Problem Encountered; need advise

[Hwingo]

Hi John. Thanks for replying. Yep, that looks like the one I was originally trying to duplicate. How did you know to do that?

Harold

[Hwingo]

At this juncture I think it?s prudent to aggressively examine cuts made using Gearotic and critically scrutinize defects and/or deficiencies that are visually present in image-results posted earlier.

Aside from obvious morphological differences between the original gear and the one I produced, I can see at least one deficiency or omissions in the gear I fabricated. I would prefer to first establish *cause* for what I am seeing followed by a means of correcting the issue. Later, either on-line or via PM, I can speak with John S. regarding tooth form. Perhaps it would be beneficial to all that the issue of tooth form (morphology) be discussed on-line as there may be others to benefit from such discussion.

Because I am a beginner, I can see only one thing that quickly ?jumps out? whereas you, the experienced, will see additional things that are not obvious to me. I will begin with that which is obvious to me. I will include a screen shot and several additional images that may be of benefit.

I will need to make a second post, immediately after this post in order that I can include the last diagram and an account of the defect that I found.

Thanks,

Harold

[Hwingo]

Take a look at the diagram included with this post. When viewing the actual image of the fabricated gear (see first image of previous post), I have identified ?ledging? with each tooth. The diagram alludes to ledges. Why has this happened and what must I do to correct this issue?

Harold

[ArtF]

Hi Harold:

  The root ledging is because the tool cannot fit in that spot to do the undercut fillet. A Ball end mill may do a better job as it wouldnt leave as
sharp a contrasted edge. Cutting low tooth count gears is always a challenge due to undercuts , most gear makers recomment no fewer than 12 teeth for that reason, OR , as John pointed out, they give a profile shift to eliinate the undercut, no ledge then exists.

      If you look critically at the 4th axis screen straight inwards, and watcht he tool move, youll see gettign that small rounded fillet
without a ball mill woudl be very difficult.. if not impossible for most gears, so the solution is to increase tooth count, add a profile shift,
( or run a round file on each tooth post production. ).

  On original look at your gears I was of the opinion they were too thin...but Im revising that as it looks more like just a low tooth count, with
low pressure angle. 14.5 is not a good angle for low tooth counts, it accentuates the undercut, makes the tooth weaker and harder to make,
as you see by looking at Johns profile shifted tooth form, same tooth count, but much better root, with much higher strength in each tooth.



Art


Art

[Hwingo]

Good Morning All,

So there seems to be at least one softly suggested message ?.. ?Gear fabrication is not strictly immutable? suggesting there may be a wee bit of ?wiggle-room? in either direction. I suppose this is where the learning curve begins to take form. Since I am a beginner, the natural progression is to ask, ?How does one begin learning how and when to modify shift? in order to make a stronger, working gear??

Is this to say that as long as you maintain the same PA and DP, shifting can be used to change the physical form of a gear and it will mesh quite well with the wheel providing long service?

Harold

[ArtF]

Hi Harold:

  Well said. Tolerances derive from purpose. Any gear is only as good as its match to its purpose.

Profile Shift:

  Consider two gears, a wheel and its pinion. If designed with zero shift, they have equal strength to
their teeth , no matter how many teeth each has except in the case of undercut. Their strength is a
function of their width, and the roots they have, its usually recommended to have no fewer than 12 teeth
generally .. but very generally. Its obvious that this cant be followed much of the time. When undercut
is in a gear, that gears teeth are weakened. Sometimes, depending on purpose, this may not matter.
Higher pressure angles can help remove undercut, but only to a point. After that..

  A profile shift can remove that undercut ( or alleviate it) and add strength. This will change the mesh,
BUT, this too is fixable by shifting the other gears profile as well in the opposite direction, the mesh is then
correct.

    There are limits of course, and they vary from gear pair to gear pair. To understand a profile shift, look at
the noncircular gear page. Its gears are created by rolling a rack around the blank of the gear. A "Profile Shift"
is simply shifting the rack outwards a set amount. On that screen is a rack tooth picture where the head normally is,
as you enter a shift, watch the pitch line on that rack tooth move. Tooth the gear to see the effect. As you shift outwards
the gears tooth gets less rack, so the root trochoids, if there are any, are much more shallow. Stronger teeth result.
BUT, since you moved the rack out for one, you must move it in on the mating gear to keep the pitchline respected.

