Constructing Janus, by Dirk

Post your build logs here

Re: Constructing Janus, by Dirk

Postby dirktheeng » Tue Aug 09, 2011 9:43 pm

metalman wrote:Dirk, I have an arm for the monitor on my CNC plasma table. It gets a workout and is very handy. Besides holding an old CRT type monitor, it also has a keyboard shelf below the monitor. Unlike yours, it's made of heavy sheet metal and has counter balance springs. The monitor is a lot heavier than your flat screen but the arm is rock solid. All the joints are held by friction, and the counter balance springs. There is a parallel link to each arm joint to keep the monitor upright. Both the top and base swivel.

I really like the look of your wooden box frame. There is one thing I noticed on your joint design that may cause a problem. The long bolt at the pivot points might need a tube between the two sides to keep the tension even on both sides and eliminate any buckling at the joint. Another option would be to use a short bolt on each side and tension as needed.

Jay


Jay,

Thanks for the encouraging words. I am attempting to address the buckling at the joint by a different means. I have put wooden cross members very close to the pivot points and these are covered by the washers. These should keep the joint from buckling. Here's an example from the shoulder:

Shoulder Assembly.jpg
Shoulder


The other joints only have 1 crossmember like the ones shown here, but the Box serves the purpose on the other side.

You can actually see this in the last photo of my previous post. Do you think that should be sufficient?
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

High friction surface

Postby dirktheeng » Tue Aug 09, 2011 11:49 pm

All,

I did a few calculations to figure out (or at least try to estimate) the clamping force I need to support the monitor. I made the assumption that the pressure exerted outward on the band clamp would be approximately equal all the way around when the band clamp is tightened. In this way, it acts as if it is a hydrostatic pressure. This should be very close to reality as the band clamp is fairly thin and flexible. Realizing this, I could use hoop-stress analysis that we use in pressure vessel calculations. The end of this analysis is that the radially applied force at the interface between the brake and clamp is 2*pi times the tangentially applied clamping force regardless of the size or width of the wheel/clamp combo (keep in mind the stress/pressures change, but the interfacial force relationship remains constant). With this in mind, the only reason you want to have a larger diameter wheel brake is to obtain a lower tortional load at the interface.

With that in mind, it comes down to static friction coefficients. The higher the friction coefficient, the less force I need to support the tortional load created by the monitor.

I went searching for friction coefficients and found a few useful values:

wood-wood dry = 0.25 (however I don't think this accurately represents the surface of laser cut parts... the heat should bring out lignans in the wood and may increase the friction coefficient... further, this is mostly endgrain exposure as it is cut layer of plywood, not long grain, so I don't know how representative this is).

plexiglass-plexiglass = 0.8

rubber-plastic = 1-4

Neglecting the weight of the arm and considering the position with the absolute highest torque (that being the arm completely horizontal), the monitor is about 41.5" away from the pivot, weighing about 10lbs, this creates 415 in-lbs of torque. if the break is 2" in diameter, this means that 415lbs of force must be supported at the interface tangentially. Dividing this force by the coefficient of friction gives the total normal force needed, which can then be divided by 2*pi to give the clamping force needed to generate the normal force. Note, this is the clamping force needed at the interface, if the force is applied on levers farther away from the interface, it is reduced by the ratio of lever arm lengths.

With a friction coefficient of 0.25 (wood to wood), it will require a clamp force of 264lbs. With a COF of 0.8 (acrylic to acrylic), it will require a clamp force of 82.6lbs. With a COF of 2, it will require 33lbs and with a COF of 4, 16.5 lbs.

This is assuming that there is only one clamp for each joint. I can put 2 on each which will half the force needed on each.

