PDA

View Full Version : Need help with air spindle



IdahoJim
01-04-2011, 11:09 AM
Guys, I'm building a Tinker tool grinder, and decided to try and make an air spindle for it. I have the spindle and the bore so the spindle is a nice sliding fit, with zero play...as far as I can measure it. When I put the air to it, it gets more difficult to turn.....more pressure, more difficult. Tried it 2 lbs up to 90lbs. At 90 it's close to being locked. It turns easiest with no air...LOL.
Any thoughts that might help me?
Thanks,
Jim

philbur
01-04-2011, 11:19 AM
Posting a design drawing might be useful.

Phil


Guys, I'm building a Tinker tool grinder, and decided to try and make an air spindle for it. I have the spindle and the bore so the spindle is a nice sliding fit, with zero play...as far as I can measure it. When I put the air to it, it gets more difficult to turn.....more pressure, more difficult. Tried it 2 lbs up to 90lbs. At 90 it's close to being locked. It turns easiest with no air...LOL.
Any thoughts that might help me?
Thanks,
Jim

moe1942
01-04-2011, 11:23 AM
You have to have a slight amount of air escape or no bearing affect..You got it too tight..

macona
01-04-2011, 11:29 AM
Air is not getting around the spindle. Instead of creating an air bearing it is pushing it against the other side clamping it in place.

Air bearings need to be honed to fit. A turned shaft is probably not going to do the job.

IdahoJim
01-04-2011, 11:48 AM
You have to have a slight amount of air escape or no bearing affect..You got it too tight..

I have air coming out the ends, around the spindle. I had a a thought that the problem was uneven air pressure, as macona suggested, but I'm not sure how to proceed from here. I'm afraid I may screw up the fit by trying to hone it, but may have to bite the bullet and give it a try. I'm a long way from being a machinist, but understand I'll have to try, or I'll never learn anything.
I'm using a spindle from one of those Chinese grinding jigs, and it measures pretty decent for parallel.
I've got an axial air groove in the bore, that intercepts the air input, and radial grooves about 1/2" in from each end that intercept the axial groove.
Thanks for the input...I'll keep you posted.
Jim

IdahoJim
01-04-2011, 11:52 AM
Posting a design drawing might be useful.

Phil

Don't have a drawing to post, Phil. The bore is .982", length of bore is 3.125".
Some of my plan was from Glen Wilson's article in Projects in Metal II, some from other's posts on various machining sites.
Jim

lazlo
01-04-2011, 12:03 PM
I'm afraid I may screw up the fit by trying to hone it, but may have to bite the bullet and give it a try.

Glenn Wilson has a great article on building an air spindle in Metalworking 2. The general idea is that you have to maintain a small uniform air gap along the length of the spindle.

Glenn's spindle has a 0.0015" air gap. The bad news is that Glenn specs the taper tolerance required for the air bearing to float at +.0002" over the length of the spindle: 4".

Toolguy
01-04-2011, 12:13 PM
On the air bearing on my Darex T&C grinder the air inlet is not direct to the bore. The bore is a sleeve with 2 sets of cutouts about 1/4 of the way in from each end. These cutouts are 120 degrees apart, looks like they were cut into the side with a woodruff key cutter. The 2 sets of 3 cutouts are indexed rotationally to match. There is a hole with a brass filter at the bottom of each cutout that lets air into the spindle. I would try another axial groove opposite the air inlet to equal things out.

IdahoJim
01-04-2011, 12:39 PM
On the air bearing on my Darex T&C grinder the air inlet is not direct to the bore. The bore is a sleeve with 2 sets of cutouts about 1/4 of the way in from each end. These cutouts are 120 degrees apart, looks like they were cut into the side with a woodruff key cutter. The 2 sets of 3 cutouts are indexed rotationally to match. There is a hole with a brass filter at the bottom of each cutout that lets air into the spindle. I would try another axial groove opposite the air inlet to equal things out.

Great idea! I was trying to decide between that method, or just honing the center slightly larger to facilitate the airflow around the spindle. I also thought about cutting an addition radial groove centered on the air input.
Right now, the shop is 22* so I'm sort of stooging around waiting for some sunshine. My shop is solar-heated, and gains 35-40* above outside temps...but only when the sun shines. Winter's a PITA
Jim

IdahoJim
01-04-2011, 12:50 PM
Glenn Wilson has a great article on building an air spindle in Metalworking 2. The general idea is that you have to maintain a small uniform air gap along the length of the spindle.

Glenn's spindle has a 0.0015" air gap. The bad news is that Glenn specs the taper tolerance required for the air bearing to float at +.0002" over the length of the spindle: 4".

Yup..I've got that article, Robert. I don't even have the means to measure to tenths.
The way I did this bore is in sections. Not having the confidence to try the full-length of the bore, I thought I might do it by doing the bore in sections....rings, if you will. There are 4 rings about 7/8" long. I bored and honed each to a nice fit on the spindle. I then bored the housing about .003 larger, assembled the rings on the oiled spindle, and then epoxied the whole works into the housing using JB Weld. I was hoping to find a method that anybody with ordinary machining equipment could use to make an air spindle. I think my problem is more of design, than fit, but we'll see. I'm fairly sure that macona, and others, are right and the trouble is even distribution of the air. Whether, or not, I can fix it remains to be seen.
Jim

HAP
01-04-2011, 01:02 PM
My Darex EM sharpener has several holes in the bore of the fixed sleeve. Also, you do not want any oil on any of the parts. It is best to clean both the cylinder and sleeve with alcohol. Makes a huge difference.

HAP

rohart
01-04-2011, 01:20 PM
I don't think your on a starter unless the air supply is balanced around the circumference of the spindle.

Remember how air speeds up over the wing of an aircraft, and the pressure drops, holding the plane up ?

Well, if your spindle moves over to one side, and the air has to speed up to get out of a narrower gap, its pressure will drop, and the spindle will get pushed further in that direction.

Your supply must be radially symmetric, and preferably three supplies at 120 degrees at least. An odd number of supplies is probably a good idea.

If I understand you to say you have three air-fed chambers that are effectively individual rings around the spindle, then you aren't providing radially symmetric control of the air as it exits the annular gap. Maybe some narrow axial channels, and a higher flow rate, may be a temporary fix.

lazlo
01-04-2011, 01:28 PM
Yup..I've got that article, Robert. I don't even have the means to measure to tenths.

I'm fairly sure that macona, and others, are right and the trouble is even distribution of the air.

We're saying the same thing: in order to provide even flow around the spindle, there has to be very little taper along the spindle length. It has to be very straight. Any deviation in gap width is going to change the localized air pressure, and the spindle won't float.

TGTool
01-04-2011, 01:37 PM
I believe you should think through the dynamics of what makes an air supported bearing. It should be like spotters around a trampoline so that if the spindle deviates from center, a force on that side pushes it back to the centered position and then the forces on all sides again balance. You can think of it either as a reduction in force at the large gap side, or an increase from the small gap. For those dynamic changes to take place there has to be a certain relationship between the air escaping around the shaft and the air being resupplied at a certain pressure and over a certain area. That is, it's a hydraulic problem with PSI x Area developing a force, and those forces have to change in response to a spindle offset. Simply blowing air around the spindle doesn't develop any balancing forces to force the spindle into any particular position.

Think of a series of holes around the periphery of the bushing. A force develops from the PSI x Area equation. Now, if the spindle is pushed to one side, you want a relative increase of force there. Since the feed air pressure is constant, this relative force is produced by the DECREASE in force at the opposite side where the annulus between shaft and bushing has increased creating a pressure drop. For the decrease to happen, the hole feeding air into the annulus must be small enough that it can't compensate for the greater area of air escape. In other words, the air being fed in is metered by the small hole so a differential force is created for recentering.

The other feature to include is the little divots Toolguy mentions in his Darex. This is to make sure that if the spindle is pressed all the way to one side covering the inlet hole, the feed there can still develop a centering force. If the small hole were blocked off completely, the PSI x Area equation says the actual force at that point becomes miniscule. The depression in the sleeve wall ensures that you maintain at least some nominal area for the force equation.

GadgetBuilder
01-04-2011, 02:20 PM
Building an air bearing spindle is more difficult than most people expect, at least according to what I've read in the Quorn group:
http://groups.yahoo.com/group/quorn_owners/ (http://groups.yahoo.com/group/quorn_owners/msearch_adv)

It is worth searching that group for comments by Frank Kurda who made a successful air bearing spindle. I believe he's German so his English is good but it clearly is difficult for him. He seems to be a talented machinist who builds from ideas rather than drawings so there is no published plan for his spindle AFAIK. However he does describe it and it uses a number of VERY small (0.3mm) holes which would likely be challenging to drill. Plus the spindle is very accurately lapped. He linked to this video (which was posted under a different name?):
http://www.youtube.com/watch?gl=DE&hl=de&v=I9YgG_soRcE

Also see Kurda here: http://bbs.homeshopmachinist.net/showthread.php?t=28528

John

IdahoJim
01-04-2011, 03:10 PM
I don't think your on a starter unless the air supply is balanced around the circumference of the spindle.