  As to how mobile the tolerances are and how worried you should be as to tolerance, that too is more a function
of purpose than gear. Look at some of the decorative gears you see on the web, many have triangular teeth with no
regard for pitchline. They function fine. Their purpose is not to drive any load, if they did a user of them would soon find
triangle teeth are natural vibrators, they cause the two gears to vibrate in velocity ( a first derivative vibration), that would
wear them out in short order. But as decorative gears, they are fine. Purpose trumps specification. For most things anyone here
would build a tolerance is allowed of quite a bit before trouble will be seen. If making a transmission gear, you want to be very accurate
indeed, if making a gear for a lathes change gears, you want close, but a bit of backlash isnt going to hurt you.

  Creating backlash with tooth width respects the pitchline, that gear will be slightly weaker in tooth from being a bit narrower,
but will wear very well. At most its pressure angle may be slightly affected. A profile shift moves the pitchline to a new position, but fully respects it so long as the other gear is shifted in the opposing direction. ( Your gear looks shifted, so the wheel is probably already shifted.. )
but theres a lot of wiggle room in all this. In the end the application is so widely varied, thats its near impossible at any point to say
"this will work" or "this wont work" in any general sense. Except in demanding requirments I suspect most do as I do, intuitively look
at a requirment and decided if your being picky enough...or too picky... Its a rare gearhead who hasnt had the experience of making gears
too weak, or too strong for the thing they were building or fixing. Wisdom always comes at a price.  :)

Art

[Hwingo]

Good Morning Art,

Within the body of this Thread, referencing reply #11, I included an image of the original gear which clearly shows its physical form in frontal view.

Following John?s lead regarding shift thus subsequence form (post #14), I have *attempted* to take careful measurements of the original gear, e.g., width at root of teeth, space between two teeth at the roots, and the tiny thin edge that?s present on the OD of the teeth. Though my illustration does not accurately depict the original gear?s tooth-form, measurements are as close as I can get. Refer to illustration.

Is there a way that I can ?come close? to accurately measuring those areas depicted in John?s submission (post #14) with those measurements of the original gear so I can compare the two?

I can only assume that if the spaces between teeth, in John?s suggestion, are less than the space between the original gear then John?s gear will likely not work. For that matter, if the widths of teeth in John?s proposal are greater than the widths of teeth in the original gear, this too would suggest his gear would not articulate with my big wheel. Are these statements correct?

Lastly, is there a way to alter space distance, as well as tooth width so as to make John?s proposal workable? The reason I ask is, it would be nice to replicate form of the original gear as closely as possible. John?s proposal certainly comes very close if going by ?looks?.

Harold 

[ArtF]

Hi Harold:

  As you shift a gear positive, the spaces shrink between the teeth, the theory being the spaces on the other will be enlarged.

What you need is the width at the pitch line on the other gear, no other measurement is real important. If the width on the mating gears pitchline is X, then if yours is <=X your gear will work. If you have the capability to shift the center distance then you wont have any trouble no matter what you do, if you profile shift your gear outwards by .6 as John did, that implies only that you need to shift the center away from the original mate by the same shift if you dont intend to recut that one.

&nbsp; &nbsp; If it were me, Id make the one John suggested with .6 profile shift ( since the DP hasnt changed, it will mesh fine ), and Id slightly shift the new gear away form the original
if I found it too tight. A profile shift is really just pushing the pitchline outwards, ( for a positive shift), so a slight increase in shaft distance will correct it just fine.

&nbsp; In all this , there is always more than one way to skin the cat, and the accuracy, specially in your rotiseri wont impact you much I dont think.

Art



&nbsp;

[ArtF]

Harold:

&nbsp; I see I didnt quite answer what you asked..

Generally, (teeth +2)/Diameter = DP&nbsp; is close enough to work in most cases to determine what you need to make, often its hard to be sure, they may have shifted,
or they may have drawn a gear in cad with no relationship to the math that should be used, when making a mate to an unknown gear it'd be hard to tell
where to measure. If I were really unsure and wanted to be close, why not just print one ( Gearotic will print to exact size) , then paste and cut from cardboard.

&nbsp; Try out a paper one, costs nothing and then youll cut one sure that it will fit fine. Only dnager is paper cuts.. and the use of dangerous scissors.. :)

Art

[Hwingo]

Good Morning to All,

Art, for ?starters?, just a few quick questions:

1. Is the term ?Profile Shift? a term which has meaning exclusive to this program or is Profile Shift common ?lingo? that?s well known throughout the industry? Stated differently, if an individual (such as me) uses the term ?Profile Shift? in the presence of *any* Tool & Die maker being familiar with gear fabrication, will they quickly understand the term ?Profile Shift??

