PDA

View Full Version : I had an idea for spindle bearings this morning. Convert a turbocharger shaft assy.



trilect
12-15-2004, 08:16 AM
PLease note that I have no idea if this is practical or not and I can't do this myself yet (no metal machining equip yet).

those needing high speed and continuous use spindles a turbo charger shaft assembly and bearings might be an idea.

machine the turbo charger case down to minimal material, retool the TC shaft and provide a 10 to 30 psi oil line with return to keep the bearings oiled and cooled. I think these bearings are good for 60,000 plus rpm.

eman5oh
12-15-2004, 08:23 AM
I dont't think the shafts would support any type of side load and these shafts are are quite small. Most of the ones I have seen are around 1/8 to 1/4 inch in diameter. I could be wrong but something to consider.

trilect
12-15-2004, 08:57 AM
I'll do more investigation in the spring, I've been trying to get the wife to let me buy a smithy 1220xp or ltd combination machine and finally succeeded.

HuFlungDung
12-15-2004, 10:26 AM
I'm not positive, but I think that turbos are built with a very loose fit on the bearings. The bearings serve to keep the thing from dragging on startup. Usually, pressurized oil is fed into the turbo shaft bearings and the shaft actually runs on a hydrodynamic film of oil inside the bearing race.

I wonder if any enterprising hobbyists have ever tried to make their own hydrodynamic spindle? The actual construction of the spindle is very slim and very simple (just uses lead bushings), but capturing and containing all the pressurized oil that keeps it afloat would be a bit of a challenge. The rpm obtainable is virtually unlimited.

eman5oh
12-15-2004, 11:33 AM
Turbos bearings when good should have less than .002 play but are designed with clearance that is for the oil. Thinking about this the thurst load would destroy the bearings in a hurry as turbos are designed to minamize the thrust load.

NeoMoses
12-16-2004, 12:31 PM
I wonder if any enterprising hobbyists have ever tried to make their own hydrodynamic spindle? The actual construction of the spindle is very slim and very simple (just uses lead bushings), but capturing and containing all the pressurized oil that keeps it afloat would be a bit of a challenge. The rpm obtainable is virtually unlimited.

Very interesting thought, Hu. Keeping the oil contained isn't all that difficult, really, just look at rear main seals on an engine. The new ones are PTFE seals last for millions of miles in today's diesels.

Actually, an engine block might be a good place to start to make something like this. You've got everything necessary for the bearings, and it's all nicely aligned. A nice aluminum block 4-cylinder would probably be the easiest to start with. Junkyard price should be less than $500, and all you're really interested in is the lower end and oil pump. The head can be destroyed and it will still be fine for this.

Using the timing chain/belt to drive the crank would be the most appropriate, and the pulleys should all be with the engine. A belt-type transmission would likely be necessary to achieve various speeds, or a DC electric motor could be used.

The RPM limit for a setup like this will be around 7000 RPM (depending on the engine used) because of oil pump cavitation.

Boring the crankshaft to accept collets might be a bit of a challenge, but should be possible at any good machine shop.
I've got to get back to work, but I'll keep pondering this. Very interesting subject, indeed!

HuFlungDung
12-16-2004, 12:59 PM
Umm....okay, NeoMoses. But it ain't gonna be pretty :D

Maybe you could just use the head from an overhead cam engine, and build a new spindle instead of using the camshaft. (You'd also have to build a box to put it in, to contain the oil and reroute the oil galleries.) Camshafts are usually cast iron and difficult to work with anyways.

You'd want to reduce the running clearance maybe to a thousandth of an inch and increase the pump pressure to 1000 psi, to make a really stiff bearing. That would require a small external oil pump unit. High pressure oil is hazardous to be around, so you'd need to use professionally made hoses and fittings and protect against pinhole leakage (high pressure oil will pierce the skin and cause blood poisoning).

ViperTX
12-16-2004, 01:25 PM
Engine bearings rely on a lot of metal, oil and water to keep them cool. Also the loads that an engine is designed for are mostly axially.

Anyway ABEC7 angular contact bearings of 15 degrees are about $138 for a duplex set...so a spindle would need almost $300 of bearings....hmmmmm.

NeoMoses
12-16-2004, 08:00 PM
Umm....okay, NeoMoses. But it ain't gonna be pretty :D

Maybe you could just use the head from an overhead cam engine, and build a new spindle instead of using the camshaft. (You'd also have to build a box to put it in, to contain the oil and reroute the oil galleries.) Camshafts are usually cast iron and difficult to work with anyways.

You'd want to reduce the running clearance maybe to a thousandth of an inch and increase the pump pressure to 1000 psi, to make a really stiff bearing. That would require a small external oil pump unit. High pressure oil is hazardous to be around, so you'd need to use professionally made hoses and fittings and protect against pinhole leakage (high pressure oil will pierce the skin and cause blood poisoning).

I've done some more thinking, and some research, and found one flaw in my plan: if the oil pump is tied to spindle speed (as it is in an engine), the pump itself will be consuming 2-5 HP. In this regard, the pump should be decoupled from the spindle and run by it's own motor. 1000 psi will be overkill for the oil pressure; engines rarely run over 75 psi.

