Another idea to add to the mix

[mondrasek]

Would it be possible to just use broli's original momentum transfer sim seen here http://www.youtube.com/watch?v=sPLeLxOBcVQ. Add to this sim a free standing flywheel. Start that flywheel spinning first with a motor that turns off shortly thereafter. Now couple that flywheel to broli's device with a timed and geared device to cause it's initial rotation. That part might be tricky, but I think we can actually use friction drives (unless WM2D actually wastes "energy" to heat, but I doubt that). It should be okay if the flywheel still maintains some motion after the broli device is up to speed. Then activate the broli device. Once that has performed it's motion, transfer the momentum back into the original flywheel using another system of time triggered gears/friction drives. Again, it might not be possible to transfer fully (ie completely stop the broli device).

Measuring the speed of the flywheel we can see if it has gained or lost speed.

Would that be a viable test? Has broli posted the .wm2d file for his device?

[DrWhat]

I wonder if Bessler has such a mech somewhere in his MTs? I'll have to have a browse.

[Fletcher]

I'm beginning to wonder what closing the loop actually means, I understood it as success in letting a device start, and having by no external drive, that means a combo of grav driven mechs, retun to start position, with all parts of mech(s) in position it started from. The difficulty if you do that is restarting the device, but I think we've vaguely discussed that p.m (no, not post mortem, private mail). The amount of energy left of that extracted by "fall", "G" or "downstroke" of beam weighted end and any added acceleration or push by parts of the device after ensuring return to start position is "usabale" or "available, or "residual".

I'd love to see somebody manage that with some sort of wheel, but don't see a load of excess or residual left to do any work.

But then I'm a beam fetishist I guess.

Assuming your oscillating device starts from a position of imbalance [with torque] then it will have Potential Energy of position, if it is a gravity driven device - once released the torque will apply & it will loose Pe - in the process of moving Pe can be extracted to do work as Ke - part of that Ke will be used up in work done as ordinary system friction losses, some of that Ke must be harnessed to be used to reset the device to the same Pe state [position] that it started the cycle with - that's the 'Stevin's problem', or closing the loop - any excess energy can do external mechanical work that you might employ it to do & it would therefore be termed mechanically useful.

It does not matter if it can only reset itself & meet the ordinary system losses & not be mechanically useful - you might consider this a mere toy but the world would take very big note of that achievement in itself.

A thing to look out for is the human intervention element i.e. sometimes inventors hand place an arrangement into a certain favourable position i.e. imbalance, max torque etc - usually this is also the highest position the device can physically attain - then they release it - you used chemical energy of your muscle to position it & after release it has to get back there by its own volition - if it ends up lower then not all the Pe is restored & you haven't closed the loop or made it self sustaining & it's winding down.

If your device Nick meets those criteria I'd say you have cracked it !

[nicbordeaux]

I agree with you on all points except the last, it's cracked when you can run a small sanyo gen off it. Then there is no question about whether it's just a kind of pendulum moving mass but producing no energy. Anything less and the debunkers will scream blue murder until the end of time, and some of them are quite clever at "proving" that a working device can't possibly work.

The main one is a self starting device from "any" position. It may be possible, but that's a another problem, so yes, all beam things so far, mine or else, require "inventor" muscle input to move things into the most favourable start position, flip a pendulum, spin a wheel, start a slide, whatever. That's why the device must be able to perform a large number of "cycles" producing visible excess so extreme no possible argument arises as to whether e produced is lower than start input. It also has to be skeletal so that no possible arguments can arise as to hidden power sources, you have to be able to dismantle and reassemble it at will so idiots who can't face the obvious can take it apart and look at every single part. Which means you have to build a different machine to what you have (if you ever have anything), a device which is radically "inferior" to the design you have, because whan you release that "inferior" design, every guy in the world is going to copy it, before you know it a guy in every country in the world has 500 patents on it.

BTW, may I at some point call on your "modelling ("modelization"?) talents to predict a given sequence of mechanical events ? Or at least subject some material which you may judge to be of sufficient or insufficient interest to be worth your time ?