Remember how air speeds up over the wing of an aircraft, and the pressure drops, holding the plane up ?

Well, if your spindle moves over to one side, and the air has to speed up to get out of a narrower gap, its pressure will drop, and the spindle will get pushed further in that direction.

Your supply must be radially symmetric, and preferably three supplies at 120 degrees at least. An odd number of supplies is probably a good idea.

If I understand you to say you have three air-fed chambers that are effectively individual rings around the spindle, then you aren't providing radially symmetric control of the air as it exits the annular gap. Maybe some narrow axial channels, and a higher flow rate, may be a temporary fix.
Yup..I understand that. I'm trying to balance the air supply with the axial grooves feeding the radial grooves. That should supply air around the spindle. What I have are 1 axial groove, feeding 2 radial grooves. I just got done cutting an additional axial groove connecting the 2 radial grooves, but diametrically opposite to the original axial groove. That didn't change anything.
Jim

IdahoJim
01-04-2011, 03:11 PM
Building an air bearing spindle is more difficult than most people expect, at least according to what I've read in the Quorn group:
http://groups.yahoo.com/group/quorn_owners/ (http://groups.yahoo.com/group/quorn_owners/msearch_adv)

It is worth searching that group for comments by Frank Kurda who made a successful air bearing spindle. I believe he's German so his English is good but it clearly is difficult for him. He seems to be a talented machinist who builds from ideas rather than drawings so there is no published plan for his spindle AFAIK. However he does describe it and it uses a number of VERY small (0.3mm) holes which would likely be challenging to drill. Plus the spindle is very accurately lapped. He linked to this video (which was posted under a different name?):
http://www.youtube.com/watch?gl=DE&hl=de&v=I9YgG_soRcE

Also see Kurda here: http://bbs.homeshopmachinist.net/showthread.php?t=28528

John
LOL...if it wasn't difficult, I wouldn't be asking questions...dohhhhh
Jim

IdahoJim
01-04-2011, 03:16 PM
I believe you should think through the dynamics of what makes an air supported bearing. It should be like spotters around a trampoline so that if the spindle deviates from center, a force on that side pushes it back to the centered position and then the forces on all sides again balance. You can think of it either as a reduction in force at the large gap side, or an increase from the small gap. For those dynamic changes to take place there has to be a certain relationship between the air escaping around the shaft and the air being resupplied at a certain pressure and over a certain area. That is, it's a hydraulic problem with PSI x Area developing a force, and those forces have to change in response to a spindle offset. Simply blowing air around the spindle doesn't develop any balancing forces to force the spindle into any particular position.

Think of a series of holes around the periphery of the bushing. A force develops from the PSI x Area equation. Now, if the spindle is pushed to one side, you want a relative increase of force there. Since the feed air pressure is constant, this relative force is produced by the DECREASE in force at the opposite side where the annulus between shaft and bushing has increased creating a pressure drop. For the decrease to happen, the hole feeding air into the annulus must be small enough that it can't compensate for the greater area of air escape. In other words, the air being fed in is metered by the small hole so a differential force is created for recentering.

The other feature to include is the little divots Toolguy mentions in his Darex. This is to make sure that if the spindle is pressed all the way to one side covering the inlet hole, the feed there can still develop a centering force. If the small hole were blocked off completely, the PSI x Area equation says the actual force at that point becomes miniscule. The depression in the sleeve wall ensures that you maintain at least some nominal area for the force equation.

Yup...got that. I know how it's supposed to work....the faster air moves, the less force it has...so when the gap gets larger, the air speeds up, and the air on the opposed side pushes the spindle back until the velocity is balanced. Same as the old beach ball above the vacuum cleaner outlet.
The grooves I have in the bore prevent the air from being blocked off, no matter the position of the spindle.
Jim

John Stevenson
01-04-2011, 03:23 PM
There was a tool and cutter grinder at the Sandown show just recently that was running off it's own compressor at 90 pounds.
When it wasn't being used the spindle was rotating at about 25 -50 rpm, on it's own for hours.

http://www.stevenson-engineers.co.uk/files/T&C%20grinder.jpg

This had air split and going into both ends of the spindle.
Another interesting thing was the wheel was driven by one of those outrunner model aircraft motors where the motor spins.

IdahoJim
01-04-2011, 03:24 PM
I'm beginning to think the problem is not enough clearance between the spindle and bore. I did some checking and honing, and I can now get a .0015 feeler gauge to slide in between the spindle and bore about 1/8", or a bit more. So, I'm getting closer. I did cut an additional axial air groove on the opposite side from the air inlet. That didn't seem to do anything.
I appreciate all the advice. I'll keep working on it, and let ya know what happens. I've got some Flitz metal polish and thought I may try using that to fine tune the fit. It's very fine, and shouldn't go overboard on metal removal.
One thing I did notice, after the last round of honing, is the air pressure is having less effect, as far as making the spindle more difficult to turn...that tends to suggest the clearance, or lack thereof, may be the problem.
Jim

IdahoJim
01-04-2011, 03:27 PM
There was a tool and cutter grinder at the Sandown show just recently that was running off it's own compressor at 90 pounds.
When it wasn't being used the spindle was rotating at about 25 -50 rpm, on it's own for hours.

http://www.stevenson-engineers.co.uk/files/T&C%20grinder.jpg

This had air split and going into both ends of the spindle.
Another interesting thing was the wheel was driven by one of those outrunner model aircraft motors where the motor spins.

Damn, Sir John, you're making me drool all over the keyboard. Nice grinder!
Jim

moe1942
01-04-2011, 04:06 PM
I have copies of HSM going back to the mid 80's. I do recall a build of one . I was interested at the time but IIRC the tolerances made me change my mind. I can try to find those issues if others can't readily produce some plans..

philbur
01-04-2011, 04:35 PM
This process for assembly of the sleeves will not necessarily result in the degree of alignment required. Think about it.

Phil:)


Yup..I've got that article, Robert. I don't even have the means to measure to tenths.
The way I did this bore is in sections. Not having the confidence to try the full-length of the bore, I thought I might do it by doing the bore in sections....rings, if you will. There are 4 rings about 7/8" long. I bored and honed each to a nice fit on the spindle. I then bored the housing about .003 larger, assembled the rings on the oiled spindle, and then epoxied the whole works into the housing using JB Weld. I was hoping to find a method that anybody with ordinary machining equipment could use to make an air spindle. I think my problem is more of design, than fit, but we'll see. I'm fairly sure that macona, and others, are right and the trouble is even distribution of the air. Whether, or not, I can fix it remains to be seen.
Jim

philbur
01-04-2011, 04:47 PM
Yes but if the air supply is to great there is no loss of pressure in the annulus when it's width increases, so you do not generate an out of balance restoring force.

The inlet holes of 0.3mm mention in a previous post limit the air supply. There is an ideal ratio between the inlet orifices size and number and the annular gap and length between the spindle and housing.

Phil:)


Yup...got that. I know how it's supposed to work....the faster air moves, the less force it has...so when the gap gets larger, the air speeds up, and the air on the opposed side pushes the spindle back until the velocity is balanced. Same as the old beach ball above the vacuum cleaner outlet.
The grooves I have in the bore prevent the air from being blocked off, no matter the position of the spindle.
Jim

TGTool
01-04-2011, 05:06 PM
Yup...got that. I know how it's supposed to work....the faster air moves, the less force it has...so when the gap gets larger, the air speeds up, and the air on the opposed side pushes the spindle back until the velocity is balanced. Same as the old beach ball above the vacuum cleaner outlet.
The grooves I have in the bore prevent the air from being blocked off, no matter the position of the spindle.
Jim

No, you haven't got it. You're assuming it's a Bernoulli effect but it's not. Stop and listen to Phil and think about the air supply and escape relationship. It doesn't depend on fast moving air. It could be modeled on oil or some other fluid that moved relatively slowly.

oldtiffie
01-04-2011, 05:09 PM
Here are some pics of my air spindle/quill:

http://i200.photobucket.com/albums/aa294/oldtiffie/Air-quill/Air-quill1.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/Air-quill/Air-quill2.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/Air-quill/Air-quill3.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/Air-quill/Air-quill4.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/Air-quill/Air-quill5.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/Air-quill/Air-quill6.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/Air-quill/Air-quill7.jpg

It is just as good as others have said. Mine will start to slide as a slope less than 1/8" per foot and will keep spinning for quite a while.

That sliding to one end can be a PITA as a simple light push will send it speeding to a stop ("crash"?). It can be a worry if that uncontrolled rush involves a cutter crashing into a spinning grinding wheel.

Air quills are nice to have but need a lot of care as any "ding" on the quill that ends up inside the bearing can be a problem.

Because of the very close fit, a lot of care is needed if you pull the quill out as I do to store it away between uses. I take a lot of care to see that there are no "dings" in storage either.

