Purely mechanical system to give a continuously variable gear ratio 3

[Nemesis4]

I would have thought this subject would be a FAQ - but apparently not. I have always wondered if an "infinitely variable gear" is possible. I am not referring to any of the currently commercially available CVTs etc. but a constantly engaged, positive drive, all teeth and chains (or whatever) system. I far as I can tell this has been an unsolved problem for at least 500 years. I have found personally that most engineers (in the transmission business) are reluctant to speculate at all - I think there is more than a touch of "pseudo science" and "wild-eyed inventors" about the subject. What is the general opinion on the subject?
I notice that currently the Mechanical Engineering department at Brigham Young University is making a major effort in this area. I think you would have to regard BYU as distinctly non "wild-eyed". There have been successive MSc. theses from Brian Andersen, Ryan Dalling and Levi Haupt (all these theses can be read on the internet).
This is all very high-powered academic stuff and they seem to think that they have solved the age-old variable gear problem. Needless to say, none of it makes a lot of sense to me and I remain unconvinced. I would also like opinions on these three theses. Thank you in advance.

 

[MikeHalloran]

I recall a positive- drive infinitely variable bicycle transmission, announced maybe 20 years ago in the mechanical fish wrappers. I think it used radially sliding teeth on the crank sprocket, and a regular chain and hub.
I don't recall ever seeing one in person.


If you want us to download and read something, at least do us the courtesy of providng a link.


Mike Halloran
Pembroke Pines, FL, USA

 

[patprimmer]

Why would it be an FAQ when it rarely comes up.

You can write up and post an FAQ yourself if you think it is warranted and worth your time.

Mike is correct about the links.

This subject has been discussed in detail several times before. A google site search might find it.

As far as I am aware VRTs have been around alsost as long as the wheel. The problems are bulk and weight vs load carrying capacity.

Belt drive over opposing cones, hydraulic drives like a torque converter, variable pitch propellers or vanes, wheel against the side of a disk are all methods I am aware of off the top of my head. Even a typical automotive friction clutch is a variable ratio drive, all be it.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[Nemesis4]

Sorry about the lack of links. This is a very specialised subject and I was hoping for some really specialised comment. We are considering a similar academic exercise to that of BYU. What comment I get may influence whether the study goes ahead or not.

http://contentdm.lib.byu.edu/ETD/image/etd1878.pdf

http://contentdm.lib.byu.edu/cgi-bin/showfile.exe?CISOROOT=/ETD&CISOPTR=1683&filename=etd2391.pdf

http://contentdm.lib.byu.edu/ETD/image/etd2635.pdf

 

[Nemesis4]

"Mechanical fish wrappers" - ??

 

[patprimmer]

I think fish wrapper equals news paper. Maybe you need to be born pre 1960 to realise.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[Peter7307]

Nemisi4,
"fishwrapper" does indeed refer to newspapers usually tabloids.
The name derives from the using of newspapers by fish sellers to wrap their products.

Pete.

 

[Peter7307]

Nemisis4,
Napier built an aircraft engine using a series of tapered discs as a variable ratio drive for the supercharger.
The engine was either the Sabre or Nomad if I recall correctly.

This may provide some leads.

Pete.

 

[GregLocock]

How serious are the stipulations positive drive, and purely mechanical?

Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[Nemesis4]

GregLocock - these are very serious stipulations. I am only interested in mechanisms of this type. Despite many claims to the contrary I don't think it can be done - but it is very difficult to prove a negative etc. Despite many years of interest in the subject I have never seen anything even approaching a solution. It is one of those things that is either done or is not done - there are no "sort-of" situations. If it can't be done, I would like to be able to define just why it can't be done. I found one statement on the internet that said that gears are by their nature "digital" and a gear varying system of infinitely small steps in ratio change must be "analogue" - so the two are incompatible. I see the point of this but human design and invention have ways of sidestepping such things. Any system involving friction or hydraulics introduces an "analogue" factor into the system - so I am talking about systems that do not have "analogue" features.

 

[ivymike]

How "continuous" do you want your gear ratios? A computer monitor uses discrete points of flashing light to make a continuous-looking image.

 

[ivymike]

here's a suggestion for a continuously variable, purely mechanical device, straight off the top of my head:

input shaft turns the crank portion of a crank-slider mechanism. Crank throw (radius) is mechanically adjustable to give variable stroke. The slider is a moving rack which drives a pair of fixed pinions, which through ratcheting mechanisms and a spring+flyweel converts the back-and-forth motion into relatively smooth unidirectional rotation.

 

[ivymike]

should add: elegant? no. proves that it's not a theoretical impossibility? yes.

you can come up with countless other concepts, if you give it a try. There's the chain-over-conical-sprocket idea, maybe some toroidal gear pairs with variable interface angle, etc... who knows

 

[BrianPetersen]

I think even ivymike's harebrained suggestion will be found to operate in discrete steps dependent on the pitch of the ratchet involved.

 

[strokersix]

Ivymike's suggestion sounds like a Zeromax to me.

http://www.zero-max.com/products/drives/drivesmain.asp

 

[ivymike]

it doesn't have to give discrete steps. you can use more than one slider, for example, to keep the velocity positive. in that instance you'd have an oscillating non-zero output speed.

 

[Nemesis4]

Ivymike - If you use a uni-directional roller clutch you can achieve infinitely small steps. Oscillating systems that are "rectified" into a variable uni-directional output are one of the main groups of suggested variable gear mechanisms - there are thousands of examples in the world's patent files.

MikeHalloran - after reading the various comments, presumably "mechanical fish wrapper" would be a magazine of the "Popular Mechanics" variety (you could only wrap very small fish with PM's page size). I think this comment demonstrates what I wrote in my original statement. Mike is presumably a typical conservative engineer - his immediate reaction seems to be that speculation on subjects like variable gear systems are not worthy of discussion in "proper" engineering publications. And to some extent I don't disagree with this view - there are an awful lot of ridiculous mechanisms proposed by people trying to promote their pet ideas. I think you would discover that the bicycle mechanism that MH mentions would have some fatal "flaw" in its basic idea. Bicycle transmissions are one of the main "hotbeds" of variable gear design. Because they transmit such low amounts of power they can get away with a bit of dodgey design.

Any comments on the BYU papers?

 

[BrianPetersen]

All of the "rectifying" mechanisms that I've ever seen discussed, are incapable of dealing with back-torque (i.e. in an automotive application, there is no possibility of using engine braking).

 

[ivymike]

hmmm. the phrase "tough shit" comes to mind. did marketing ask for engine braking?

 

[BrianPetersen]

No, but the motor vehicle safety standards do.