Best Of The Overland/Wiseman brass Western Maryland Shay thread

[u18b]

I want to thank Max for his offer to tackle this.  This has been a learning adventure for me.  Max even purchased additional pieces to his lathe to tackle the little gear.

His explanation above is clear- and I can’t add to that which is described since I don’t have a lathe.

However, what he has posted above is actually step 2.

For those who might like fuller explanations, let me fill in step 1- since that relates to what you see in step 2 above.

The original problem was that the worm gear hole had become eccentric and was vibrating badly.   The gear spins on a special M1.4 screw that has a smooth upper portion (the threads are on bottom).  So the gear spun on the smooth part of the gear.  You screw the gear in and it stops when the thread bottoms out on the frame.



You absolutely HAD to have that special screw because a regular screw would not work (I tried).  If you used a regular screw, then the spinning gear in one direction would un-screw the screw.  And running the loco in the other direction would cause the screw to get tighter and lock up the loco.

So the suggestion from Max was …. instead of relying on that highly specialized screw, let’s just use an insert.  We will open up the hole in the gear and place an insert in there.  Then any old screw M1.4 would work if the tube-insert were the correct height.

After Max bored the hole out, we agreed there was a problem we did not originally anticipate.  The tube did not max out the space that was a part of the stock gear.

Here is what I mean using the following shots and drawings.

The original stock gear is like a hat and the head of the specialized screw sits INSIDE the gear.  That opening is about ….  .100”

Unfortunately, the tube was smaller that the stock opening in the gear for the screw head.   So when Max bored the smaller hole originally, the gear only made contact about HALFWAY down- the upper portion of the hole had a gap between the gear wall and the tube insert.



We agreed this was a prescription for excessive wear and I would be right back where I was with gear wobble pretty quickly.

But here was the problem.  We did not have any available tubing that went from 1.4 mm inside hole to that which was needed for the current hole.  Max suggested there WAS a tube that went from 1.6 mm- but that would mean I would have to drill out the frame and tap it for a new hole.

And I didn’t want to do that.  It was an option- but I wanted to keep screw sizes consistent at M1.4 mm.

Thus…. The solution was to nest TWO thin tubes.  The first goes from M1.4 hole to .0785”.   The second has an inside diameter hole of about .0785” and goes to an outside dia of .0995”.

Max mailed what he had to me, I purchased the other tubing and gave it all a try.

I then had to report to Max that the gear still had an eccentricity to it.  It was also still a bit noisy.

Well, Max being Max wanted to know exactly what went wrong.   I didn’t think his boring was off, but it was possible.

So I sent everything back to him so he could double check his previous work.  Max confirmed his boring was excellent (no surprise) - but the problem was the tubes and the slop they introduced.

So I learned a valuable lesson- when using tube inserts, it’s “one or none”.  Multiple tubes introduces too much error.   So Max has created a custom single insert.

And THAT catches you up and you now know the rest of the story. 

When I get the gear with new custom insert, I look forward to trying it out- and I’ll report back.

After the first failure, I experimented with the Polish custom gear and found it to be a great solution.   So all of what Max and I have done with the stock gear is now academic- since this lathe work is a whole lot of trouble when a Delrin gear works as good or better.

[u18b]

I received the gear back from Max with the wonderful crafted inner bushing.

It is indeed quieter.  However….. there is still a slight eccentricity.

Max says he bored it well.  I still see an eccentricity.
After a bit more analysis, I think we are BOTH correct.
I’ll show you what I mean.

Let’s start with fact.   Max bored the gear.  He double checked to show that it was true.

But here is a short video to demonstrate what I see.

In this video, you will see three parts:  The worm up top, the worm gear in the middle (this is the gear in question), and the cylinder head gear.


In the next short video, will notice that there is a wobble in the cylinder head gear.  I’ve determined the bearing on which this gear runs is worn and will need to be replaced (the subject of future posts).   The point is- ignore the cylinder head gear wobble- it is not relevant to the current investigation.

For example, see this 17 second video with the new Delrin gear I got from Poland.  It runs pretty true- regardless of how the cylinder head gear is not true.