[Fletcher]

Since I can't picture what you are talking about I can't say whether it can be modeled in WM2D but its probably beyond me anyway - some like greg, lust or broli could possibly do it ?

Sorry guys, I don't have the time lately to help you out with the sim, I do not even progress with my own design and I am happy if I can read most of the post on the forum.

This thread is interesting...

I would need a mechanism too which can fully transfer momentum from one type of weight to a flywheel, so one day I have to design some bloody mechanism which is capable to do the trick. At the moment I haven't got it. An automatic gearbox or cvt is a bit complicated, we cannot just throw that inside a wheel. Something more simple mech or a different approach is required for this problem.

Otherwise, Am I got it right?

By transferring momentum from a heavy slow moving weight to a light standing weight, at the end of this process the (now) fast moving light weight will have more KE than the heavy weight had.

So, according to standard (accepted) physics there is really a gain? What is it mean anyway?

Keep up the good work!


P.S.: I hope Jim will solve all this stuff soon, so I do not need to sit back to the drawing board & try to save/destroy the world myself, LOL :D

That's a pity Greg - you once built a useful adjustable pulley system that might have been of use with some tweaking.

Here's a quick pic of a linear representation of momentum transfer theory - it requires that the drive weight fall as far as possible unfettered by 'hard' connection to the next stage, & so on - this would need some sort of clutch arrangement or variable transmission that could be turned on & off when optimal.

[nicbordeaux]

Probably another silly post because the stuff is going to be too heavy... but over here, we have 49.9 cc mopeds in their hundreds of thousands lying abandoned. The "simpler ones have a centrifugal clutch which it's pretty "easy" to adjust as regards kick in rpm by playing with the spings and weights and pads and stuff. The later and "better" ones have variators : the motor moves under rpm (works against springs), and on the primary drive little bits move out as the pulley tension slackens, and make that primary drive larger dia, eg higher gearing. Hit a hill, the rpm drop, the motor moves and "constrains" the little gizmos, thereby giving a lower gearing. Point of motor moving about is that the pulley slack allows the gizmos to move in and out, changing the gearing. Actually, said gizmos are also "modifiable" by interchanging 'em for differently weighted ones, so the "variation" can be controlled too. Or you can remove a few gizmos and that also has an effect, but ya burn out the clutch, which isn't so cool when you still have three generations of revolving credit repayment on the moped.

Complicated in words ?

http://img19.imageshack.us/i/photo1lyj.jpg/

There are other systems with identical "variator" effect where it ain't the motor what moves about, but other things I don't purport to comprehend. Nor want to.

Point being you could lift one of these from an old moped no hassle. Or use the systems as a source of inspiration.

Maybe googling "variateur mobylette" would lead to a load of mumbo jumbo in French, could be "moped automatic transmission" might get you somewhere in English ?

Come to think of it, seem to remeber helicos and washing machines have variable drives, probably centrifugal clutch to ? Forget washing machines, too big. But a scale model helico ?

[Michael]

The reason I mentioned that a perfectly balanced see-saw will find it's own level is because I have built one, and was a little surprised that this is what it does. Prior to actually building, I assumed (as you are now) that you could park it in any position and it wouldn't move. Sure - with sufficient friction in the bearings, you can do that - but it is very clear that there are forces that work that ensure the beam will have a preference to stay level.




The only reason a see-saw wouldn't rest at any position is if it's not perfectly balanced;bottom heavy.
Edit; I see it's already been covered.

[rlortie]

Fletcher,

Love your drawing! Cannot help but notice that if you put the light wheel inside the heavy you have a morphed version of Doc's Pop Keenie wheel.

I hope this goes farther than the original, but instinct says it won't.

http://www.patentstorm.us/patents/5254041.html
http://www.freepatentsonline.com/CCL-474-14.html
http://www.springerlink.com/content/q6u5458m4r531513/
http://books.google.com/books?id=XCEDAA ... ch&f=false

Ralph

[Fletcher]

fetching isn't it ;7) - no, it probably won't coz of the mechanical issues.

