Gearify

Software helpfull to Gear makers.
Post Reply
User avatar
Mooselake
Old Timer
Posts: 522
Joined: Sun Dec 26, 2010 12:21 pm
Location: Mooselake Manor

Gearify

Post by Mooselake »

Gearify is aimed at non-circular gears, and especially non-circular planetary gears that I've been casually watching for the last year or so.  It was written by a grad student, no updates on YouTube or his site for 5 or 6 months, so I'm trying to get him to respond and see what it's current status is.  Judging from the tutorial video views it hasn't gotten a lot of attention.

It's got something similar to Gearotic's functional gears (including turning a picture into a gear, like make an Art gearhead), plus a tool to automatically create non-circular planetaries.  Certainly not in Gearotic's class,  but looks like it could create some fun looking toys.  Something else to make with my new K40 laser engraver/cutter...

Anybody here have any experience with it, seen it before, or have any comments on it?

Kirk
User avatar
ArtF
Global Moderator
Global Moderator
Posts: 4586
Joined: Sun Sep 05, 2010 6:14 am
Contact:

Re: Gearify

Post by ArtF »

Hi Kirk:

  Looks like great software for decorative gears. Its gears do not roll though, near as I can tell they rub.
So dont use them for loads at all,they are subtractions of a shape during a rotation, so they dont really follow a pitchline, as a result, they would have a speed wobble equal to the number of teeth per rev. and a wear pattern due to the mesh rub. Or at least thats what I saw last I looked. But as I say, for decorative gears, definitely a nice package.

Art
Nate
Old Timer
Posts: 107
Joined: Fri May 08, 2015 6:11 am

Re: Gearify

Post by Nate »

ArtF wrote:   Looks like great software for decorative gears. Its gears do not roll though, near as I can tell they rub. ...
Do you mean that they act by surface friction rather than normal forces on the faces?  (I think we all know that gears have to slip against each other a little bit to work properly.)  It does look like these gears 'roll' more than they mesh, but I'm not sophisticated enough to tell by looking.
User avatar
ArtF
Global Moderator
Global Moderator
Posts: 4586
Joined: Sun Sep 05, 2010 6:14 am
Contact:

Re: Gearify

Post by ArtF »

Nate:

The theory of gearing relies on the pitchpoint being the tangent to the pitch circle at all times. Youll notice in an involute set the teeth always
contact only on the tangent to the pitchcurve, this means the ratio of the two gears remains a constant at the radius of one vs the radius of the
other. To have this true, you need involute teeth.

  Watch those gears as they "mesh" and youll see the contact point moves from the outside of a tooth to the root of the tooth
during the rotation. Since the motion ratio at any point in time is the instantaneous ratio of the two radii , that radii
is changing by the distance of the top of tooth to the bottom. This causes a harmonic motion equal in frequency to the #teeth times the
speed of the gear. Thats what causes noisy gears. If youve heard silent transmissions when their new, iits because the pitchline contact
tangents are perfect, as they wear they can get noisier as the contact point starts to wobble about the perfect pitch radius.  The greater the distance
delta from pitch circle the greater the harmonic vibration, and the greater the wear. A worn out gear wears out much faster than one
that isnt worn..Its an accelerating effect.
    Those round teeth on a gearify gear have a harmonic equal to the full tooth depth from my look at them. They "seem" to mesh well,
but unless you have an involute profile of one sort or another, you cannot or should not bear a load.. 

  My non-circulars are based on similar mechanics, I too use subtraction for them, but you have to subtract a virtual hobb, you cant just
subtract from the mating gear. And if you do that then the arc length of the pitchline comes into effect and things get a lot more
complex.  Though as I said they are pretty and at the low price of the software, probably a bargain for that type of design.  :) )

Art


  Just my reading of the theory though..
User avatar
ArtF
Global Moderator
Global Moderator
Posts: 4586
Joined: Sun Sep 05, 2010 6:14 am
Contact:

Re: Gearify

Post by ArtF »

Nate:

>>I think we all know that gears have to slip against each other a little bit to work properly.

  Actually,  this statement is the key, gears should roll, not slip, slip is friction, frictions kills and is noisy.. could saved myself
a paragraph and some mental math.. lol

Art
Nate
Old Timer
Posts: 107
Joined: Fri May 08, 2015 6:11 am

Re: Gearify

Post by Nate »

ArtF wrote: >>I think we all know that gears have to slip against each other a little bit to work properly.

  Actually,  this statement is the key, gears should roll, not slip, slip is friction, frictions kills and is noisy.. could saved myself
a paragraph and some mental math.. lol
Involute gears have to flex or slip a little at the contact point.  Consider for example a simple case of 1:1 circular gears in some frame.  The only place where the gears are locally moving with the same velocity is where the pitch circles meet.  At any other point on the pressure line, the idealized contact surfaces of the gears will be moving at different velocities.  (If gears really worked better when they rolled against each other without slipping, then we wouldn't lubcricate them, right?)

