Table Tilt Lock

Forum for Maintenance and Repair topics. Feel free to ask questions or contribute.

Moderator: admin

User avatar
JPG
Platinum Member
Posts: 35600
Joined: Wed Dec 10, 2008 7:42 pm
Location: Lexington, Ky (TAMECAT territory)

Post by JPG »

algale wrote:Did Shopsmith mill off the area shown in your post #58 or did they create the ledges shown in #58 by milling lower/deeper into the pad area where the lock wedge sits/travels. If it is the latter, the wedge would sit lower relative to the trunnion and the wedge would need to travel further in order to apply the same amount of contact force as a lock wedge sitting on a higher pad. Does that make sense?

It appears to me that the milled area in #58 is 'deeper' than that in the later post, and I think that explains the 'wear' also. The milling cutter also cut into the lip of the 'ledge'.

I measured the 'rod'/wedges/clamp nut and determined that the clamp nut will run out of thread when the outer edges of the wedges are 16 1/2" apart. When clamped, the separation is 16 5/8". That is only 1/8" 'to spare', but is approximately the same as the wedge to the 'bottom' of the milled area when clamped.(about half on each end) The rear end of the rod is bottomed out about 5/8" into the wedge.

The 'acute' angle is about 30 °, so the vector force would still have a significant horizontal magnitude. The wedge should slide smoothly on the milled area so it will not 'cock'/lock as a bar clamp would.

I agree the clamp will exert force normal to the trunion angle that results in 'squeezing' the trunion against its pivot. The milled area and flat side of the wedge counteracts that force as the clamp nut provides the lateral force to the wedge. I still think the three ridges merely limit trunion deflection as they are squeezed. That will provide some friction, but small compared to the wedge/trunion surface. The fact that the $ shim has any effect indicates the trunions do indeed deflect ever so slightly. FWIW, that also affects the resultant 'normal' force to the trunion and may explain the reason for weak clamping.

I would move the rear trunion so it butts up against the trunion pivot and see if propensity to slip is reduced.

I have come to the 'conclusion' that table alignment should merely reposition the table to the trunions and that the trunions must bear against the pivots horizontally(that is why the clamp is tightened and the $ used as a shim since it holds the trunions in position).

Out loud mental meandering!
╔═══╗
╟JPG ╢
╚═══╝

Goldie(Bought New SN 377425)/4" jointer/6" beltsander/12" planer/stripsander/bandsaw/powerstation /Scroll saw/Jig saw /Craftsman 10" ras/Craftsman 6" thicknessplaner/ Dayton10"tablesaw(restoredfromneighborstrashpile)/ Mark VII restoration in 'progress'/ 10
E[/size](SN E3779) restoration in progress, a 510 on the back burner and a growing pile of items to be eventually returned to useful life. - aka Red Grange
User avatar
algale
Platinum Member
Posts: 4841
Joined: Thu Nov 04, 2010 8:13 am

Post by algale »

JPG wrote:It appears to me that the milled area in #58 is 'deeper' than that in the later post, and I think that explains the 'wear' also. The milling cutter also cut into the lip of the 'ledge'.

I measured the 'rod'/wedges/clamp nut and determined that the clamp nut will run out of thread when the outer edges of the wedges are 16 1/2" apart. When clamped, the separation is 16 5/8". That is only 1/8" 'to spare', but is approximately the same as the wedge to the 'bottom' of the milled area when clamped.(about half on each end) The rear end of the rod is bottomed out about 5/8" into the wedge.

The 'acute' angle is about 30 °, so the vector force would still have a significant horizontal magnitude. The wedge should slide smoothly on the milled area so it will not 'cock'/lock as a bar clamp would.

I agree the clamp will exert force normal to the trunion angle that results in 'squeezing' the trunion against its pivot. The milled area and flat side of the wedge counteracts that force as the clamp nut provides the lateral force to the wedge. I still think the three ridges merely limit trunion deflection as they are squeezed. That will provide some friction, but small compared to the wedge/trunion surface. The fact that the $ shim has any effect indicates the trunions do indeed deflect ever so slightly. FWIW, that also affects the resultant 'normal' force to the trunion and may explain the reason for weak clamping.

I would move the rear trunion so it butts up against the trunion pivot and see if propensity to slip is reduced.

I have come to the 'conclusion' that table alignment should merely reposition the table to the trunions and that the trunions must bear against the pivots horizontally(that is why the clamp is tightened and the $ used as a shim since it holds the trunions in position).