As can be seen, it is highly advantageous to get the highest COF possible. This made me think about rubberizing the interface. I found a spray on, non-skid coating called super grip. It lays down a silicone rubber that is very sticky. They say it doesn't attract dirt or dust and remains high friction. I looked at the MSDS and it contains several solvents which are known to be slightly solvating to acrylic. I was thinking about trying to apply an extremely thin coating of this stuff to acrylic. The idea is to dilute it with a solvent like MEK and brush it onto the acrylic surface. The solvent should loosten the suface exposed polymers so they can hopefully intermingle with some fo the rubber compounds. When the solvent evaporates, hopefully I will be left with a somewhat rubberized surface with higher friction properties. Another way may be to lightly sand the brake and clamp in the axial direction to creat ridges that will increase the effective friction of the interface.

I can easily test this out either by buildign the shoulder and upper arm and making various friction surfaces on clamp/brakes or by testing them on incline planes. The coefficient of static friction should be the tangent of the angle at which a static body begins to slide on an incline surface.
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Re: Constructing Janus, by Dirk

Postby metalman » Wed Aug 10, 2011 4:17 am

The cross members in the joint will be a big help. The closer they are to the bolt the better. A small box would be better yet. Thin plywood even only a couple inches long will tend to flex under load. A box section will eliminate that movement. Wood is a great material but it would probably be best to cut a test joint and do a torture test or three.

Jay
metalman
 
Posts: 33
Joined: Tue Aug 31, 2010 3:03 am

Plate clutch design for arm joints

Postby dirktheeng » Fri Aug 12, 2011 10:01 pm

All,

Sorry I haven't been so active here lately. I've been working on a project at work and have recently gotten some excelent results that may lead to several publications and patents. I am very excited about that... That said, I have also been putting a decent ammount of time into thinking about the monitor arm design and how to very quickly and easily freeze the motion and let you move it freely.

I came up with a plate clutch design that will allow me to produce enough force to hold the monitor strait out with as little as 1/4 lb of applied force. This will allow me to use a magnetic coil and small rare earth magnet to actuate the clutch. I can run this off of a small AA batery. When the coil is engaged, the clutch is released and the user is free to move the arm as if it had no resistance. When the magnetic coil turns off, a small return spring and the magnet secure the clutch and it will freeze.

The math is based on a constant pressure assumption for clutch design (rather than a constant wear assumption). This should hold for a very long time as the clutch wear is absolutely minimal. As long as the user doesn't force the movement when the clutch is engaged, this assumption should hold indefinately. If the clutch experiences a lot of motion when engaged, it will not wear eavenly (the outside will wear first), and the contact patches change over time which can reduce the allowable torque transfer. This is why a new clutch in a car can feel grabby when new and can loosen up a bit over time (not the only reason, but a contrbuter).

I calculate that I should be able to use 1/4 lb of force to generate 15lbs of force with a simple lever at the clutch point and that can be used with 15 clutch interfaces on the shoulder, 9 at the elbow, and 2 at teh wrist to fully support the weight with clutch plates made of acrylic. I will use 0.06" acrylic for the plates.

I am working on the drawings now and will post them as soon as I am done.
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Re: Constructing Janus, by Dirk

Postby TLHarrell » Fri Aug 12, 2011 11:44 pm

This is so totally way over the top, over engineered. *grabs popcorn* I can't wait to see this thing done.
40w Full Spectrum Engineering 5th Gen Hobby 20"x12" w/ Rotary Engraver
South San Francisco Bay Area - Sales and Support Representative for Full Spectrum Engineering
408-47-LASER - Skype: whitelightlaser-thomas - Facebook: White Light Laser
TLHarrell
 
Posts: 419
Joined: Thu Jul 28, 2011 11:30 pm
Location: Morgan Hill, CA

Re: Constructing Janus, by Dirk

Postby dirktheeng » Sat Aug 13, 2011 1:28 am

TLHarrell wrote:This is so totally way over the top, over engineered. *grabs popcorn* I can't wait to see this thing done.