Non-air quills work very well and there is no real advantage for an air quill in a small shop.

Here is my non-air quill which is excellent:
http://i200.photobucket.com/albums/aa294/oldtiffie/Universal_grinder/Universal_grinder20.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/Universal_grinder/Universal_grinder21.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/Universal_grinder/Universal_grinder22.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/Universal_grinder/Universal_grinder23.jpg

Here are some pics of using my "jury rigged" - for "demo" purposes - "Spindexer" as a quill - it works pretty well - but it is a pretty good "Spindexer" too:
http://i200.photobucket.com/albums/aa294/oldtiffie/ER-32_C5/ER32-C5-24.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/ER-32_C5/ER32-C5-26-1.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/ER-32_C5/ER32-C5-27.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/ER-32_C5/ER32-C5-28.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/ER-32_C5/ER32-C5-29.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/ER-32_C5/ER32-C5-30.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/ER-32_C5/ER32-C5-31.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/ER-32_C5/ER32-C5-32.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/ER-32_C5/ER32-C5-33.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/ER-32_C5/ER32-C5-34.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/ER-32_C5/ER32-C5-35.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/ER-32_C5/ER32-C5-36.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/ER-32_C5/ER32-C5-37.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/ER-32_C5/ER32-C5-38.jpg

IdahoJim
01-04-2011, 05:30 PM
Yes but if the air supply is to great there is no loss of pressure in the annulus when it's width increases, so you do not generate an out of balance restoring force.

The inlet holes of 0.3mm mention in a previous post limit the air supply. There is an ideal ratio between the inlet orifices size and number and the annular gap and length between the spindle and housing.

Phil:)

Aahhh...I see where you're going, Phil. I hadn't considered the volume of the air supply...I apparently missed the post mentioning the .3mm holes.
I've been back and forth between the house and the shop.
So, this being the case, why couldn't a an adjustable jet be installed to control volume?
Thanks,
Jim

philbur
01-04-2011, 05:44 PM
Reaction time related to the compressible reservoir volume, between the control valve and the point of inlet to the annulus, could possibly be a problem.

Phil:)


Aahhh...I see where you're going, Phil. I hadn't considered the volume of the air supply...I apparently missed the post mentioning the .3mm holes.
I've been back and forth between the house and the shop.
So, this being the case, why couldn't a an adjustable jet be installed to control volume?
Thanks,
Jim

oldtiffie
01-04-2011, 05:51 PM
So far as I can see, the body of my air quill has a manifold that surrounds the precision-honed hardened steel sleeve which has two rows of three quite large holes in it to carry the air to the precision-lapped quill spindle.

Air supply requirements are as per the panel on the body.

(See pics posted in my previous post).

IdahoJim
01-04-2011, 05:56 PM
This process for assembly of the sleeves will not necessarily result in the degree of alignment required. Think about it.

Phil:)

Yeah...I figured that out in the shop awhile ago...too much movement on the spindle. Should have shimmed them tight...still might not have worked, but would have been better. I can actually feel the spindle trying to float, but it's being pushed to one side at the right end. When I put a slight pressure against that movement it moves REALLY easily.
So what you're saying is an air spindle doesn't float from a Bernouli effect? In fact it's the opposite? The gap increasing on one side causes a drop in pressure, not from increased velocity of the air, but because the volume the air fills becomes larger, but there isn't enough air supply to fill the volume so the pressure drops? If that's the case then the size of the supply opening, and the air grooves on Glenn Wilson's rig are critical dimensions. Interesting.
Jim

Toolguy
01-04-2011, 05:59 PM
You might try putting another inlet hole opposite the first one in the new groove and tee off the air line to feed equally from both sides.

TGTool
01-04-2011, 05:59 PM
Reaction time related to the compressible reservoir volume, between the control valve and the point of inlet to the annulus, could possibly be a problem.

Phil:)

I think it's even more than that. Ideally you'd like a needle valve at each hole to adjust the volume at that point, not just the aggregate volume. The hole has to be small enough, relative to the pressure behind and the change in annulus section to let the pressure there drop when the spindle moves away from the hole. In other words, it has to be insufficient to make up the pressure so a differential develops relative to the opposite side. I know you understand this, Phil, this is really for the OP.

IdahoJim
01-04-2011, 06:13 PM
So far as I can see, the body of my air quill has a manifold that surrounds the precision-honed hardened steel sleeve which has two rows of three quite large holes in it to carry the air to the precision-lapped quill spindle.

Air supply requirements are as per the panel on the body.

(See pics posted in my previous post).

Thanks, Tiffie....and for the pics. I don't feel bad about this thing not working. As usual, I learned alot from you guys. I'm not quitting on this. I never feel bad because a project doesn't work the first time. I'll monkey around with what I've got, but will probably start over with a new housing. I ain't too smart, but make up for it with persistence.
Jim

IdahoJim
01-04-2011, 06:17 PM
You might try putting another inlet hole opposite the first one in the new groove and tee off the air line to feed equally from both sides.

I can't do that with this plan....there would be an interference.
What I don't understand is why the internal radial grooves don't accomplish the same thing, assuming they are large enough? Apparently that works fine on Glenn Wilson's spindle. There seems to be some difference of opinion amongst the experts on why this works...or doesn't.
Jim

lane
01-04-2011, 06:54 PM
1 1/2 -2 thousand clearance is what you need.spindle MUST be round and polished . Hole must be at least honed. I have built one and it works on 65- 90 pounds of air. Better at 90.
http://sites.google.com/site/machinistsite/TWO-BUDDIES/lanes-home/tooling/LTooling1.JPG?attredirects=0
3 holes 1/8 dia 120 degrees apart about 1/2 inch - 3/4 from each end. Will do it.

IdahoJim
01-04-2011, 07:37 PM
1 1/2 -2 thousand clearance is what you need.spindle MUST be round and polished . Hole must be at least honed. I have built one and it works on 65- 90 pounds of air. Better at 90.
http://sites.google.com/site/machinistsite/TWO-BUDDIES/lanes-home/tooling/LTooling1.JPG?attredirects=0
3 holes 1/8 dia 120 degrees apart about 1/2 inch - 3/4 from each end. Will do it.

Lane, will yours float, unloaded, at lower pressure?
Thanks,
Jim

PeteF
01-04-2011, 08:18 PM
How ironic this thread should come up now as I was only last night reading about building one. I could be completely wrong here, but my understanding is that the bore is not actually perfectly parallel, and is actually shaped to have a slight double taper making it slightly larger in the middle of the "body". The spindle on the other hand is perfectly parallel. So the clearance would be greatest in the middle of the device and taper down to least at the two ends. The air is fed in at the point or points where the clearance is greatest. Is my understanding incorrect?

Pete

IdahoJim
01-04-2011, 08:30 PM
How ironic this thread should come up now as I was only last night reading about building one. I could be completely wrong here, but my understanding is that the bore is not actually perfectly parallel, and is actually shaped to have a slight double taper making it slightly larger in the middle of the "body". The spindle on the other hand is perfectly parallel. So the clearance would be greatest in the middle of the device and taper down to least at the two ends. The air is fed in at the point or points where the clearance is greatest. Is my understanding incorrect?

Pete

I've read the same thing, Pete. There seems to be more than one way to skin this cat....LOL. If nothing else, as I was telling the wife tonight, my little rig is going to make a nice testbed before I start over. First thing I'm going to try is reducing the available air volume. May install a valve.
Jim

lane
01-04-2011, 08:33 PM
Lane, will yours float, unloaded, at lower pressure?
Thanks,
Jim

What do you mean . at 90 pound pressure it will float and spin all day

lazlo
01-04-2011, 08:39 PM
The air is fed in at the point or points where the clearance is greatest. Is my understanding incorrect?

I've only seen two amateur designs: Glenn Wilson's and the "Greatly Improved Quorn" design by Walter (can't remember his last name). Both suggest .0015" gap, with a very low taper tolerance (less than 2 tenths). Both indicate you have to hone the bore to get that straightness.

My commercial K.O. Lee air bearing is also straight, and floats forever on about 30 lbs of air.

That said, I can believe there are many other designs that would work.

This is my KO Lee:

http://i164.photobucket.com/albums/u15/rtgeorge_album/AirBearingHoles.jpg

PeteF
01-04-2011, 08:57 PM
I've only seen two amateur designs: Glenn Wilson's and the "Greatly Improved Quorn" design by Walter (can't remember his last name). Both suggest .0015" gap, with a very low taper tolerance (less than 2 tenths). Both indicate you have to hone the bore to get that straightness.

Hmmm, ok. I'll have to do some back tracking and see if I can find the source where I read that. Hey at least if Jim read it I know I'm not going crazy ... alternatively Jim and I are BOTH in trouble :D

Knowing absolutely nothing about these other than what I read, I know when I was reading through the description, having that "double taper" for lack of better description, seemed to be logical in my mind. I basically figured that the portion in the middle doesn't need to do anything anyway (if this was a conventional spindle that area wouldn't be supported) and would act as a "reservoir" for the air supply and help distribute it evenly. The air would then flow out toward the ends, with the gap between the two decreasing until it reached a critical dimension at which point it would be able to float the spindle, decreasing the gap any further than this would simply allow less air pressure to be used.

