Need help in re-designing rotating joint 1

[dkmulford]

Hello everyone. I am a civil engineer by education, however, I have a side business that manfactures amateur and professional camera mounts.

My professional camera mount, called the Pro-Mount, is used by professional videographers to videotape from within trees (i.e., whitetail deer and bear hunts for the Outdoor Channel, Sportsman Channel, ESPN Outdoors, etc.).

My Pro-Mount is able to level vertically and horizontally. I am trying to redesign the "shoulder joint" which provides the horizontal leveling. Please see the pictures below.

Currently, the shoulder joint is two 3 inch diameter plates. The front plate has two slots with two 3/8" hex bolts passing through the slots and into keenserts located in the back plate. There is a stainless steel 1/2" diameter shaft extending through the center of both plates.

My problem is that when the arms are fully extended 90 degrees to the tree (and in line with the rotation of the shoulder joint), the joint can slip. Obviously slipping is dependent upon the force applied to the two bolts and the load applied to the end of the arm. Altough the TV type cameras weigh about 30 to 40 pounds, I would like the arm to be able to handle about 80 lbs without slipping. Additionally, I would like to incorporate a handle in place of the allen bolts so the user doesn't have to carry an allen wrench. Doing this reduces the amount of force the user can place on the bolts.

Is there a book out there that might help us design this joint to maximize the clamping effort?

By the way, we can't increase the diameter of the two plates and the two materials must be aluminum. Additionally, the joint must provide enough rotation to allow leveling (say minimum of 15 degree each direction).

Thanks for your help,
Darren

 

[dkmulford]

btrueblood (Mechanical) Jul 15, 2004
How will making the bolts larger in diameter help?

Simply that you can support a higher clamp load with a bigger bolt. Of course, this takes up room on your 3" plate, and the torque required to generate clamp load goes up with increasing diameter.

OK, then i don't think increasing the bolt size is going to matter much to us since the applied load is governed by what the person can apply to the quick adjust handles or with an L-key allen wrench. The 3/8" bolts we are using now have been able to handle these loads.

In fact, I was wanting to go down to a 5/16" so I could have more room to work. As long as the 5/16" bolt can handle the load then I should be ok, right?

 

[CMcF]

Would it be to expensive to just not clamp the joint. Attach a reasonably long lever to the "moving" half of the joint. The free end of the lever could be anchored to the tree with a seconed lighter strap.

 

[MadMango]

I think ergonomics would fly out the window if you had to wrestle with another arm or device that you had to strap down.

I know you mentioned a desire to test a new prototype this weekend. Please keep us updated [blue]dkmulford[/blue].

Ray Reynolds
"Computers in the future may weigh no more than 1.5 tons."
Popular Mechanics, forecasting the relentless march of science, 1949
Have you read faq731-376 to make the best use of Eng-Tips Forums?

 

[Barry1961]

You can't move the joint out to end of arm and you don't want to move it back to the "tree plate" and teeth probably won't work. Hummmm... Well the main things that I think of for clamping torque is more lever, more pressure (bolts), more area and a shallow wedge.

Instead of having the pivot in the middle of the disk can you move it to the top of the disk? This would give you a longer lever. If you moved the horz. pivot sleeve up you might have room for 3 bolts at 4,6 and 8 o'clock.

I think a larger bolt may help since in aluminum it is the yielding of the aluminum that limits the clamping force, if you are coming up against that limit. Also make sure your washer is not sinking in.

A shallow taper would clamp great but want to stick. Also, I am not sure of the longevity of a taper in aluminum.

Great problem in a small box!!! Good Luck!

Barry1961, Titular Despot Emeritus

 

[CMcF]

If you were to attach a spirit level to your mechanism, and advise your customers that angular adustment is best done with the adjustable support, the flexbility you would need from the coupling would be much reduced if not eliminated.

 

[sreid]

Another though. You could try under cutting one of the rotating plates so that the clamping force is guaranteed to be at the OD. Maybe just a 1/8" land.

 

[dkmulford]

The latest news is that we did not do a prototype of the t-slot/wedge design like I show above. We started to lay it out and we immediately saw some areas where the aluminum would be thin where it needed to be strong. This was the case even when going up to a 3.25 diameter from a 3.0.

After this disappointment, we decided to take another look at the OD clamping design that we had previously made (see below). I made the clamp out of round stock instead of square and designed a slot and pin to limit the rotation. See below. The slot and pin are not shown. When going to round clamp, we had to counter sink the quick adjust levers (also not shown) so they had a flat surface to clamp against. There was barely enough space for the steel keenserts.