Now look at the same configuration with the brass gear which has been bored with a new insert.

You can see the eccentricity in the gear.

So….. how can Max and I be BOTH right?

Well here is the gear Max returned to me.
This is a bottom view.  The “boss” is facing us.  The boss is the part Max grabbed when lathing out the center hole.

And that hole is true…. Or at least true to the boss.



So if the hole is true, then why does it spin with an eccentricity in that video?

There are a couple of possibilities (or a combination).
The answer, I believe, is on the other side.

Here is the top of the bored out gear.  We already established it is true from the bottom.  But what about this side?



I blew up the photo and made some measurements with Corel Draw.

I measured the distance between the inner wall and outer wall.



When I do that around the gear in this photo, I do not get the same measurement- and I don’t think it is distortion from a good- but not perfect camera angle.

When you look at this gear, the side with the blue arrow looks fatter than the red side.  Thus the top is not true.

If the bottom is true but the top is not, what would be causes?

1.  Human error.  Maybe when Max chucked the boss on the lathe, the gear was slightly canted.  Not likely but possible.

2.  Maybe the boss itself is not centered with the teeth of the gear.

3.  Something else?


Being who we are, Max and I will continue to investigate.  This has become a puzzle we want to solve.

This time I’m sending him not only the gear and insert he worked on before, but also the extra Shay frame from the destroyed Shay, along with a spare worm.



This will allow Max to actually install the gear and test its operation.

I’m also sending him an unmodified worm gear for comparison.

[Ed Kapuscinski]

You boys are the maddest of lads.

I love it.

[mmagliaro]

The only thing I want to add is that in the first of Ron's photos, I could see that the boss itself is not actually in the center
of the gear, and all along, we've been banking on that.  I am chucking the gear in the lathe, so if the boss isn't concentric
with the gear teeth, then the whole thing will spin eccentrically, even if the bore is perfectly round.

Note, in this photo, how much that outer edge of the boss is actually not centered within the gear teeth.
Distances "A" and "B" should be identical, but they are not.



So.... what to do, what to do.
The only way I can think of to hold the gear so the bore will be concentric with the teeth is, crazy as it sounds, to actually
chuck it in the lathe by the teeth.  I am using a collet chuck for this, so we have 6 rounded jaws applying even pressure all the way around.  It's not like a nasty 3-jaw drill chuck biting into the teeth.  I *think* I can get away with just putting the gear itself into the collet, and then reboring the hole that way, which should center it.

[u18b]

And I was thinking also....

Since I'm sending you an unmodified gear.... you should be able to chuck that in the lathe by the teeth, but with the boss pointing outward.
As long at the gear surface is perfectly perpendicular to the axis of the center-line.... then when you spin the gear, you should be able to see if the boss itself is concentric with the teeth.

Heck..... I haven't mailed anything yet because of the crazy Christmas rush..... but I may do some more checking myself this week.
I could also mount that gear upside down and spin it and try to use a straight edge to make a determination of how true everything is.

[peteski]

. . .
Heck..... I haven't mailed anything yet because of the crazy Christmas rush.....

And weren't holidays supposed to be time for relaxing and unwinding . . . getting a break.   Sure. 

[u18b]

But model trains is relaaaaaaxing!

 

[peteski]

But model trains is relaaaaaaxing!
 

Well yes, until something goes horribly wrong, or that tiny screw unique to the specific  model, bounces off your jeweler's apron (worn specifically to prevent such incidents), and lands somewhere on the unfinished cement floor, becoming totally invisible!  Then the "relaxation" part goes out of the window, at least for some time. 

But on a serious note, our hobby is generally more relaxing than aggravating. If it wasn't we wouldn't be in it. It is nice to immerse oneself in some exciting modeling project, forgetting the crazy reality around us. 

[mmagliaro]

And... after a hiatus... this thread is back!
I have been sitting on Ron's worm gear and frame for over a month, just too busy to get to it.
But I finally did today.