That was my take on it, but I'm probably talking out my clacker.

Pete

Edit: further to the above. It seems to me also that it would be easier to build something with a very slight taper going down to a critical dimension, rather than trying to hold that dimension along a length.

lazlo
01-04-2011, 09:02 PM
Hmmm, ok. I'll have to do some back tracking and see if I can find the source where I read that. Hey at least if Jim read it I know I'm not going crazy ... alternatively Jim and I are BOTH in trouble :D

Sounds like Jim is trying to follow the Glenn Wilson design, but I'm not sure...


Knowing absolutely nothing about these other than what I read, I know when I was reading through the description, having that "double taper" for lack of better description, seemed to be logical in my mind. I basically figured that the portion in the middle doesn't need to do anything anyway (if this was a conventional spindle that area wouldn't be supported) and would act as a "reservoir" for the air supply and help distribute it evenly.

I really meant that I'd only seen those two, not that other designs wouldn't work! :) That beautiful Stent that John posted sure looks like a unique air bearing design.

I vaguely recall someone discussing a (commercial?) design that had recessed groves offset from the front and rear. I think the idea was to provide a pressure reservoir to stabilize the spindle, like you're describing.

The other design I've seen is using porous sintered material (graphite) on the air bearing liner, so you effectively have a million tiny jets of air, uniformly distributing the pressure. That seems pretty far out of bounds of a HSM'er...

IdahoJim
01-04-2011, 09:05 PM
What do you mean . at 90 pound pressure it will float and spin all day
The reason for asking was that I've been using about 40 lbs for most of my testing. I wondered if yours, as have others I've read of, would float on as little as 5psi. I was curious if it was possible that one of these wouldn't work at all without pressure in the neighborhood of 65 psi.
Jim

lazlo
01-04-2011, 09:08 PM
The reason for asking was that I've been using about 40 lbs for most of my testing. I wondered if yours, as have others I've read of, would float on as little as 5psi.

Never tried it, but I can drag it out tomorrow. But the amount of air required is proportional to how finely tuned the spindle is. In the video that TGTool posted, the guy cranks his air valve open all the way, and you can hear the air rushing out :)

PeteF
01-04-2011, 09:19 PM
The guy at the end of this video talks about the specs of his air bearings

http://www.righthealth.com/topic/Air_bearing/Video

JCHannum
01-04-2011, 09:20 PM
The original air bearing project in HSM was by Phil Duclos. It is in the book "The Shop Wisdom of Phil Duclos" and probably one of the earlier HSM bound books. Later variations were done by Glenn Wilson and Walter Mueller. They are all about the same in concept.

IdahoJim
01-04-2011, 09:36 PM
Sounds like Jim is trying to follow the Glenn Wilson design, but I'm not sure...

Yup...that was the idea.....obviously I did a poor job of it....LOL
Jim

lazlo
01-04-2011, 11:08 PM
Yup...that was the idea.....obviously I did a poor job of it....LOL

It's not a poor job if you're learning something! Note that the video of the working air spindle that TGTools posts is "Mark II" :)

lazlo
01-04-2011, 11:12 PM
The guy at the end of this video talks about the specs of his air bearings

http://www.righthealth.com/topic/Air_bearing/Video

Ah, those are the New-Way sintered graphite bearings I saw:

http://www.youtube.com/watch?v=lOTWx69mghM

Bruce Griffing
01-05-2011, 12:02 AM
You still don't have it. You cannot use and adjustable jet, unless you can figure out how to make many of them. The point is that many tiny holes are needed, so that the airflow is limited by each hole. Then the response to an off center spindle varies by position. Airflow is limited by the local hole, so that pressure drops where the spindle to bore opening is larger. Conversely, where the spindle to bore spacing is smaller, the pressure rises.

oldtiffie
01-05-2011, 12:14 AM
How ironic this thread should come up now as I was only last night reading about building one. I could be completely wrong here, but my understanding is that the bore is not actually perfectly parallel, and is actually shaped to have a slight double taper making it slightly larger in the middle of the "body". The spindle on the other hand is perfectly parallel. So the clearance would be greatest in the middle of the device and taper down to least at the two ends. The air is fed in at the point or points where the clearance is greatest. Is my understanding incorrect?

Pete

Pete.

There are six holes about 1/4" + in the sleeve of my air-quill and it is fed directly from the 1/4" air hose. The annulus between the sleeve and the quill acts as a choke and impedes the air flow to regulate pretty constantly over a wide range of pressures. It is important that both the air pressure and volume are maintained.

http://en.wikipedia.org/wiki/Orifice_plate

http://en.wikipedia.org/wiki/Venturi_effect

http://en.wikipedia.org/wiki/Calibrated_orifice

The compressor needs to be pretty good to keep up the necessary "Free Air Delivery" litres or cubic feet per minute - and that needs to account for air-line losses between the compressor regulator and the quill of the air-quill.

Here is my compressor - note the Free Air Delivery - ie "Air Out"as compared to "Air In". The compressor range is 7 Bar (105 psi) "Switch On" and 10 Bar (145 psi) "Switch Off". Many compressors have a high of 7 Bar (105 psi) and a low of 6 Bar (about 90 psi) so they struggle with any high pressure high FAD demand put on them (same applies to Plasma cutters).

http://i200.photobucket.com/albums/aa294/oldtiffie/Compressor/Compressor_1.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/Compressor/Compressor_2.jpg

PeteF
01-05-2011, 12:29 AM
Tiff where did you get your air spindle from BTW?

oldtiffie
01-05-2011, 12:54 AM
Pete.

It came with the first of my two T&C grinders that I bought from H&F Victoria. It is superseded and was replaced by a new model that does not have and does not need an air quill.

Here are the links to the manufacturers sites:

http://www.chitseng.com/image/sa6.jpg

http://www.chitseng.com/product2.htm

http://www.chitseng.com/product5.htm

I prefer not to use an air quill. I use either a non-air quill or no quill at all.

When I was T&C Tool Room grinding the Cincinnati T&C grinders did not need a quill as we moved table. I still do it on occasion with my T&C grinders when grinding the spiral flutes of end-mill cutters, slab mills and "staggered" and non-staggered side and face cutters.

An air quill is not necessary for good T&C work at all although it makes some but not all jobs a bit easier.

I still use my "Spindexer" as well for the same purpose at times just to "keep my hand in".

MuellerNick
01-05-2011, 01:41 AM
The principle is the following:
* Restricted inlet
* distribution-area with constant pressure
* variable outlet
* even distribution of supports on a circular part of the bore


Restricted inlet:
A jet with a small bore. Small is something like the already mentioned 0.3 mm. All those jets are supplied (often) by a circular groove in the housing. Inside that housing sits a bushing with the jets.

Distribution area:
A recess connected to the jet that helps distribute the air over a smallish area of the spindle. Circular, longish, square, all works (depending on other factors)

Variable outlet:
Well, that is actually the gap between the bore and the shaft. If the shaft is moved in one radial direction, it reduces the gap and thus restricts the airflow and thus increases the pressure in that area and thus pushing back the shaft to equilibrum (SP?).

Distribution of supports:
Think of two circular areas at both ends of the bushing that do make the support. You have to keep in mind, that you want drop in air pressure when the gap increases. You can get that with minimal play between shaft and bore. You also have to think about the air-flow. If you look at two distribution-areas, you want to make sure that you get a drop of pressure in the area between them. So making them too close together, won't make much drop, because air can't escape. You could either reduce the number of distribution areas, or have a relieve channel between two distribution areas. With the later, you need a very small gap to reduce air consumption.

The shaft should be ground and lapped to be accurately cylindrical. It even goes that far that any smallest scratches from lapping that are helical will start rotating the spindle.

You could cast a bushing out of ABCoating (AirBearingCoating made by Diamant, makers of Moglice). To get the pockets, you coat the spindle with their seperator, stick shapes of wax onto the spindle where desired, cast the ABCoating, separate both parts and maybe lap down the spindle a bit more (you can have very tiny gaps by applying several coats of separator and polishing that. Can make gaps of a few m that way. But casting that stuff needs some experimenting!

Sidenote: The guy Frank Kurda mentioned here is a friend of mine.


Hope that helps (didn't read all postings)!
Nick

PeteF
01-05-2011, 02:15 AM
This manufacturer provides a very good video of their air bearings http://www.alfredjaeger.de/air-bearing-bushing.html

Also some very good information here under the support section http://www.newwayairbearings.com/

macona
01-05-2011, 02:24 AM
The reason for asking was that I've been using about 40 lbs for most of my testing. I wondered if yours, as have others I've read of, would float on as little as 5psi. I was curious if it was possible that one of these wouldn't work at all without pressure in the neighborhood of 65 psi.
Jim


This is my Darex air bearing at about 90.

http://www.youtube.com/v/XzPl-Mm0z4M

IanPendle
01-05-2011, 02:43 AM
This thread is interesting because I have just acquired a T&C grinder and was thinking about air spindles. I understand that a good air spindle is virtually friction free and glides effortlessly. It also has the benefit of keeping grit out of the bearing surfaces due to the continuously venting air.