We ensured that there was enough space to sufficiently clamp the two halves together (without the halves touching) and we also ensured we tapered the sharp ends around the od of the male piece (where the two 45 degree angles meet) and at the front of the female piece. During our test, we used the quick adjust levers.

During our first test of this prototype, the joint slipped with little weight applied to it. This was VERY upsetting. We decided to use a red magic marker to paint the internal components to see where the parts rubbed and it was deterined that we had good contact around the entire parts. After this we decided to rough up the surfaces by sand blasting everything.

After blasting, it held like a charm. However, the problem we now had was that when we wanted to rotate it (to level it up), we had difficulty getting it to turn even when the bolts were loosened. The pieces were just too rough. In order to rotate the joint, you had to use the arms for leverage. Although this might have been acceptable, we decided we didn't like the feel of things.

We have to get an order done THIS WEEKEND for 15 of these. In order to get this order out the door, we are going to go back to the old design and rough up the surfaces by sandblasting. As long as the customer uses a long allen wrench to clamp the two plates together, he can provide enough leverage to prevent the plates from slipping. Going from a 3.0" diameter to 3.25" should help also.

After this order is out the door, I'm going to further investigate the worm gear. This was the ideal design that we identified long ago, however, we knew it would add cost and weight to the design. I'm guessing I'll need a 1.5" to 2" diameter main gear in order to get it and it's worm into a 3.25" diam. housing. I'm not sure if the teeth on a worm this small will hold up to the force applied with a 240 ft-lb moment (80 lbs on a 3 foot arm)? I guess I'll tackle that problem after this order.

Darren

 

[jomor]

Just as thought. If the top square tube were a Hollow Split Round tube and a Male Round Peice(with lever arm attached) could be inserted, rotated, and clamped down. A pin and slot would allow for stops and keep it from falling out. The level adjustment/tierod could still be attached to the bottom of the split tube. With a little planning they could nest/fold for transporting. Should be low cost & simple. Holding Contact Surface Area would be a function of Length and OD(right now you can only vary face Diameter)


Again, just a thought.

Jomor

 

[ccw]

Hi dkmulford,
Revisit knurling (actually a form thereof on flat surfaces according to ANSI std). Knurl one side of the joint or one surface only. I never met a machinist that did not like to knurl things. Have it hard coated (a deep anodize that puts a hard garnett casing of a few thousandths into the surface, not decorative anodizing). One would probably redesign the part with the slotted holes so no welding is required (a machined block, with a cross drilled hole for the vertical axle, unlike what you have now). After hard coating the part can only be ground, no more machining or welding. Now you have hundreds of super hard little pyramids biting into the opposite soft surface, giving ample friction. Make sure there is a close slip fit with the 1/2 dia. center shaft, so it takes all the radial and most of the bending load across the joint.

An alternative would be to add a piece: a 3" dia. flat washer form 1/16 to 1/8" thick, with 1/2" clearance center hole and two 3/8" clearance holes that inserts between the existing joint. The washer would be knurled on both sides. It could be made of steel or hard coated aluminum.

At 240 ft.lbs. I would be concerned with structural integrity throughout, including the tree. It may see that much now periodically in the field as people plop down their heavy cameras on the mount. However, if you rate it for 80 lbs weight on the end, it will probably see more than that from the handling. Next, you will have people building articulated tree stands out of two of them.

 

[notnats]

I have just skimmed through the above posts and I don't think anybody suggested a multi-disc clutch type of arrangement.

2 rotating plates sandwiched between 3 stationary plates will give 4x the holding torque. Of course the locating pins on the plates will reduce some of the effective area.

Another simple thing would be to relieve the inside of your friction faces so that all your clamping force is applied to the largest diameter possible.

Jeff

 

[Lasker]

I didn't caculated the required torque in your case but ,i had a good experience with "notnats" kind of solution.
We used 10 discs (10 inners and outter)with one screw at the center.
We bought the friction discs (not the whole assembly) from :

http://www.moenninghoff.de/

The moment generated by the preloaded multi discs can be found in "shigley".

Lasker

 

[EdDanzer]

Multiple disks will not provide greater holding force with the same coefficient of friction and pressure, just better heat transfer.
Torque is what you are trying to hold, a larger clamping diameter will give the most for the least, and, or try a fine thread screw for higher clamping force.