As I planned, I chucked the gear in a collet chuck by the teeth.  Then I used my same micro boring bit to
bore out the center.  The bore on this gear, relative to the teeth, and the boss, was far worse than we realized.
I had to take the bore up from .105 to .118 to get rid of enough material from the hole to true it up.

After doing that, I went around the gear with calipers, measuring between the inside of the hole and the outer edge
of the tooth adjacent to where I was measuring, and I got between .036" and .038" everywhere around the gear.  So this time, I was confident that I had it pretty good.

Here's a video of the gear running in the spare frame Ron sent me, powered by a motor I just connected off to the side through a flexible tube.  The bad eccentric wobble from the first attempt is gone.  It's also quiet.  Most of that noise is coming from the metal gearhead on the motor.

The only thing I do notice here is that it looks to me like the worm and worm wheel should be pushed a little closer together.  The mesh looks pretty loose to me.  But since the worm wheel is mounted on a screw into the frame, one way to do this is to slot the frame hole and put a nut on the back for the screw.  Then it could be gingerly nudged closer, tightened, tested, and so on, and then fixed in place with some LocTite or epoxy.  But for now, at least this is bored right.

I agree with Ron, that a Delrin gear is superior to fixing this brass one.  It will be bored correctly right from the get-go, it will be quieter, and it will last longer.  Brass-on-brass causes a lot of wear.  Putting a Delrin gear in the middle there is a big improvement, I think, all the way around.

Here's the video link:
https://www.youtube.com/watch?v=TVg_ZN79CDk

[woodone]

I might have tried to bore an hole in scrap rod that was larger than the gear. Bore a hole so the gear could be pressed into the bored hole to hold it. Then bore the gear to speck and then press the gear out of the holding rod. Just a review 20-20 hind sight thought.

[u18b]

Thanks Max.
I look forward to getting this back and trying it out.  I'm interested in doing a decibel check as before.
It is CLEARLY going to run far better than before.

Hey Max...... did you measure the boss on the underside as  over and against the teeth?

Was the boss concentric with the teeth?

[mmagliaro]

Thanks Max.
I look forward to getting this back and trying it out.  I'm interested in doing a decibel check as before.
It is CLEARLY going to run far better than before.

Hey Max...... did you measure the boss on the underside as  over and against the teeth?

Was the boss concentric with the teeth?

The boss was definitely NOT concentric with the teeth.   Besides boring out the hole,  I then chucked the gear in the
collet with the boss sticking out,  and it was painfully obvious that the outside surface of the boss is not at all concentric with the teeth.  I could visually see it wobbling around.  So I just grazed a little off to true it up.  I did that just because it bugged me watching it.  Ha ha.     That surface doesn't affect how the gear runs, of course.

Woodone:
You idea is interesting ( about using a scrap of rod as a "holder" to bore the gear ).   I'd have to get that bored hole "just" a little bigger than the teeth so that it would press in snug without it trying to wander off center and without the brass teeth deforming.  Even so, I don't see how this makes things any better than the 6-jaw collet.  With the collet, I know the run-out and I know that when it closes down around the teeth, it is closing within a certain specified amount of precision.  It is also easy to make sure I don't tighten it too much.

[woodone]

Max, like I said 20-20 hind sight my idea. I guess it would depend has to where each six flanks would contact the gear. If one were to fall over a tooth gap the gear center might be off a bit. Two or more might make it worse.
My thought would be to bore the fixture rod a bit under size- heat it up insert gear and cool. Like I said just thinking outside the box.
These small parts are very hard to hold on to and trying to keep any close tolerance is next to impossible.
The flex of the boring tool can give you fits. Small hole , small tooling = flex. I even set my compound at 60 degrees so one turn moves the tool 1/2 the distance.

[u18b]

Max, thank you for this investigation.

Who would have thought that the boss was not concentric.
It would appear clear now that this was the original source of the error and problem since we assumed it was perfect and you grabbed the gear with the collet on that boss.

[Ed Kapuscinski]

I have been sitting on Ron's worm gear and frame for over a month

That sounds rather... uncomfortable.