Out loud mental meandering!

Just intuitively, it seems to me that if the trunnion isn't flush against the three milled ridges on the table bar when the lock is applied, there's going to be a lot less locking force. I've been saying that from the beginning of this thread.

But what I find curious is why Shopsmith went to trouble to form those three separate ridges behind the trunnion and then to mill them very precisely, when one raised ridge would seem to do the trick if backing up the trunnion were the major point of it.

So, without anything more than intuition, I'm guessing there's a significant friction factor also at work. Those three milled ridges on the table bar actually provide much more surface area to contact that trunnion than the lock wedge itself. If you look at the lock wedge in the photo you posted of the disassembled locking mechanism http://shopsmith.net/forums/showpost.htm?p=179058&postcount=53, only the very edges of the lock wedges touch the trunnion.
User avatar
JPG
Platinum Member
Posts: 35600
Joined: Wed Dec 10, 2008 7:42 pm
Location: Lexington, Ky (TAMECAT territory)

Post by JPG »

algale wrote:Just intuitively, it seems to me that if the trunnion isn't flush against the three milled ridges on the table bar when the lock is applied, there's going to be a lot less locking force. I've been saying that from the beginning of this thread.

But what I find curious is why Shopsmith went to trouble to form those three separate ridges behind the trunnion and then to mill them very precisely, when one raised ridge would seem to do the trick if backing up the trunnion were the major point of it.

So, without anything more than intuition, I'm guessing there's a significant friction factor also at work. Those three milled ridges on the table bar actually provide much more surface area to contact that trunnion than the lock wedge itself. If you look at the lock wedge in the photo you posted of the disassembled locking mechanism http://shopsmith.net/forums/showpost.htm?p=179058&postcount=53, only the very edges of the lock wedges touch the trunnion.
Again the acute angle affects resultant forces. There will be greater force against the trunion bevel than the back side against the three ridges.

I think the three ridges are to increase the 'span' of the three pads without excessive surface area to interfere as the trunion is rotated past them.

I think they are milled to match the trunion pivot width.(slightly less???)

Also smaller area results in greater contact force/area which is what actually provides resistance to moving.
╔═══╗
╟JPG ╢
╚═══╝

Goldie(Bought New SN 377425)/4" jointer/6" beltsander/12" planer/stripsander/bandsaw/powerstation /Scroll saw/Jig saw /Craftsman 10" ras/Craftsman 6" thicknessplaner/ Dayton10"tablesaw(restoredfromneighborstrashpile)/ Mark VII restoration in 'progress'/ 10
E[/size](SN E3779) restoration in progress, a 510 on the back burner and a growing pile of items to be eventually returned to useful life. - aka Red Grange
User avatar
reible
Platinum Member
Posts: 11283
Joined: Thu Jul 20, 2006 12:08 pm
Location: Aurora, IL

Post by reible »

Well this has been quite a thread. I think I will have to read it a second time and spend sometime thinking on the subject.

I wish I could go and look at mine but it is over 5 pounds so that will have to wait.

I can tell you that I have not experienced the movement issue talked about here. I can also tell you that during some testing I did that table movement can come from places other then the tilt area. When I did my study on the subject (posted here some where) I was looking to stabilize the table and showed the effects of using the tubes connecting to extension tables. The improvements by doing that were major and is the method I use now when trying to get precision cuts. Of course when tilted that is not possible so things change.

Second I can tell you that I have had 3 table tops on one of my machines and simply moved the parts and did an alignment with no effect on table clamping. The first table was from a 510r that I got as a basket case. The PO had spilled something on it and it was only removable by force since it was stuck in the valleys, the top was sanded to remove it from the flats but it still looked like........ The second table was another 510 table and it was in pretty good condition but it had the full set of ribs which I found I did not like all that much. I then replace it with a table that had the smooth center section. The under pinning of the table had the hole issue which I drilled out, it was really hard to align before that so I saw that as a major improvement.

While I understand why one would want to fix this issue I think some of the ideas posted are not correct. I would not expect any "spring" action to be involved in the design except when a spring is used. In fact I would believe the design would be done to minimize that action.