The rollers that gearify produces, on the other hand, may well be able to roll against other without slipping, and the reason that they may be unsuitable for transferring torque is that their mechanism relies on friction, rather than on pressure between the contact surfaces.
User avatar
Mooselake
Old Timer
Posts: 522
Joined: Sun Dec 26, 2010 12:21 pm
Location: Mooselake Manor

Re: Gearify

Post by Mooselake »

I finally heard from the developer, who says he's got some updates in the works. 

The 60,000+ pound pile of 8 foot logs in my yard is going to be eating up some free time as it's being prepared for the 4 letter white stuff season, but trying the gearify trial is still high on my list.

Kirk


User avatar
ArtF
Global Moderator
Global Moderator
Posts: 4586
Joined: Sun Sep 05, 2010 6:14 am
Contact:

Re: Gearify

Post by ArtF »

Nate>

>>Involute gears have to flex or slip a little at the contact point.

No. They dont. Let me put it this way, two round gears in a closed frame with perfect involutes and no wear, have no slip, no rub and
the only friction is rolling friction. Slipping is not part of the design. Since the contact point is always in tangent to the pitch circle, it rolls
on its mating surface at all points. Now DO they slip, you bet, we live in an imperfect world.

  Seriously though,I think its a mistake to think two involute gears slip by design..I dont believe they do.. at all or at any point in
the tooth. The whole point in most gear design is to eliminate most if not all slippage, rolling is the way they work. Gearify gears cannot
roll, if they did, their teeth would be involutes. One measure of slipage is the amount difference from involuted shape a tooth is. Thats a
quick measure, the more different from involute the shape, the more slipping there is. This is a generality, but math is math.. 

>>the rollers that gearify produces, on the other hand, may well be able to roll against other without slipping

  Respectfully, I disagree.. :) , the math says no... ( While Im often wrong, the math never is...)


Art
User avatar
ArtF
Global Moderator
Global Moderator
Posts: 4586
Joined: Sun Sep 05, 2010 6:14 am
Contact:

Re: Gearify

Post by ArtF »

Nate:

(Im not arguing the point here, but I do think a discussion on this is valuable to people watching who may want to understand more on this topic,
It is after all defined as "The fundamental law of Gearing" , and as it is the basis of all gearing, I think it bears a discussion, and in such things
I do not pretend to expertise, Im simply explaining what I have learned from experience in the math, so where Im wrong, please let me know..
(I hate carrying wrong information. :) )

First, a quote from the involute gear wiki and repeated in another form by Faydor Litvin ( My god of noncircular gearing who is reponsible
for getting me interested in noncircluar gearing.....)

"Where the line of action crosses the line between the two centres it is called the Pitch Point of the gears, where there is no sliding contact".

  Any gear which does not maintain a line of action, is by definition a gear that slides. Involution though , as you are thinking intuitively, is
not the only shape a gear tooth can have, BUT, its the only shape that allows for each gear to share the shape. Youll notice in epicycloidic
gears the pinion has a differing shape from the wheel. Same with cage gears and such. The shape of the mating gear to any gear is really
defined in the shape of the wheel's tooth by the cycloidic math of the motion of the originating shape. Involution gives each the same shape
and has the benefit of no slippage at any point in their contact. Any shape other than involute should never be run fast, it will wear quickly
as no other shape Im aware of has no slippage.

  Now, as to the instantaneous speed ratio, in involutes as you watch the contact point rise from bottom of tooth to top, youd be
forgiven for thinking this is causing an instaneous shift in speed, .. but it doesnt. This is because the contact point is moving out from center
at the exact same curve as the pitch curve. At any point the ratio is identical. So long as the contact point on any tooth is tangent
to the point where the two pitch circles meet, the ratio is exactly the same ratio between the two circles. It never varies at all. In good gears,
this makes them silent and very smooth. In a polished gear you'd need no more lubrication than a trains wheel, and it would be as quiet.
Measuring vibrations caused by speed variances is a good way to measure how worn a gear is.

  In any other case, the more off tangent you are, the more the ratio varies. So in round teeth, ( watch then carefully in slow
motion) the contact point basically moves around all the curves. So is starts at the root and the contact rises to the top and then back
to the bottom of the next tooth. Note the tangent at each point. they point wildly from straight out( at tip to bottom contact) and then
to straight up when the two teeth meet in the middle. This violates "The fundamental law of gearing" and the line of action varies from
0 to 90 degrees on each tooth. This causes the speed to vary by the ratio differences involved,the deeper the tooth , the more the vibration
in speed. This vibration has to happen, the math demands it, and as you rotate such a gear youd feel it vibrate, the deeper the tooth , the
higher the amplitude of vibration and frictional wear. Now as a disclaimer I can say I have never made such a gear, perhaps when youve made
a set you can let us know if Im right or wrong, Im simply picturing the numbers not the reality.