However, would a linear ball bearing sleeve and a ground stock shaft not serve the same purpose? I have no experience of them, but they are quite cheap. Maybe there is still too much friction? I don't think that these things were around when air bearings were first used - perhaps a case of 'new' technology having an application here?

Anyone have any experience of this application?

Regards, Ian.

oldtiffie
01-05-2011, 03:27 AM
Ian,

the air-spindle only has those characteristics while the air pressure is on. If the air pressure is off the quill is just a very accurate cylinder in a very accurate bore, that because there is no air or oil to lubricate it makes it all but useless.

I get sick of mine wandering off on its own unless it is held captive. Despite what the videos may seem to say, it is not necessary for the air-quill to rotate before it starts to move.

I like a bit of "resistance" and tactile "feel" on a quill and the air quill can be a bit short in that regard.

I quite often have my quills restrained axially and just use them for rotation as I sometimes use the table for axial motion.

The old faithful "Spindexer" works well in that regard.

Quills are not much use for face-mills and for sharpening TC inserts and other stuff like side and face cutters, slab cutters, saws etc. as many of them require a universal head or a mandrel/arbor as well.

Using a quill requires very good set-ups as well as very good physical dexterity and co-ordination - and reflexes at times.

Safety is paramount as your hands will be very close to the wheel and one slip or lapse of attention may lead to a serious "gottcha" as your hand/fingers get dragged in.

If I haven't used my grinders for a while I go through several "cold runs" (wheel switched off) until I not only have it right but that it is pretty well second nature again - and that can take a while. It is not something that I will rush or take for granted.

A linear rail is more a possibility than a probability and if it were adequately practical we would be seeing more of them.

IanPendle
01-05-2011, 03:54 AM
Hello Tiffie,

Thanks for the reply. You said:

"I like a bit of "resistance" and tactile "feel" on a quill and the air quill can be a bit short in that regard."

If this is acceptable / desirable, then maybe a linear bearing sleeve would work - even if I have to put a felt wiper at both ends to keep out the grinding dust. After all, when used on a T&C grinder, the shaft and collet only need to slide a couple of inches max. whilst rotating a fraction of a turn at the same time.

In the UK there are 20mm ID linear ball bearing sleeves on ebay for about 8.00 each. At this price I will probably give one a try. I will report if successful.

Rgds. Ian.

oldtiffie
01-05-2011, 05:17 AM
Ian,

the dust and grit is not a big problem - or not as bad a some may have you believe - as witnessed with the stuff that gets on the ways on surface and T&C grinders. Its a fair guess there is a lot of loose air-bourne grit on many machines in many shops and people don't get too anal about that.

If people leave their air-quill assembled it is quite likely that it won't be covered adequately - if at all - and a fair amount of grit will be the quill when it is moved to the grinder and during that process the quill will move or be moved in its cylinder and if there is a time that grit will get in that's it.

Grit can get in from the air hose as there may be grit in the hose or the connector etc.etc.

People don't worry too much about the quills on milling heads or lathe tail-stock quills and they are pretty fine fits and finishes as well.

A bit of shop hygiene and a lot of common sense goes a long way.

In theory your linear rail may well be OK but it will need to have a bore big enough for a 5C collet draw-bar - unless you use an ER-32 adaptor and collets which need no bore or draw-bar at all.

I use ER-32 collets with my ER32>C5 adaptor in the C5 taper in my air quill.

I'd say give it a go as it looks promising and it might be OK but you will never know unless you try it.

The OP is doing exactly the same with his air-quill and he is going to stick with it and see where it takes him.

You could do worse with your linear rail concept.

Toolguy
01-05-2011, 09:09 AM
The linear bearing is a wonderful thing when moving lengthwise, but it will not want to rotate. If you want to use it to sharpen endmills it has to rotate effortlessly or you will never get a good cutting edge. I sharpen all my own endmills above 1/4" and make custom cutting tools with a Darex air bearing like the one in the video. It is a dream to use. I have the machine tilted just slightly so the spindle will retract if let go, to keep it from running the cutter into the grinding wheel. I also have the taper and radius attachment, but that is another subject.

IdahoJim
01-05-2011, 09:14 AM
You still don't have it. You cannot use and adjustable jet, unless you can figure out how to make many of them. The point is that many tiny holes are needed, so that the airflow is limited by each hole. Then the response to an off center spindle varies by position. Airflow is limited by the local hole, so that pressure drops where the spindle to bore opening is larger. Conversely, where the spindle to bore spacing is smaller, the pressure rises.

I don't want to appear argumentative, Bruce, but go look at the pic in post #40...Lazlo's KO Lee unit. Those are NOT tiny holes. There is more than one way to accomplish this....and I try to listen to ALL the info presented.
Thanks,
Jim

IanPendle
01-05-2011, 09:16 AM
Hello again Tiffie,

I was planning to go the ER32 route as I already have a full set of ER32 collets. I have just bought a ER32 collet chuck with 20mm x 100mm parallel shank (through hole and 3/8 whit drawbar thread) - ex Hong Kong. 19 delivered to the UK - how do they do this quality at this price?. Excellent quality with less than a thou run out. I just wish that I had bought the one with a 150mm parallel shank as it would be even better for this application.

Regards, Ian.

Bruce Griffing
01-05-2011, 09:42 AM
Idaho-
I realize there are many ways to make this work. I was only responding to your point about using a metering jet. My point was that if you meter, you have to do it at the inlet holes. Sorry about the tenor - I'm not trying to be argumentative either.

philbur
01-05-2011, 10:14 AM
If the KO Lee design is the same as the Darex then the orifices are set back in the large pockets that you see in the photo in post 40.

In my searching several years ago it appeared that there were at least 3 different design concepts that resulted in a floating spindle. Two amatuer and the Darex concept.

What you need for a precision spindle is rigidity, that is low deflection under load. The rigidity (once you have it floating) will largely be a function of width of the annular gap. Smaller gap means stiffer spindle.

Phil:)


I don't want to appear argumentative, Bruce, but go look at the pic in post #40...Lazlo's KO Lee unit. Those are NOT tiny holes. There is more than one way to accomplish this....and I try to listen to ALL the info presented.
Thanks,
Jim

GadgetBuilder
01-05-2011, 10:15 AM
I don't want to appear argumentative, Bruce, but go look at the pic in post #40...Lazlo's KO Lee unit. Those are NOT tiny holes. There is more than one way to accomplish this....and I try to listen to ALL the info presented.
Thanks,
Jim
Back at post #14 TGTool provided the basic design concepts and in post #15 I provided links so you could consider Frank Kurda's design which solves some of the problems mentioned in TGTools post. The last reference I provided in #15 has a picture of Kurda's bearing (click thumbnail below) showing the 0,3mm jets and the larger area interior divots needed to develop sufficient restoring force as described by TGTool. I submit that if you looked into the end of Kurda's bearing your conclusion would be that he didn't use tiny holes either - looks can be deceiving.

Your response to Bruce seems to indicate that you're not considering ALL the info presented, in particular not following the links provided, or you have rejected Kurda's approach without stating why.

http://img534.imageshack.us/img534/6082/airbush1.th.jpg (http://img534.imageshack.us/i/airbush1.jpg/)

Kurda indicates his spindle floats at 15psi; 60psi was used in the video to support the heavy chuck.

John

Uploaded with ImageShack.us (http://imageshack.us)

IanPendle
01-05-2011, 10:26 AM
Toolguy,

I seem to have overlooked your post - sorry. You said:

"The linear bearing is a wonderful thing when moving lengthwise, but it will not want to rotate. If you want to use it to sharpen endmills it has to rotate effortlessly or you will never get a good cutting edge"

I had somehow thought that this may be the case - perhaps I read it in the past and filed it away as "must remember" and then forgot. As I already have the collet chuck with a parallel shank, I will get one linear bearing sleeve to use on a trial basis. I'm sure that I can find another use for it if it doesn't work.

Regards, Ian.

IdahoJim
01-05-2011, 11:36 AM
Back at post #14 TGTool provided the basic design concepts and in post #15 I provided links so you could consider Frank Kurda's design which solves some of the problems mentioned in TGTools post. The last reference I provided in #15 has a picture of Kurda's bearing (click thumbnail below) showing the 0,3mm jets and the larger area interior divots needed to develop sufficient restoring force as described by TGTool. I submit that if you looked into the end of Kurda's bearing your conclusion would be that he didn't use tiny holes either - looks can be deceiving.

Your response to Bruce seems to indicate that you're not considering ALL the info presented, in particular not following the links provided, or you have rejected Kurda's approach without stating why.

http://img534.imageshack.us/img534/6082/airbush1.th.jpg (http://img534.imageshack.us/i/airbush1.jpg/)

Kurda indicates his spindle floats at 15psi; 60psi was used in the video to support the heavy chuck.