EDIT: lol Ralph - well observed [is there such a phrase ?] - seems they not only mechanically look alike but suffer the same problems by trying to use momentum as energy & all that entails, JMO's.

P.S. great links Ralph - thanks for taking the time to find them.

[rlortie]

Typing error corrected and it ain't hte fritrs tiem! Sorry about that, had my mind on collecting the links.

rulpf

[mondrasek]

I believe it is easy enough to simulate the actions of a CVT type momentum transfer in WM2D, just not the true mechanism. If you press two circular elements together they can pass angular velocity between themselves. For example, push two together with springs (maybe proper construction can achieve perfect contact without the springs but I have not tested that yet). Start one spinning (drive wheel) with the momentary use of a motor. The two will interact like a friction drive, but will slip based off of the friction coefficients assigned to those circles. You will see the angular velocity of the drive wheel slow in a perfect linear fashion as the driven wheel is likewise linearly accelerated. At the point where both circles are rotating at the correct speeds for them to not be slipping, both velocities immediately stabilize. Doesn't this model an ideal (lossless) CVT?

Of course this will not work in the real world since the slipping would result in losses as heat. But it should be a simple way to further test broli's original design. Does anyone know if the sim of that momentum transfer device http://www.youtube.com/watch?v=sPLeLxOBcVQ is available? Or broli, could you make it available? It would be nice to start with that perfectly tuned model rather than have to create one. Plus, if this works, others could also play with exactly the same file.

Thanks,

M.

[Fletcher]

Have a go mondrasek - it's worth spending some time on I think - I don't believe broli ever posted up the WM file, just the animation on U-tube - if he doesn't respond you or I can build another one or something similar - I'd do it just to try & get some sort of cvt sorted out, if it can be done - I have a friend who is pretty good with the program who often finds simple work-arounds that only comes from experience - but I wouldn't send him the file until we were stuck so he doesn't waste too much time as he's generally pretty busy & at work.

Just a word of advice, I try to stay clear of polygon shapes for componentry as they sometimes behave erratically - I tend to use pin jointed rectangles - more joints sucks more calculating memory but sometimes avoiding polygons makes a difference especially when pieces collide.

EDIT: I see on the 'energy producing experiments' thread broli is building a real world model to confirm the expected physics - that data can then be fed back into the sim if they are divergent for some reason - that's actually a good approach, IMO.

[Fletcher]

mondrasek .. here's a friction dampener that you might be able to tweak for your purposes & turn it into a friction driver of sorts.

[mondrasek]

Fletcher, yep, it was disappointing that broli chose not to provide his wm2d file even when asked. Especially after he has asked for others to assist and collaborate. That is exactly what I was trying to do. I figured if we could clearly show the device being started up by a flywheel and then returning more speed to that same flywheel (if possible) it would give some of the builders the pupose they do not have yet to build and test some real world models.

It only took a few minutes to sketch up the geometry and find values for the initial rotation and weights that would duplicate the action of broli's video. But things have gone down hill from there. I was able to get one friction drive working so that I could start the system with a flywheel. But as I tried to get the output back to the same flywheel things became very erratic and then just stopped working much at all. I've run into the same problems before in wm2d and always attributed it to having messed around too long in the same file (to many edits). I had always started over by re-imported my geometry (from CAD) and then re-building using the least amount of steps I could. But perhaps your advice to avoid using polygons is what has been the problem all along.

Either way I have run out of patience and time for much more today. I'll give it a go again tomorrow trying again to mimic broli's original video geometry as closely as possible but with primitives. Maybe he will change his mind and provide the original file?

Any worries about importing primitives from CAD vs. creating them in wm2D? I become very frustrated with the lack of CAD tools in wm2d as well as prefer to have my geometry files saved for when I want to update or slightly modify just one or a few parts.

[mondrasek]

Double post.