Well... that's undoubtedly true... but I am having fun and it will demonstrate what you can do with a laser... there's nothing here that can't easily be made with the laser plus just a few common fasteners from mc-master and a little glue. I don't even think the raw materials will total $50. Plus once it's out there, people can do with it what they want. If you don't want to do anything with it, you don't have to, but others have expressed interest.
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

First preview of the clutch... not finished yet

Postby dirktheeng » Sat Aug 13, 2011 1:46 am

Here's the first preview of the clutch on the shoulder. This gives you a general idea of how it will work. It's really not that complex, but the dimensions and materials were carefully chosen to obtain the right supporting torque.

ShoulderClutch.jpg
Early version of clutch


There ended up being more room in the middle of the clutch (actually 2 separate clutches, one on each side), so I added more plates than I needed for a margin of safety. The mechanism that applies force to the clutch will reside in the middle between the bolts. It will consist of a pair of counter-rotating levers. I don't need the levers to produce much motion, so I have very long folcrums which magnify small forces, which means I only need very little energy to make the system activate.

The highest load on the plastic occurs at the inner bolts. I calculated an interfacial stress of about 2500psi at the worst position. Acrylic compression strength is abot 18000psi, tensile strength is 9600psi, shear is about 9000. unless there is something to concentrate the stress, which there shouldn't in this design, this should not cause the acrylic to crack. The big thing is to get a tight clearance fit about a smooth shaft. However, the iron test will be building the thing. The good news is there is plenty of room to add more plates and spread the load out if needed. The local stresses on the plywood should not exceed 1500psi... information on plywood is harder to find... strengths range from 3500-6500 psi. I found a bunch of informaton on screw holding strength... similarly sized screws to these bolts were able to hold 800+lbs before failure... the load on the bolt shouldn't be mroe than 250lbs spread over about 1 inch thick material. Barring any beter info, I am going to have to go with the comparitive shear tests and say it's safe.

I also changed the pivot method of the axis. This will help spread the load out and provide a bit more friction. There won't be more than 15 lbs on this joint, so it's not that critical. It should be plenty strong enough.

ShoulderPivot1.jpg
NewPivot 1


ShoulderPivot2.jpg
NewPivot 2
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Questions about the Laser tube... Please post

Postby dirktheeng » Sat Aug 13, 2011 1:02 pm

All,

I received the laser yesterday (my wife called me at work and told me that my big dog bone had arrived... I knew exactly what it was after seeing other peoples packaging). It got here in one piece which was good... but they must spend about $30 on packaging that thing plus labor.

Anyhow, when I opened it and inspected it, I had a couple questions:

1) my tube did not come with the 2 short tubing pieces that connects the mirror cooler to the main body. Is that normal?

2) the inner ring (I think the anode) in the back does not appear to be aligned with the tube. It's not blocking anything, but it definitely isn't strait. I don't think that should be a problem, but I wanted to ask the experts here.

Here's a couple pictures:

DSCN4133.JPG
Close up of back end


DSCN4134.JPG
laser tube


unfortunately, I can't fire the laser to test it yet because my 12V power supply has not arrived yet to power the pump. It should be here soon.
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Re: Constructing Janus, by Dirk

Postby metalman » Sat Aug 13, 2011 1:57 pm

Dirk, Your new pivot lock design looks much better. It should eliminate all the problems the band clamp had.

Not sure about the anode ring, but if doesn't interfere with the beam you may be ok. If you want to borrow a 12v power supply to try out the tube, I can lend you one of mine until yours arrives. I have to make a run up your way later today or tomorrow.

Jay
metalman
 
Posts: 33
Joined: Tue Aug 31, 2010 3:03 am

Re: Constructing Janus, by Dirk

Postby dirktheeng » Sat Aug 13, 2011 3:38 pm

Jay,

sounds good. I sent you a pm with my phone number.
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

PreviousNext

Return to Build Logs

Who is online

Users browsing this forum: No registered users and 17 guests

cron