I think the design would have portions would be looked at as a fix portion and section that would be a moving portion. Fix portions that are designed to not move and not expected to do so. This can be thought of a constraint of the design. This can be accomplished by limiting the forces applied to that area. An example would be the lever that is used to apply the force to the moving parts. You will only get just some much force from the tightening, this can be figured out or measured and it will be much less then can start moving the fix portion of the design.

Other design factors such as the area that contact from the fix portion to the moving portion were also factors of the design. That fit would have been designed to allow movement when in one position and to resist movement of those connected parts when tight.

I might even question if the design was to having a cam action, it quite possible that the angle was used to gain area, the sloped parts will have much more area then if they were just flat.

I'll not go farther then this on that subject. At this point I can only guess.

What I would like to see happen is to have this called in to shopsmith. They have the drawing/note books/design intent to look at. Let them have a look at it and see what they have to say. Perhaps them might even want to look at a system that is not working correctly to find out if something is out of tolerance/damaged/or for a potential design flaw. I think they have to potential to solve this issue better then any of us do.

Ed
User avatar
BuckeyeDennis
Platinum Member
Posts: 3814
Joined: Tue Jul 24, 2012 10:03 pm
Location: Central Ohio

Post by BuckeyeDennis »

reible wrote:Well this has been quite a thread. I think I will have to read it a second time and spend sometime thinking on the subject.

I wish I could go and look at mine but it is over 5 pounds so that will have to wait.

I can tell you that I have not experienced the movement issue talked about here. I can also tell you that during some testing I did that table movement can come from places other then the tilt area. When I did my study on the subject (posted here some where) I was looking to stabilize the table and showed the effects of using the tubes connecting to extension tables. The improvements by doing that were major and is the method I use now when trying to get precision cuts. Of course when tilted that is not possible so things change.

Second I can tell you that I have had 3 table tops on one of my machines and simply moved the parts and did an alignment with no effect on table clamping. The first table was from a 510r that I got as a basket case. The PO had spilled something on it and it was only removable by force since it was stuck in the valleys, the top was sanded to remove it from the flats but it still looked like........ The second table was another 510 table and it was in pretty good condition but it had the full set of ribs which I found I did not like all that much. I then replace it with a table that had the smooth center section. The under pinning of the table had the hole issue which I drilled out, it was really hard to align before that so I saw that as a major improvement.

While I understand why one would want to fix this issue I think some of the ideas posted are not correct. I would not expect any "spring" action to be involved in the design except when a spring is used. In fact I would believe the design would be done to minimize that action.

I think the design would have portions would be looked at as a fix portion and section that would be a moving portion. Fix portions that are designed to not move and not expected to do so. This can be thought of a constraint of the design. This can be accomplished by limiting the forces applied to that area. An example would be the lever that is used to apply the force to the moving parts. You will only get just some much force from the tightening, this can be figured out or measured and it will be much less then can start moving the fix portion of the design.

Other design factors such as the area that contact from the fix portion to the moving portion were also factors of the design. That fit would have been designed to allow movement when in one position and to resist movement of those connected parts when tight.

I might even question if the design was to having a cam action, it quite possible that the angle was used to gain area, the sloped parts will have much more area then if they were just flat.

I'll not go farther then this on that subject. At this point I can only guess.

What I would like to see happen is to have this called in to shopsmith. They have the drawing/note books/design intent to look at. Let them have a look at it and see what they have to say. Perhaps them might even want to look at a system that is not working correctly to find out if something is out of tolerance/damaged/or for a potential design flaw. I think they have to potential to solve this issue better then any of us do.

Ed
Ed, I didn't dream up the structural compliance theory out of thin air. Mechanical engineers routinely calculate the deflection under load of simple shapes such as beams and rods, be they made of wood, steel, cast iron, or whatever. For more complex shapes, they use structural finite-element analysis software. The software breaks the complex shape down into a "mesh" of elements with simple shapes, for which solutions are known. Then it solves a whopping big system of simultaneous equations to determine the deformed shape of the object.

Having originally forgotten that "if there's no pic, it didn't happen", here's a pic: :D

[ATTACH]26792[/ATTACH]

Both images show a model of a wrench under load. In the image on the left, the colors indicate the internal stress (i.e. force per unit area) of the material. Given the "modulus of elasticity" (sort of a 3-dimensional spring constant) of the material, then the deformation ratio of each element can be calculated. The image on the right uses color to indicate the total displacement of each element from its original unloaded position (represented by the line drawing).