 

[notnats]

EdDanzer
I beg to differ. Multiple discs give multiple friction faces. Each face has the same force applied to it. Each face has the same coefficient of friction, so the torque is a direct function of the number of friction faces.
If anything, heat dissipation is worse because the heat must travel through the entire disc stack.

Jeff

 

[EdDanzer]

notnats
We designed a brake to hold 50,000 ft lbs torque a few years ago and found this formula:
F = µ * N F is force, µ is the coefficient of friction, and N is normal force. The force is independent of the pressure per unit area.
If a brake or clutch is going to slip then the area of contact becomes another problem. The surfaces become a bearing surface under load. At this point in design, the area subjected to friction becomes important and it will need to large enough for adequate life.
In the case of this product, it should not be under load when being rotated.

 

[Maui]

dkmulford, there are many good suggestions in the posts on this thread. Perhaps you would consider one more. Are you familiar with a belt driven metal lathe that uses collets to hold the workpiece instead of an a keyed headstock? I have made collet assemblies from aluminum alloys, and they do a remarkably good job in the right application. Could you use that type of design approach to solve this problem? Imagine a knurled 3" diameter outer sleeve that you can use to tighten the joint by hand. The inside of this sleeve is threaded, and there is a 2" diameter central pin extending through the center of this side of the joint and coaxial with the sleeve. The sleeve contains a taper at the contact point for the "collet" on the mating part. You could lenghten the mating part that contains the pivot tube to an appropriate amount, and thread it to accept the threads for the sleeve. Extending out from this threaded section would be a series of concentric slots (the "fingers" of the collet)and an appropriate taper to match that of the knurled sleeve. As you tighten the sleeve, the fingers are brought into contact with the 2" diameter central pin and the joint is secure. If you need to rotate it, loosen the sleeve by backing off on the threads, the collect loses contact with the pin, and it will freely rotate. Tighten it again, and you are good to go. And you can experiment with the surface roughness of the contact points to get the kind of grip that you need. I hope that this description makes sense.


Maui

 

[notnats]

EdDanzer,
The basic equation for torque in a brake or clutch is:

1) T = F * R = µ * N * R where R is the effective friction radius.

Now, in a clutch such as an automotive clutch, the driven plate is sandwiched between the flywheel and the spring loaded driving plate. There is friction material on both sides of the driven plate, both of which are loaded by the driving plate, so the equation becomes:

2) T = n * µ * N * R where n is the number of friction faces, in this case 2, making the equation:

3) T = 2 * µ * N * R

If there are multiple driven plates on a common splined output shaft sandwiched between multiple driving plates driven from a common housing, all loaded by the one set of springs, equation 2) applies.

Multi disc clutches are commonly used to transmit large torque loads in a confined space. Because heat dissipation is a problem they are often enclosed in an oil bath. The resulting reduced value of µ is more than compensated for by the greater torque capacity.

You might have been able to greatly reduce the size of your 50,000 ft lb brake with a multi disc design, although it might have caught fire if it ran too long.

Jeff

 

[EdDanzer]

What book did you find that calculation in? The 50,000 ft lbs brake is a multiple disk and runs at 2,000 psi. The formula I posted is what was we used to determine pressure at a given coefficient of friction, this is the same in 4 different books. We used aluminum bronze against 4150 with oil fill. This brake package is 10” dia., and 5” long. Some of the design problems were transmitting the torque from the plates to the housing. We tested the torque and were with in 5% of calculated and has worked very well. What I will stand by is the number of disks has very little effect on torque, only µ and pressure.

 

[unclesyd]

In line with the post by Barry1961

I think you could use a toggle or over center coupling/clamp to grab the outside edges of the mating surfaces. You might put a little taper on each half to made the clamp much more effective.
There are several types used in vacuum service and can be found in any alloy.
I have an SS one for 3" SS pipe that uses a wing nut instead of a toggle. This one clamps with a rubber friction ring inside the metal clamping part that grabs the pipe, it will slip. Once clamped you will collapse a Sch 10 pipe before it slips. It only has a part number and I have no idea who made it.

This is just one type
Take a look at the toggle coupling in the middle of the page

http://www.intercast.au.com/cast_standards.php

Worst comes to worst you could make one. There all types of things you could do with this clamp.
Taper the edges of the mating parts.
Make a taper fit male /female of the two components.

 

[tracetrimble]

I will second a couple ideas here, knurling and hardening one face, or using a sanitary Tri-Clover style clamp around the outside. The clamp would be a lot less trouble to try first.

 

[tracetrimble]

Also, tapering the OD of the round plates to match the clamp would be a very good idea.