John

Uploaded with ImageShack.us (http://imageshack.us)

I'm sorry, but I haven't rejected anything. You have to understand that in 24 hours I've received several days worth of reading and study. Not only that, but I'm also doing things in the shop. If I'm not keeping up to anyone's satisfaction there's not much I can do about it. As far as Bruce's post...if it requires many tiny openings then why does Glenn Wilson's unit operate? It has one small hole. So, if I'm not getting it, there are also others who aren't either.
Jim

TGTool
01-05-2011, 11:41 AM
Ian,

I'm not saying that the linear bearing is a good idea but it may not be shot out of the water yet. There are two types to my knowledge, one intensely linear and the other not particularly. The type usually seen with linear rod slides have a linear bearing race with a track for recirculating the balls. These are often even preloaded slightly to eliminate or minimize backlash.

The other variety are those I've seen most often in my experience in press dies. It's simply a tubular bearing ball retainer with balls distributed in a pattern through its length and circumference. This would have no particular bias to move lengthwise versus rotationally and so might lend themselves to the spiral movement of endmill sharpening. Their downside is that the bearing retainer sleeve will move half the distance of the inner and outer elements making the movements between parts problematic. If they were restrained from movement to keep them from falling out or being otherwise mispositioned, then you lose the rolling element and the balls are then just skidding. Here's one illustration of this type http://www.anchorlamina.com/p_pinsbushings.html

lazlo
01-05-2011, 11:58 AM
If the KO Lee design is the same as the Darex then the orifices are set back in the large pockets that you see in the photo in post 40.

It's not -- the bore is perfectly smooth. There are no annular grooves or pockets.

MuellerNick
01-05-2011, 12:07 PM
It's not -- the bore is perfectly smooth. There are no annular grooves or pockets.


Did you look into the 3 bores?

It is part of the concept to get a pressure drop when the gap increases. If the supply of inrush air is unrestricted, the drop will get insignificant.



Nick

lazlo
01-05-2011, 12:13 PM
Back at post #14 TGTool provided the basic design concepts and in post #15 I provided links so you could consider Frank Kurda's design which solves some of the problems mentioned in TGTools post. The last reference I provided in #15 has a picture of Kurda's bearing (click thumbnail below) showing the 0,3mm jets and the larger area interior divots needed to develop sufficient restoring force as described by TGTool.

Kurda indicates his spindle floats at 15psi; 60psi was used in the video to support the heavy chuck.

John, I was curious about that, so I pulled-up Franks' post on the Yahoo Group.

It sounds like he lapped the spindle straight -- where's the thumbnail you reference that shows inside the bore?



http://groups.yahoo.com/group/quorn_owners/message/5145

"Runs at a pressure of 4atm [60 PSI] in the video - without such high load at the end it will loose friction at 1atm

The bushings have a gap of 0,015mm and every one has 36 holes 0,3mm diameter without "windows".

spindle and bushes are lapped (what a lengthy job) and have no measureable taper"

lazlo
01-05-2011, 12:18 PM
Did you look into the 3 bores?

It is part of the concept to get a pressure drop when the gap increases. If the supply of inrush air is unrestricted, the drop will get insignificant.

Yes, the inlet hole into the sleeve is much smaller -- is that what you mean?

Here's another shot, don't know if it's any better. That's a reflection in the fore-front of the lower air inlet -- it's perfectly smooth:

http://i164.photobucket.com/albums/u15/rtgeorge_album/AirBearing1.jpg

MuellerNick
01-05-2011, 12:46 PM
Yes, the inlet hole into the sleeve is much smaller -- is that what you mean?


Yes. Post #54:
Distribution area (3 big bores) & restricted inlet (much smaller inlet holes).


Nick

philbur
01-05-2011, 01:10 PM
We may be missing one another in the wording. From memory the Darex looks exactly like your photo of the KO Lee. IIRC it required a mirror in the bore to see the orifice hole in the recess. There was a photo link and dimensions, including a measured orifice size, somewhere on the Yahoo site.

Phil:)


It's not -- the bore is perfectly smooth. There are no annular grooves or pockets.

GadgetBuilder
01-05-2011, 01:11 PM
Lazlo,

Not sure why you're not seeing the thumbnail in my previous post but here's a link that might work:
http://img534.imageshack.us/i/airbush1.jpg/

It shows the air bearing removed from its holder. Kurda has a 3D drawing (no dimensions) of the overall spindle here:
http://groups.yahoo.com/group/quorn_owners/files/Frank%20Files/

A couple pictures of it here:
http://groups.yahoo.com/group/quorn_owners/photos/album/438477972/pic/list


As I understand what Kurda did, he lapped the spindle to a constant diameter then lapped the bearings to fit. Not sure how he did this, perhaps by trial and error (tricky because the bearings are brass so they would retain abrasive). I think he tried bearings with 24 holes (Mark I) and then a second set with 36 holes (Mark II). I believe the holes were enlarged with a Woodruff cutter on the ID of the bearing. He has very small clearance between the spindle and bore so I expect it took a LOT of lapping to get the spindle and bores perfectly round to allow this. The air bearings use O rings so this may help to align between ends (which could be tricky too). He seems to be a very competent machinist who does beautiful work.

My guess is that Kurda's design would be more straightforward to get working than the Darex design but I haven't attempted either so this is just a guess based on how he addressed the issues noted earlier by TGTool.

There is also an analysis of the Darex type air bearing at that site, see "airbearing pressures":
http://groups.yahoo.com/group/quorn_owners/files/

John

philbur
01-05-2011, 01:28 PM
here is the original photo of the Darex orifice.

http://i186.photobucket.com/albums/x36/philbur/airspindle1.jpg

Phil

philbur
01-05-2011, 01:42 PM
When Kurda posted his design a few years back I could not work out why it actually worked. He seemed to have to many orifices (flow area) for the size of the annular gap (flow area). Also he had no air pads. With no disrespect to Kurda I think it may have worked (floated the spindle) in spite of the design flaws. I think the Darex is the "correct" technical design.

Also I think the design published some years ago in HSM under the title "A much improved Quorn" was not an optimum design. It included a circumferential groove, which supplied air around the spindle annulus, but of course at the same time provided a means to equalise the pressure around the annulus, which is exactly what you don't want. But again the design did float the spindle.

Just some thoughts
Phil:)

TGTool
01-05-2011, 01:44 PM
GadgetB,

Those patent drawings are interesting. I must have looked at them once before and forgot it. Their approach is to just drill large holes for the pressure pad area, then meter air to them with the small V-grooves feeding from the central gallery. It certainly takes the pucker factor out of the tiny holes but I wonder if they really worked out the calculations for restriction to flow in that cross section and length or just developed it empirically. That might give one a way to try different settings by making some sort of adjustable restriction in those grooves but it might also be hard to get them all balanced.

One ought to also be able to get into the ballpark of the Kurda design by calculating the aggregate hole area versus annulus area to get a sense of proportion. If his first attempt with 24 holes wasn't successful but 36 holes was, that should give you a range and at least a lower limit.

lazlo
01-05-2011, 02:10 PM
Lazlo,

Not sure why you're not seeing the thumbnail in my previous post but here's a link that might work:
http://img534.imageshack.us/i/airbush1.jpg/

Crap -- I just realized you're using ImageShack, which OpenDNS blocks :mad:


Kurda has a 3D drawing (no dimensions) of the overall spindle here:
http://groups.yahoo.com/group/quorn_...Frank%20Files/

Ah, now that I can see (the 3D model). The outer chambers are interesting.

Phil: the K.O. Lee is a much bigger and heavier design than the Darex unit. About 50 lbs, all assembled.

But I'll stick a mirror up there when I get home from work.

GadgetBuilder
01-05-2011, 02:46 PM
Phil,

I'm not up on the terminology, but aren't the larger divots made with a Woodruff cutter in Kurda's bearing air pads? It seems tricky to get the restoring force to work with fewer holes, perhaps that's why Kurda used so many, so there was a pad almost directly in line with the required restoring force.

His first bearing worked but the clearances were large so it used a lot of air, video of his Mark I:
http://www.youtube.com/watch?v=65Nv7VjEqGM

He wasn't happy with this because the flow was too large and also because it would sometimes motor in one direction.

He said he would glue the 36 hole bearings in place prior to lapping in order to achieve reduced clearances = less flow. Not clear why he increased the number of holes from 24 to 36 but he may have done this and glued them in place making it difficult to say which change produced what effect. An air bearing is a difficult thing to experiment on because it takes a LOT of work for each variable tried.

The picture of the Darex holes is interesting because the bore looks to be honed rather than lapped, not what I expected.


TGT,

Success is in the eye of the beholder here - I'd probably count Kurda's Mark I as a success and buy a big compressor ;-)

I think the idea of the metering hole feeding the large area is as you explained in post #14 - without that enlarged area, blocking a hole would increase the pressure but on a tiny area so restoring force = pressure*area, would actually drop because area would decrease radically. At least that's my theory (based on no experience).