Here's the site from which I stole that pic. It does a much better job than Wikipedia at explaining the concept, in something close to plain English.


---
Attachments
Wrench FEA.jpg
Wrench FEA.jpg (18.61 KiB) Viewed 2748 times
charlese
Platinum Member
Posts: 7501
Joined: Fri Dec 08, 2006 10:46 pm
Location: Lancaster, CA

Post by charlese »

Thanks, Dennis~ I gotta file this in the great big pile of other things I don't know about:) :cool: :rolleyes:
Octogenarian's have an earned right to be a curmudgeon.
Chuck in Lancaster, CA
User avatar
reible
Platinum Member
Posts: 11283
Joined: Thu Jul 20, 2006 12:08 pm
Location: Aurora, IL

Post by reible »

I think we are on very different tracks here. I am not saying that the design could not have had that effort done but then it is from experience having worked in the field during those times (pre 1985 years for this design we are speaking of). If you are old enough to have been working the field pre 1985 you would be surprised how new these software packages are that are used now. The investment of personally and software back then would have only been done on critical designs.

I worked for a large company with deep pockets and it wasn't until the mid 1990's that we had design packages with even modest tool sets. They had things like checking min and max of tolerances for you on system level designs and added sheet metal functions to do bend radius but they were not anywhere near being able to do finite-element calculations. While I was still working hands on design responsibility before retirement we only talked of having that capacity, while I'm sure it available now it sure was not back then.

This is the perspective I'm coming from. I mean this was the days when a vax11/70 was a big deal...... our computer center was throwing around the word TB of online storage........... for a campus of 8,000 people. We were just phasing out of punch cards machines...... line printed spilled out boxes of paper and ascii graphics ruled the screens and graphics.

Anyway you can feel free to think the design was done anyway you want and I will do the same. Who really knows is shopsmith and again I think this is their problem to solve.


Ed


BuckeyeDennis wrote:Ed, I didn't dream up the structural compliance theory out of thin air. Mechanical engineers routinely calculate the deflection under load of simple shapes such as beams and rods, be they made of wood, steel, cast iron, or whatever. For more complex shapes, they use structural finite-element analysis software. The software breaks the complex shape down into a "mesh" of elements with simple shapes, for which solutions are known. Then it solves a whopping big system of simultaneous equations to determine the deformed shape of the object.

Having originally forgotten that "if there's no pic, it didn't happen", here's a pic: :D

[ATTACH]26792[/ATTACH]

Both images show a model of a wrench under load. In the image on the left, the colors indicate the internal stress (i.e. force per unit area) of the material. Given the "modulus of elasticity" (sort of a 3-dimensional spring constant) of the material, then the deformation ratio of each element can be calculated. The image on the right uses color to indicate the total displacement of each element from its original unloaded position (represented by the line drawing).

Here's the site from which I stole that pic. It does a much better job than Wikipedia at explaining the concept, in something close to plain English.


---
User avatar
BuckeyeDennis
Platinum Member
Posts: 3814
Joined: Tue Jul 24, 2012 10:03 pm
Location: Central Ohio

Post by BuckeyeDennis »

reible wrote:I think we are on very different tracks here. I am not saying that the design could not have had that effort done but then it is from experience having worked in the field during those times (pre 1985 years for this design we are speaking of). If you are old enough to have been working the field pre 1985 you would be surprised how new these software packages are that are used now. The investment of personally and software back then would have only been done on critical designs.

I worked for a large company with deep pockets and it wasn't until the mid 1990's that we had design packages with even modest tool sets. They had things like checking min and max of tolerances for you on system level designs and added sheet metal functions to do bend radius but they were not anywhere near being able to do finite-element calculations. While I was still working hands on design responsibility before retirement we only talked of having that capacity, while I'm sure it available now it sure was not back then.

This is the perspective I'm coming from. I mean this was the days when a vax11/70 was a big deal...... our computer center was throwing around the word TB of online storage........... for a campus of 8,000 people. We were just phasing out of punch cards machines...... line printed spilled out boxes of paper and ascii graphics ruled the screens and graphics.

Anyway you can feel free to think the design was done anyway you want and I will do the same. Who really knows is shopsmith and again I think this is their problem to solve.


Ed
Ah, I did misunderstand what you were saying. I thought that you didn't believe that "rigid" structural members had any significant deflection under load.

No, I highly doubt that Shopsmith used finite-element analysis back in 1985. And yes, I was a practicing engineer back then.