  Ive been asked about this topic for over  a year now and haven't wanted to comment, I havent wanted someone think Im picking on
a competitor with some technicality, Im really not. I like what the software does, I think its decorative and looks nice and at its price
point is a very good value. Its a clever subtraction method and Ive not added it to Gearotic (for those that have asked) because its done,
the solution exists at a good fair price point, and Id rather spend my time researching and programming things that haven't been done
or aren't offered at a fair price.

  So, thats my take on it, correct me if you feel Im wrong, but let us all know how the gears work, might get a copy myself. :)

Art

JustinO
Old Timer
Posts: 137
Joined: Mon Jul 30, 2012 2:13 pm

Re: Gearify

Post by JustinO »

Hi Art,

My understanding is that for gears that obey the fundamental law of gearing, the pitch point is the only point where there is pure rolling; it is the only place where the two gears are going the same direction at the same velocity. Litvin's explanation of centrodes helped make this clear to me.

Gearify looks like an interesting little narrow utility. If I were interested in gears with "round, square, or triangle teeth" I might buy it.

Thanks Art,
--Justin
User avatar
Mooselake
Old Timer
Posts: 522
Joined: Sun Dec 26, 2010 12:21 pm
Location: Mooselake Manor

Re: Gearify

Post by Mooselake »

Don't forget about Mike's eccentric planetary gear calculator, "Astronomer", which is my main interest.

Been enjoying this discussion; nearly mentioned in the OP that they obviously weren't involute teeth and not suitable for any kind of load, but y'all have that one covered.  All I want to do is make cool looking gadgets that don't really do anything useful - kind of like me these days.

I should contact Mike again and see if he wants to chip in here.

Kirk
User avatar
ArtF
Global Moderator
Global Moderator
Posts: 4586
Joined: Sun Sep 05, 2010 6:14 am
Contact:

Re: Gearify

Post by ArtF »

>>My understanding is that for gears that obey the fundament al law of gearing, the pitch point is the only point where there is pure rolling;

"The angular velocity ratio between two gears of a gearset must remain constant throughout the mesh".

  This is the fundamental law of gearing engraved in stone. :)

  Give this a moment of thought experiment. If an involute tooth slipped on its contact point at any point in its travel , this law by logical inference
must be disobeyed. The speed at the time of slipage has just varied. It must at all times be a constant or the math simply falls apart. This is why
it is soo difficult to involute tooth a noncircular gear, you MUST maintain the pressure angle and tangent to the pitchpoint, or you violate the law.
  As yet, I havent seen any men in black approaching so I suspect Im on the right side of the law.

Truths:

  1) Good Gears dont slip, they roll, on every point of contact.
  2) Taxes never really go down,
  3) You die.

-- you know, prove me wrong and Ill really feel the idiot. :) lol

Art
JustinO
Old Timer
Posts: 137
Joined: Mon Jul 30, 2012 2:13 pm

Re: Gearify

Post by JustinO »

Not my responsibility.
Nate
Old Timer
Posts: 107
Joined: Fri May 08, 2015 6:11 am

Re: Gearify

Post by Nate »

I think that we agree about what's going on but have different a different understanding of what the words mean.

Lets suppose we have a pair of circular involute gears, asin(1/2) degree pressure angle (because it makes the math convenient) both with a pitch radius of 10.  One gear centered at (0,10) rotating counterclockwise at 60 rpm, and one centered at (0,-10) and rotating clockwise at 60 rpm.

The pressure line is at 30 degrees so let's say that two teeth are meshing, momentarily, at the point  (2 , 1)

Now, the distance from the center of rotation of the top gear to the contact point is sqrt(4 + 91) and the distance from the center of rotation on the bottom gear is sqrt(4 + 121) so, at the point of contact the bottom gear is moving faster than point of contact on the top gear.  They're moving at different speeds at the point of contact so they can't be rolling against each other.

Discounting the effect of imperfections, gears behave like idealized rollers that do not slip so the angular velocity ratio of the gears as a whole will stay constant, but the action of the gears - in other words the meshing of the teeth - has to have some slippage to work, even with idealized gears.

As Justin correctly notes, the only point where ideal gears can actually roll against each other is the pitch point.
...If an involute tooth slipped on its contact point at any point in its travel , this law by logical inference
must be disobeyed ...
The idealized sliding is parallel (i.e. tangent) to the instantaneous contact surface, while mechanical action of the gears is perpendicular (i.e. normal) to that surface so the sliding doesn't influence the rate of rotation per se.
Last edited by Nate on Sat Sep 19, 2015 11:35 am, edited 1 time in total.
User avatar
ArtF
Global Moderator
Global Moderator
Posts: 4586
Joined: Sun Sep 05, 2010 6:14 am
Contact:

Re: Gearify

Post by ArtF »

Nate"

Well said. I agree. And in the case of the gears we're discussing Im suspicious of the line of action, I dont believe there is one. More of a constant
rotation. But as I said, Id like to hear of the experience in building a set..

Art
Post Reply

Who is online

Users browsing this forum: No registered users and 4 guests