In practice I'm a barbarian and use plastic bearings in my helix sharpener; easy to make, reasonably low friction, no compressor needed.

John

MuellerNick
01-05-2011, 02:48 PM
I remember when Frank (Kurda) started a discussion about that in "my" forum. No one had definite answers and especially, there were no dimensions at all.
So he made a wild guess and built it without messing around. And it worked.

He might reduce the number of inlets and increase the size of the pockets. In my eyes, the gap is too big with 0.015 mm.

Air spindles need an air filter and water separator in front of them.

My feeling is, that you either have luck with your dimensions, or you'll have to make a bunch of prototypes to get a perfect one.

Blockhead is an US manufacturer, maybe you can find more there.


Nick

lazlo
01-05-2011, 03:23 PM
Blockhead is an US manufacturer, maybe you can find more there.

A Block-Head runs $3,000 - $5,000 :) New-Way (that Pete poster earlier) is around the same price.

Diamond turning with Blockhead air bearing
http://www.youtube.com/v/QQa9mB6OgV8?

philbur
01-05-2011, 03:40 PM
I think these type of bearings have very, very narrow annular gaps, which makes for a very stiff spindle capable of taking high radial loads. It's a whole new ball game.

Phil:)


A Block-Head runs $3,000 - $5,000 :) New-Way (that Pete poster earlier) is around the same price.

Diamond turning with Blockhead air bearing
http://www.youtube.com/v/QQa9mB6OgV8?

PeteF
01-05-2011, 04:11 PM
A Block-Head runs $3,000 - $5,000 :) New-Way (that Pete poster earlier) is around the same price.

I've seen people using New Way air bushings in linear tracking turntables, and a relatively recent post on a DIY audio BB I read quoted the price of the air bushings at between US$2-300 for the bushing and matching pillow block. However that was for a small size suitable for a tonearm, a size we're talking here is possibly considerably more expensive.

While it may not help the OP one iota, from what I have been reading, the real breakthrough in making these things truly affordable has been with the availability of a porous medium from which to construct the bore. In addition, it seems the finish of the spindle itself is not critical (with that configuration at least) and New Way make a point that their bearings are often replacements for conventional bearings where standard precision ground shafting is used.

I know it was dismissed as an alternative very early on, but with the above in mind, I'm now wondering if it would be easier to make these using some form of porous material? I recall many years ago having a vacuum pump that used carbon rotor vanes, but more the point some form of porous carbon that was used as a final filter on the inlet. I've tried searching for that material but I've no idea where to start. Anyone have any ideas? I've also read about people using porous bronze(?) liners for this purpose. Again, something I'm personally not familiar with but some here no doubt are. Since there's no contact, I see no reason at all why the bore needs to be any form of metal, and I guess it could be quite soft material.

I think the end goal needs to be kept in mind here. At the end of the day the primary intent in this application is a very low friction spindle. Rigidity will be somewhat important to get a good finish, but recall these are cutters being sharpened and even when new the edge grinding is clearly evident. By design these spindles have a very low gap, and even if the spindle went to the limit of that gap (obviously impossible as that's a crash), it's still not much, and I feel the miniscule deflection is a non-issue. Finally, the load bearing of a spindle large enough to take collets is absolutely miniscule, the cutter isn't exactly shoved into the wheel, so again I feel that's a non-issue and those two factors should considerably ease the design constraints ie if it floats it works!

Pete

Edit: FWIW, provided suitable material could be found, I thought the video by New Way on the gimballed air bearing pads was the most likely to be the design I personally would be pursuing. It's quite different to the traditional approach, but I thought using the difference in stiffness to create the appropriate gap was very clever. Also the gimballed design was self centering. Maybe it was just deceptively simple, but lapping pads to a shaft and installing on gimbals was something I definitely saw as possible in a home shop environment.

lazlo
01-05-2011, 04:28 PM
from what I have been reading, the real breakthrough in making these things truly affordable has been with the availability of a porous medium from which to construct the bore.

New-Way is using a porous graphite - is there a source for that? It would be nasty to machine, but still intriguing.


In addition, it seems the finish of the spindle itself is not critical (with that configuration at least) and New Way make a point that their bearings are often replacements for conventional bearings where standard precision ground shafting is used.

Wouldn't the spindle sleeve still need to be incredible straight? The porous material gives you fantastic pressure distribution, but seems like you'd still have stability issues if the bore wasn't perfectly honed.

PeteF
01-05-2011, 04:44 PM
Well recall reference to my clacker, and where most of my opinions are currently emanating from, but yes I would think the spindle would need to be "straight" but not necessarily lapped to a fine finish. A few manufacturers have made the point that the gap basically averages out any roughness in the spindle finish. I would guess that as long as it's not so bad there's contact, it would be fine. However the smaller the gap (hence the stiffer the bearing) the more finely finished the spindle would need to be.

It seems to me an analogy could be made to a hovercraft. Forget the flexible skirt, that's just an addition, but even if the skirt was rigid, the hovercraft would happily skim over obstacles up to the size of the air "cushion" upon which it is riding. Ok the obstacles may only be an inch high, or however high is that gap, but providing they're smaller than the gap the hovercraft doesn't even "see" them.

That's my take on it anyway :)

PeteF
01-05-2011, 04:52 PM
Oh sorry I initially missed your point about the bore being perfectly honed. Yes I think that's where the secret is ... to a degree. The manufacturers make a point that any scratches of the porous media don't effect performance; you simply lose load bearing in that small area. Again it seems to me that if the bore wasn't perfect, and deviation to increase the bore size would simply reduce load capacity in that region. The shaft should self-centre provided the bore was perfectly round, I would guess it would be that latter point that would be the most difficult and why I thought the gimballed pads had the most hope for success.

Pete

Toolguy
01-05-2011, 05:02 PM
An oilite bearing is very porous. If you can find one the right size, just wash the oil out with solvent.

PeteF
01-05-2011, 05:36 PM
BTW this was the video to which I was referring in case I lost anyone with my tangent :)

http://www.youtube.com/watch?v=lOTWx69mghM

IdahoJim
01-05-2011, 08:44 PM
It included a circumferential groove, which supplied air around the spindle annulus, but of course at the same time provided a means to equalise the pressure around the annulus, which is exactly what you don't want. But again the design did float the spindle.

Just some thoughts
Phil:)

That's more like what I'm trying, Phil. This thread is sure interesting. It's going to take some time to digest all the info, though I doubt I'll ever try to build these more complex designs.
I did make some progress with mine today...when I first tried it, the air actually locked it up. When I quit this afternoon the air was making it turn easier. I did put a needle valve on the single air inlet to control the volume. That did help, though as the lapping progressed the volume supplied became less critical. That was probably because the gap was growing. I think it would be usable right now, as a tool grinder....fairly rigid from 65-90 psi, and turns very smooth, but it does NOT float.....LOL
I still haven't conceded.....though I may. More work to do tomorrow.
Thanks to all,
Jim

Bruce Griffing
01-05-2011, 10:04 PM
Given your starting point, I think there is another way to modify your design to make it work. The key is to make the overall bearing bore so it is divided into 3 or more parts. The division must be axial. So if you have 3 air supply holes in the axial center, distributed at 0, 120 and 240 degrees you can make three axial grooves at 60, 180 and 300 degrees along the inside of the bore. These grooves go from one end of the bore to the other. They serve to keep the airflow from each segment separate by providing an easy path to the outside. The same idea would work with 4 or more segments, but I am guessing that three would be optimal. You could augment the design by having more sets of air supply holes displaced from the axial center in both directions, but at the same angles. Just something to try.

IdahoJim
01-05-2011, 11:02 PM
Given your starting point, I think there is another way to modify your design to make it work. The key is to make the overall bearing bore so it is divided into 3 or more parts. The division must be axial. So if you have 3 air supply holes in the axial center, distributed at 0, 120 and 240 degrees you can make three axial grooves at 60, 180 and 300 degrees along the inside of the bore. These grooves go from one end of the bore to the other. They serve to keep the airflow from each segment separate by providing an easy path to the outside. The same idea would work with 4 or more segments, but I am guessing that three would be optimal. You could augment the design by having more sets of air supply holes displaced from the axial center in both directions, but at the same angles. Just something to try.

Thanks, Bruce....that's an interesting concept, and something I hadn't thought of. That's the first time I've heard of axial grooves that go all the way to the outside. It wouldn't work on my current model, with just one air inlet, but I may try that, assuming I can't get this one to work. At some point I'll start all over in any event, just for fun. This whole air bearing idea is an interesting concept, with what looks like, lots of room for experimentation.
My current housing has 2 radial grooves that supply air to the offside. They are fed by an axial groove from the inlet. I also added a second axial groove between the radial grooves, opposite the air inlet. That's one reason that your idea won't work on this unit. The other reason is difficulty with multiple inlets...they'd interfere with other things on the grinder. I do like the idea of separate pressure areas. I'm really interested in anything that reduces the amount of precision work required. It would be great to come up with a plan that would allow people (like me...LOL) with lesser skill & equipment to have the benefits of the air bearing. May not be possible, but fun to try.
Thanks,
Jim

oldtiffie
01-05-2011, 11:07 PM
Providing that I keep my air-pressure within the specified limits the pressure along the quill is self-regulating.