Your mention of the DEC computer sure brought back some memories. I programmed a PDP 11/45 to control a six-legged walking robot back around 1980, and then ported everything to a PDP 11/70 around 1982. Great fun! :)
User avatar
JPG
Platinum Member
Posts: 35600
Joined: Wed Dec 10, 2008 7:42 pm
Location: Lexington, Ky (TAMECAT territory)

Post by JPG »

So, let us think in terms of what was most likely common practice 'back then'.

I have taken the trunion off my 510 table and verified some things I assumed earlier.

The triple pad area is indeed in the same plane as the trunion pivots. The pads/pivots are 14 15/32" apart.

With the trunions slightly further apart than the pivots(creates end play so to speak), the trunions do indeed deflect until the triple pad is reached when the clamps are tightened and 'springs' back when the clamp is loosened. Whether this is steel trunion deformation, or the aluminum table top is moot. Point is there is motion of the trunion and the trunion base does not slip against the table bottom.

The rod has 5/8" of thread on both ends.

I continue to think my earlier conjecture is accurate.

As for SS having this problem, I think the more common absence of the problem relieves them of having a design 'issue'. There are a handful of folks experiencing this. Which of you can do without your table as SS determines why it behaves as it does?

Dusty has shown that his rear trunion does not bear against the triple pad pad when the clamp is tightened. That to me explains why the table slips. Question is, why the gap???
╔═══╗
╟JPG ╢
╚═══╝

Goldie(Bought New SN 377425)/4" jointer/6" beltsander/12" planer/stripsander/bandsaw/powerstation /Scroll saw/Jig saw /Craftsman 10" ras/Craftsman 6" thicknessplaner/ Dayton10"tablesaw(restoredfromneighborstrashpile)/ Mark VII restoration in 'progress'/ 10
E[/size](SN E3779) restoration in progress, a 510 on the back burner and a growing pile of items to be eventually returned to useful life. - aka Red Grange
User avatar
dusty
Platinum Member
Posts: 21530
Joined: Wed Nov 22, 2006 6:52 am
Location: Tucson (Wildcat Country), Arizona

Post by dusty »

JPG wrote:So, let us think in terms of what was most likely common practice 'back then'.

I have taken the trunion off my 510 table and verified some things I assumed earlier.

The triple pad area is indeed in the same plane as the trunion pivots. The pads/pivots are 14 15/32" apart.

With the trunions slightly further apart than the pivots(creates end play so to speak), the trunions do indeed deflect until the triple pad is reached when the clamps are tightened and 'springs' back when the clamp is loosened. Whether this is steel trunion deformation, or the aluminum table top is moot. Point is there is motion of the trunion and the trunion base does not slip against the table bottom.

The rod has 5/8" of thread on both ends.

I continue to think my earlier conjecture is accurate.

As for SS having this problem, I think the more common absence of the problem relieves them of having a design 'issue'. There are a handful of folks experiencing this. Which of you can do without your table as SS determines why it behaves as it does?

Dusty has shown that his rear trunion does not bear against the triple pad pad when the clamp is tightened. That to me explains why the table slips. Question is, why the gap???

I have played with this for about four days now and I am convinced that what I experienced was/is a fluke. A self induced fluke maybe but none the less a fluke.

I can reproduce the failure mode but in normal operation the conditions required to do so are not likely to exist.

When the trunnion bolts are loose, if the trunnions are pulled apart (by whatever force) and the tilt lock is then applied, the front brake will secure itself on the trunnion while the rear brake will close some but will not fully lock (thus the illustrated gap). It is my conjecture that the gap is the sole cause of the tilt lock slipping.

On the bench, pushing the trunnions inward tight against both spindles before applying the tilt lock eliminates the propensity for slippage of the tilt lock. The gap is gone!

At this point, securing the trunnions to the table and performing a table alignment is all that remains. Once that is done, I believe that the slippage problem is eliminated and can not reoccur unless the trunnion bolts are loosened and the table tilt lock is released both at the same time. Only under this condition will the trunnions be able to move apart.

I am currently performing some verification tests but I am satisfied that this issue is resolved.

I have a floating table mounted to the main table and extended as far from the main table as the extension tubes allow. This simulates the typical load condition for the table tilt lock.
"Making Sawdust Safely"
Dusty
Sent from my Dell XPS using Firefox.
Post Reply