I'm looking for a way to build real world one-way latches that will hold a swinging pendulum at the end of its swing just as it starts to reverse its swing direction. This would be like a one-way clutch or ratchet that slips or turns easily in a first direction and binds up or grabs in a second reverse direction. It could be at the pendulum pivot or out on the end of the pendulum.
The latch must release when the wheel has rotated around to about a certain angle. The release mechanism should be actuated by gravity in some way, while not being affected by centrifugal forces. That is difficult to accomplish.
Each pendulum needs two of these latches, one for each swing direction. Once they are engaged they need to hold a few pounds of force and then require very little force to release them.
If anyone has any ideas that might get me thinking in the right direction then I'd appreciate the help.
As ingenious as your animations are.....I'm almost surprised you ask....
I should think that your upright wheel support would provide a stable platform from which to negotiate the activation of a latch paddle as it comes by.....
Peace,
TS
As most of humanity suffers under tyrants, misled by the devil and his cohorts who've recently been thrown down here, nothing short of Yahshua, King of Kings, will remove these oppressors and bring everlasting peace.
On one design I used a 3/8 x 3/8 x 4-3/4 inch long steel lever with ratchet teeth. It was spring loaded and had a roller in a slot at the end. The roller would hit a cam that would release the latch. The problem I had was when the pendulum came flying in at a high speed and hit the first teeth then the lever would swing out real fast like. It would then swing in and bounce off the pawl. The end result was many times the pendulum would swing right back out without latching. The tooth spacing was 0.080 inch so it lost some reverse travel each time. I also found that with the mating tooth surfaces being square then the pressure alone would unlatch it. The teeth need a little hook like angle to pull the pawl into the teeth when pressure is applied. The parts were not hardened and showed considerable wear after very little use. This type of design has problems!
I've thought of using a vee-belt in a pulley. The belt can be levered tight to hold or loosened to slip. I'm just searching for ideas. I usually think of complex ideas first and then latter find the simple ones.
Latches are a headache! But, here's something that came to my mind after reading your post.
The idea is that one uses a small secondary pendulum (which I refer to as a ballast rod in the sketch) to catch a pendulum as it swings to one side. As the wheel carrying the pendulum continues to rotate CW, at some point the ballast rod assumes a horizontal position and its ballast weight then forces it to disengages from the pendulum weight so that the pendulum can swing back to a vertical position. Note the small tooth on the pendulum weight.
Hope this helps...
ken
Attachments
On 7/6/06, I found, in any overbalanced gravity wheel with rotation rate, ω, axle to CG distance d, and CG dip angle φ, the average vertical velocity of its drive weights is downward and given by:
Spring loaded pin & socket activated by weighted lever or rolling weight that depresses the release mech. These latches require precision placement & are not very forgiving of bounce or poor alignment.
Variations include pin engaging a cross bar mounted on a ladder type arrangement. Advantages are that it is more tolerant of alignment & slop in the build & the 'latching bar' can be fitted with roller bushes for near frictionless activation therefore small activation force required.
I guess you've considered various types of gate latches ?
If willing to dispense with true gravity activated mechanical catches & latches I would consider using a solenoid activated electromagnetic latch. Far simpler to arrange & timing can easily be adjusted. One major advantage is that bounce or miss latching is less likely to occur.
N.B. If you have a self turning design then it can self charge the battery to activate the solenoid so no net loss of energy :)
Thanks for the ideas. The problem I'm having is that if the pendulum swings half way and starts to reverse then the latch must engage and hold at that point. If the pendulum makes a more complete swing before reversing then tha latch must engage at that different point. The point at which the latch must engage is not fixed. It is determined by how far the pendulum happens to swing. It is determined by the pendulum's reversing direction. The latch point is not fixed. It varies.
I can buy one-way clutch roller bearings which would work perfect except they don't release. Maybe I can come up with a combination clutch bearing inside an on/off latch? The on/off latch would lock and the pendulum would be free to swing only one-way. Then the on/off would release for the reverse swing. The on/off latch could be pin and socket or whatever works. It can be engaged in the middle of the pendulum's swing. The clutch bearing will handle the one-way latching that is needed at the end of the swing.
Jim-Mich; The use of a sprag- clutch such as used in the converter of a 1955 or /and a1956 Buick Dynaflow converter. The necessary inner and outer race is also available. A junkyard is a good source for a test model. jim kelly
Sprag clutches from auto trannys would be way too big and heavy for my prototype use. I can get small clutch roller bearings that fit 1/4 to 5/8 dia shafting for about $6.40 to $7.00 each from McMaster-Carr. If I use eight weights then I need 16 clutch bearings, about $110. I'm sure the junk yard would not part with 16 old auto transmissions for that price.
Here is a quicky I just dreamed up. With discarded banding strap and a couple of $0.79 springs a piece of all thread, two "L" brackets for mounting. and your pendulums.
Not knowing when and where of the position you want to catch and release the secondary pendula would have to be mounted accordingly.
Ralph
Attachments
very cheap catch and release device
Last edited by rlortie on Sun Oct 23, 2005 11:37 pm, edited 2 times in total.
It seems that you require the flexibility of variable catch positions. I assume that the optimal release position (vertical height of release) is always the same.
What about a rim gear (saw tooth) arrangement i.e. a series of close together & short one-way levers located on the inside surface of the rim. As the pendulum passes over them in sequence it lays them flat (counter balanced for min frictional losses) but they act as a one-way clutch pawl.
The rim levers can then be spring loaded (or cam) to release when a certain pressure of weight from the pendulum overcomes them. The advantages of this setup is that the pendulum is caught at its highest point b4 it can exert its downward pressure causing release.
Alternatively an escape mechanism as used in clocks with gravity activated release to disengage the escape mech.
I think I got a usable solution. It requires four small inexpensive 3/8 inch ID bearings, one inexpensive clutch bearing, one 3/8 inch shaft, one custom cog toothed wheel (maybe a standard gear), one custom catch pawl operated by gravity, and one simple custom bearing housing that the pendulum rod connects to.
The shaft rotates on bearings, one at each end fastened to the wheel sides. Fixed to the shaft is a cog. A gravity actuated pawl pivots on a pin attached to one side of the wheel. A bearing housing has a hole bored through for three press fit bearings. The middle bearing is a one way clutch. These bearings press onto the shaft. The pendulum rod attaches to the outside of this housing.
It seems precision enough to eliminate most friction yet cheap enough to prototype.
Thinking about this further, what about a sort of clutch pressure plate. One half has nodes & the other indentations. They are held together by adjustable pressure springs but one half face is able to rotate around the other. Then use a weighted lever operated cam to disengage the plates similar to a car clutch.
On the subject of latches; can a simple one-way rotation, other way hold joint (pivot, motor connection, etc...) be made easily in the Working Model demo version? Sorry if covered before!
Disclaimer: I reserve the right not to know what I'm talking about and not to mention this possibility in my posts. This disclaimer also applies to sentences I claim are quotes from anybody, including me.
Not only a simple gravity latch but demonstrates one reason for curved boards reportedly in Bessler wheel.
Hope someome finds this of value.
Attachments
gravity latch static test before and after
Looking for fellow inventors who want to share and go public (without patents/selling) to jointly develop a couple different bessler wheels :)
303.921.1554 cell anytime
