch, 800 lbs?:eek: Is that right?

I don't think I'll be needing 1200 rpm even if it were possible with this meager little machine. I'm looking at 10 - 15 inches per minute