You're right Hu, an engine block would be pretty ugly in a machine tool. Let's look at this from the perspective of building a hydrodynamic bearing for the spindle from scratch.

If the bearing and spindle were both made from iron, 15-30 microns for radial clearance should work well. Oil flow through these bearings increases cubically with clearance, so much more than 80-100 microns will be flowing a huge amount of oil, requiring a very large oil pump. Also, aluminum expands much more than iron, so ideally both parts should be iron, with a thin bearing material inbetween. Perhaps the spindle could be made to use standard engine bearing inserts for cost, the same with the lip seal.

ViperTX, main bearings do see signifigant radial loading. Think about the firing loads happening when a smallblock is generating 400 ft-lbs. of torque! Also, there aren't coolant passages near main bearings, the oil does the cooling.

HuFlungDung
12-16-2004, 08:30 PM
I have my doubts that you could force any oil at all into a space narrower than .001" You do want some flow to ensure removal of debris and to cool the bearings.

A typical gear type oil pump (such as I am used to working on, being in a agricultural area) has no significant loss of pressurized oil until the clearance exceeds about .002". At that point, then the user will notice that the shaft seal starts to seep, and then he thinks it needs a new shaft seal. But, he's wrong: a lip seal pressed into the outside housing (from the outside of the pump) will certainly not be holding in 1500 psi :) The mechanical clearance of the shaft in the bushing is what keeps the high pressure oil inside the gear chamber.

In a hydrodynamic bearing, I think you would want to feed it high pressure oil, because you want to ensure that the spindle stays centered. In an engine, there is no real requirement that this happen, and the relatively low pressure oil is only for lube and heat removal.

I suppose that you could figure out the bearing surface area, and deduce some kind of typical deflection force that the spindle is liable to experience when cutting, and figure backwards from that, what psi you'd need. I'm guessing that it would be well above 75 psi, but that's just nothing more than a guess.

The bearing clearance determines the rate of throw-off of the oil, so, in good condition, I don't think the volume requirement for high pressure oil would be extreme.

NeoMoses
12-16-2004, 10:03 PM
For those who don't work with this stuff everyday, here's a good starting point: http://www.machinedesign.com/ASP/strArticleID/55762/strSite/MDSite/viewSelectedArticle.asp

I happen to be a lubrication development engineer, so much of my knowledge will likely be engine related. I'm not saying I'm an expert (yet), but I do work with this daily. If you can guess what car company I work for, you'll win a :cheers: (Hint: My favorite engine just celebrated it's 50th birthday) Anyway, 25.4 microns = 0.001", so we're talking about the same clearances, Hu.

After doing some re-thinking, I've realized that aluminum isn't all that bad of a choice for the hobbyist. These machines won't be seeing nearly as much thermal cycling as an engine (hopefully!). Aluminum eliminates the whole need for a bearing insert, so it would be much easier to whip one up in a weekend (cheaper, too!). And, if the sealing land is sufficiently long, the lip seal could be left out and there would only be a small amount of oil that weeps by. It probably won't even be noticeable in a machine shop, especially if there's a flood or mist coolant system.

trilect
12-17-2004, 08:05 AM
You could use a modified power steering pump off an older car. Feed the bearings dextron 3 transmission fluid.

I think that most ps pumps can be modified to produce alot of pressure if you needed it too and it can be coupled to an external motor.

HuFlungDung
12-17-2004, 10:30 AM
There ya go, Trilect. Make it so :D

ViperTX
12-17-2004, 11:06 AM
NeoMoses,
25.4 microns isn't that 25.4 x 10 -6 which is like 0.0000254....oops micron is metric......hmmmmmm....... 1 microinch = 0.0254 microns....okay 25.4 microns is equal to 0.001 inches.

Fifty years...hmmmmmm....the fuel injected engine that appeared in the early vette???? Let's see the V8 from Ford is much older.....

sbrpollock
12-17-2004, 06:17 PM
I just read that chevy's first V-8 was introduced in 1917, but only lasted for a few years.

At least they never built a flathead.

Neo.....My guess is the small block chevy. Didn't it come out around '54 as a 265?

How many Small block chevy's have been produced since it came out?

ESjaavik
12-18-2004, 05:36 PM
This is starting to get interesting! Now who will be the first to make a spindle powered by a V8 big block! ;) Why bother converting it? Would that be what they call High Speed Machining?

HuFlungDung
12-18-2004, 05:46 PM
LoL, yes that would be a crowd pleaser, wouldn't it. :D 5000hp @ 10,000 rpm! What couldn't you cut in less than a second :D

NeoMoses
12-20-2004, 01:37 PM
You're right, sbrpollock. I'm a GM guy, and a big fan of the smallblock: http://www.circletrack.com/thehistoryof/58038/index.html

Anyway, back on subject. First question: why use Dexron rather than engine oil? We're looking for lubrication, right? A good hydrodynamic film, low friction bearing? For those reasons, I'm leaning towards engine oil.