The clearance between the bore and the quill acts as a "choke" or "self/auto regulating/rated orifice" (annular space) and all works fine.

http://i200.photobucket.com/albums/aa294/oldtiffie/Air-quill/Air-quill2.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/Air-quill/Air-quill4.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/Air-quill/Air-quill1.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/Air-quill/Air-quill6.jpg

In the case of my air quill the air pressure limits are:
6.4 kg/cm^2 x 14.22 = 91 psi

8.6 kg/cm^2 x 14.22 = 121 psi

That may well take some doing with the normal range of compressors - hence why my compressor works between 7 Bar (~101 psi) and 10 Bar (~ 145psi) and while the compressor regulator controls the pressure it can only do it if the Free Air Delivery - FAD - (in litres or gallons per minute) can be sustained.

I bought that compressor for the air quill as well as the plasma cutter which has similar requirements.

PeteF
01-05-2011, 11:12 PM
Bruce I was thinking the same thing. Furthermore, going back to the video link I posted a few posts ago, I really like the design concept of using the dynamic stiffness of the air bearings to overcome the static stiffness of the housing, thus automatically providing the appropriate gap. It's probably not something Jim could incorporate in the design as it stands, but smarter minds may come up with a solution down that path.

I'm intrigued by the design and wondered if a spindle such as this would have enough length for typical cutters? http://cgi.ebay.com.au/ws/eBayISAPI.dll?ViewItem&item=160378334294&ssPageName=STRK:MEWAX:IT#ht_1769wt_907

I envisaged using the bearing material itself as a lap for the spindle, thereby auto-generating an appropriate fit. As long as the pores of the material could be made to flow freely, the non-contact nature of the bearing means any remaining lapping abrasive is immaterial. Once lapped, saw the bearing into either 3/4 segments, then mount the segments in appropriate holders on threaded gimbals in a somewhat compliant housing that would allow negative clearance as per the New Way design. Once the air is introduced the bearing stiffness would overwhelm the compliance of the housing and the appropriate gap should be obtained. Well that's my theory anyway :D

Jim, if all else fails and assuming you don't solve the situation any other way, maybe you could saw the housing into quarters or thirds and rebuild the housing using compliant bands (like a number of large hose clamps)? It may close up the gap, yet the compliant nature of the bands may allow a gap to form. ... Maybe :D

PeteF
01-05-2011, 11:19 PM
In the case of my air quill the air pressure limits are:
6.4 kg/cm^2 x 14.22 = 91 psi

8.6 kg/cm^2 x 14.22 = 121 psi

That may well take some doing with the normal range of compressors - hence why my compressor works between 7 Bar (~101 psi) and 10 Bar (~ 145psi) and while the compressor regulator controls the pressure it can only do it if the Free Air Delivery - FAD - (in litres or gallons per minute) can be sustained.

I bought that compressor for the air quill as well as the plasma cutter which has similar requirements.

Tiff are you saying the compressor has a hard time meeting the air demands of your air spindle? :eek: If so that's a LOT of air it's using. In contrast look at the air lines leading to and specs of some of the other commercial spindles here. They really don't seem to use much air at all, indeed I was quite surprised. Could it be that the designer of yours simply uses "brute force and ignorance" (ie a lot of air) to make that design fly?

oldtiffie
01-05-2011, 11:34 PM
Nope Pete.

The pressure is only maintained while the free air delivery matches or exceeds the air consumption of the air quill.

The FAD is very small indeed for the air quill but is very large for the plasma cutter. I move my compressor as close as I can to the plasma cutter to minimise line losses but that is not a problem with the air quill as the compressor is only about two metres from it.

The big problem for smaller or "normal" compressors is their pressure range - say 6 Bar (~ 90 psi) to say 7 Bar (~105 psi) when my air quill range is 90>120 psi - at the quill - not at the regulator - as line losses are a problem. The best that small compressor can do is to put out its FAD with a line pressure only in the bottom half of the air-quill specs.

With a low FAD, line losses are minimised.

IdahoJim
01-05-2011, 11:35 PM
Jim, if all else fails and assuming you don't solve the situation any other way, maybe you could saw the housing into quarters or thirds and rebuild the housing using compliant bands (like a number of large hose clamps)? It may close up the gap, yet the compliant nature of the bands may allow a gap to form. ... Maybe :D
I've been thinking of something like that since seeing that New Way video. I doubt I'd do it with the current housing, but it's definitely an idea for a fresh start. I find myself constantly glancing at the shop junkpile for holey material...Haha You guys sure do cause me alot of mental turmoil....LOL
Jim

IdahoJim
01-05-2011, 11:39 PM
Providing that I keep my air-pressure within the specified limits the pressure along the quill is self-regulating.

The clearance between the bore and the quill acts as a "choke" or "self/auto regulating/rated orifice" (annular space) and all works fine.

.

One thing I noticed in the pics, Tiff, is the temperature range. I'm working in a mostly 20 to 50* shop at this time of year...I imagine the clearances in my unit are going to open up once the weather warms. I wonder how much larger they would get on a roughly 1" diameter.
Jim

oldtiffie
01-05-2011, 11:43 PM
Jim.

Try the Navy - they may still have some holy stone.

http://en.wikipedia.org/wiki/Holystone

IdahoJim
01-05-2011, 11:52 PM
Jim.

Try the Navy - they may still have some holy stone.

http://en.wikipedia.org/wiki/Holystone
I was Air Force.....and, thankfully, I didn't have to scrub any decks....LOL, and I don't plan on startin' any time soon.
Jim

oldtiffie
01-05-2011, 11:57 PM
Originally Posted by oldtiffie
Providing that I keep my air-pressure within the specified limits the pressure along the quill is self-regulating.

The clearance between the bore and the quill acts as a "choke" or "self/auto regulating/rated orifice" (annular space) and all works fine.
One thing I noticed in the pics, Tiff, is the temperature range. I'm working in a mostly 20 to 50* shop at this time of year...I imagine the clearances in my unit are going to open up once the weather warms. I wonder how much larger they would get on a roughly 1" diameter.
Jim

http://i200.photobucket.com/albums/aa294/oldtiffie/Air-quill/Air-quill2.jpg

Jim,

the standard temperature for calibrating a lot of tools and guages is 20*C (~ 68*F).

+/_ 3.5*F is OK as 3.5*F x 5/9 ~ 2*C

So it is 68 +/- 3.5 = 71.5/64.5*F which is pretty wide. My shop is mostly within the 20 +/_ 2*C = 22/18*C and if it isn't I have noticed any deterioration of the air-quill performance. The warmer compressed air pretty well sorts that out from a practical view-point.

Height guages and slip guages (micrometers too) are calibrated at 20*C/68*F and its rarely an issue and it can be "worked around" most times if needs be.

IdahoJim
01-06-2011, 12:18 AM
http://i200.photobucket.com/albums/aa294/oldtiffie/Air-quill/Air-quill2.jpg

Jim,

the standard temperature for calibrating a lot of tools and guages is 20*C (~ 68*F).

+/_ 3.5*F is OK as 3.5*F x 5/9 ~ 2*C

So it is 68 +/- 3.5 = 71.5/64.5*F which is pretty wide. My shop is mostly within the 20 +/_ 2*C = 22/18*C and if it isn't I have noticed any deterioration of the air-quill performance. The warmer compressed air pretty well sorts that out from a practical view-point.

Height guages and slip guages (micrometers too) are calibrated at 20*C/68*F and its rarely an issue and it can be "worked around" most times if needs be.

Well..that's good. I was afraid I might be looking at loss of rigidity, or a significantly larger air requirement.....this all on the assumption I'll eventually get this mess sorted out.
Jim

MuellerNick
01-06-2011, 03:43 AM
You researchers will find a lot more information if you look for hydrostatic bearings. In fact, it is one with just the medium replaced.


Nick

lane
01-06-2011, 07:07 PM
Some photos of mt darex and home made air spindle /.
http://i178.photobucket.com/albums/w277/lane5263/Air%20bearing%20spindles/Airbearings003Darex.jpg

http://i178.photobucket.com/albums/w277/lane5263/Air%20bearing%20spindles/Airbearings002Darex.jpg

http://i178.photobucket.com/albums/w277/lane5263/Air%20bearing%20spindles/Airbearings005Homemade.jpg

http://i178.photobucket.com/albums/w277/lane5263/Air%20bearing%20spindles/HomemadeAirbearing-1.jpg
The home made one has 2 groves cut in for the air to escape.

oldtiffie
01-06-2011, 07:14 PM
Very nice work Lane - as always