2. When you speak of ?Pitch Line?, are you specifically referring to ?Pitch Circle Diameter??

I think I am getting closer to understanding how to use this program. I have learned that I can actually populate a field with a fractional value, e.g. DP 11.05 or shift .53 or .58. I am not bound to whole numbers. This does make a difference. In testing this I have learned that distance between teeth is altered and can be measured using the "measure function".

Harold

[ArtF]

Hi Harold:

&nbsp; Indeed anyone from a gear making industry would recognise the term "Profile Shift". When I refer to the pitch line I do
mean the pitch circle ( though in my world its not always a circle. :)&nbsp; ) Gears mesh on that pitch circle, right on its line,
which is why I refer to it as a pitchline. The area the pitch line is the addendum. Now given a gear blank, a shaper will
cut teeth into the blank to a depth where the shapers pitchline matches the gears pitchline. This will generate a zero shifted gear.
But if I only cut the shaper into the blank less than to where the pitchlines match, then the gear is said to have a positive profile
shift. (Meaning in english the profile of the tooth is shifted outwards by a set amount. This changes the shape of the tooth )

&nbsp; Use the calculator button as well Harold, it will show you what fractional DP migth best suit you for a given center distance.

Art

[Hwingo]

Good Morning Art.

Profile shift applied to one gear requires (basically) the application of an opposite shift to the opposing gear. Correct? What happens when a shift (perhaps out of necessity) is applied to the gear being fabricated but the opposing gear cannot be easily altered nor easily re-fabricated? Is this where a wee bit of "wiggle factor" may be allowable without severe consequence?

Harold

[ArtF]

Harold:

Yes. Ideally, the two gears are meant to mesh right on the pitch line, what happens if you pull them out a
bit is the pressure angle changes, but they still mesh, albeit a bit more loosely as the tip of the tooth thins.
&nbsp; In most applications Ive come across there is more wiggle room than youd think. Thats why I think
printing a blank to see and judge a fit is a pretty good way to see how good intuitive fits are..

Art

[Hwingo]

Thanks Art. Message received.

I am in the process of doing a comparison between the original gear and the purposed gear. I should have the results fairly soon and I will post for your inspection.

Harold

[Hwingo]

It?s now obvious to me, though it has been obvious to you guys from the beginning, that I am dealing with a non-standard pinion gear. Though the original pinion gear can be defined as non-standard for several reasons, e.g., tooth form and tooth number (and perhaps other reasons such as DP and PA), the designer of this ?system? has, indeed, designed a ?system? that seemingly works.

I have no real way to neither alter the original wheel-gear nor fabricate a new wheel-gear (which is too large) so that it can articulate with a new pinion gear which has been given ?shift?. Conventional logic would be to do my best to replicate the original pinion gear as closely as possible. After all, the original pinion gear, with all its anomalies, has been designed to work with that system. The problem that I am having is, I have no ?gear making wisdom? since wisdom comes from experience. So, I haven?t the education/experience to know if I can have a ?wiggle factor? of 0.001? or one of 0.010?. One important tip that you have given me is, the Pitch Circle diameter is the most important thing to consider. Decision making is very difficult at this stage of my educational development.

Here are knowns from original gear:
1. The original pinion gear is 0.996? OD
2. The gear has 9 teeth
3. The Pitch Circle diameter is N x OD / N + 2 thus:
&nbsp; &nbsp; 9 x 0.996 / 9 + 2 =&nbsp; 0.8149? Pitch Circle diameter
4. Using the formula N + 2 / OD, the Diametral Pitch is 11.0442
5. Measured distance between teeth, at roots of teeth, is 0.100?
6. Measured distance between teeth on circle is 0.13799? (see first image)

Here are knowns from new gear:
1. Recommended OD of purposed pinion gear = 1.0828?
2. The gear will have 9 teeth
3. Pitch Circle dia = 0.8858
4. Diametral Pitch = 11.01
5. Measured distance between teeth, at roots of teeth, is 0.100? using Gearotic measuring tool
6. Measured distance between teeth on circle is 0.167? (see second image)

Here is non-factional information:
1. Using a 14.5 deg gear pitch gage, the *best fit* for the Pinion Gear is the 12 pitch blade.
2. A 20 deg pitch gage does not fit the pinion gear
3. There is result-disparity between the 14.5 deg pitch gage and the mathematical formula.
4. At this stage of my development, I trust the formula more than the pitch gage but there remains plenty room for doubt inside my vacant brain.
5. It appears that the distance between teeth on the pitch circle is greater in the proposed gear than the original gear. I suppose this means that it?s back to the drawing board and come up with a gear whose pitch circle diameter is equal to or less than the original gear. Right?

Harold