Please check my flywheel calculations

[mmagliaro]

...and also provide variable momentum.


I suspect that "electronic flywheels" are against Max's DC religion. 

I really don't want to make my mechanism dependent on a decoder in order for it to run well.
That would be too much like giving up.  That would be tantamount to what the "B" company is doing now,
and let's just say...

ummm. 

how shall I put this....

Maybe you had it right all along, peteski.  It's against my religion.

[nscaleSPF2]

Not sure if you have thought of this, Max, but making the flywheel out of tungsten instead of brass would double its effectiveness (if I remember my college physics).  Don't know if this would help enough, or not, though.

[mmagliaro]

Not sure if you have thought of this, Max, but making the flywheel out of tungsten instead of brass would double its effectiveness (if I remember my college physics).  Don't know if this would help enough, or not, though.

Thanks!  And yes, I thought of it.   That would get me .045 x 2 = 90 mSec of spin time, because tungsten is indeed 2x more dense than brass.
The diameter of the flywheel is the biggie.  The momentum increases as the square of the radius.  But I have no more room.

The big thing is... who can machine a tungsten disk, or worse, center-bore a 1.5mm hole in said disk?  Yes, some machine shops can do this, but I sure can't, and the last time I solicited a service to make such flywheels, the cost was something like $20 each,
in quantities of 50 or more.

[GaryHinshaw]

At the risk of starting a holy war... I'll just point out that making your mechanism dependent on a decoder flywheel in order for it to run well is not really any different.  Back-emf sensing is just an electronic flywheel.   

None of this is intended to impugn your magnificent craftsmanship, however!

[mmagliaro]

At the risk of starting a holy war... I'll just point out that making your mechanism dependent on a decoder flywheel in order for it to run well is not really any different.  Back-emf sensing is just an electronic flywheel.   

None of this is intended to impugn your magnificent craftsmanship, however!

I only cursorily see your point.  The flywheel is a part of the mechanical system that is the locomotive.  The decoder is an electronic control, separate from the locomotive (even though it is carried inside).  In other words, the challenge to myself is to make the locomotive mechanically correct so that it runs well.  Back emf would be like relying on "somebody else" to make my engine run right.  Putting a flywheel in would be relying on me to do it.

[peteski]

I only cursorily see your point.  The flywheel is a part of the mechanical system that is the locomotive.  The decoder is an electronic control, separate from the locomotive (even though it is carried inside).  In other words, the challenge to myself is to make the locomotive mechanically correct so that it runs well.  Back emf would be like relying on "somebody else" to make my engine run right.  Putting a flywheel in would be relying on me to do it.

Once you get as far as you can perfecting the mechanical aspects of the locomotive (achieve smoothest running mechanism and include a flywheel) and you want to go even further, that is when you add the electronics (virtual flywheel).  At that point you have achieved N scale nirvana. 

[Cajonpassfan]

Peteski, never argue with a man's religion; his Nirvana may not be your Nirvana
And I can respect that...in life and in modeling.
Otto K.

[GaryHinshaw]

I was just making a (perhaps-too-snarky) response to your earlier comment:

But... (there's always a but)... The rotational motion of the drivers isn't perfect.   
...But if a flywheel could "sweep it under the rug", I'd be all for it. 

If "sweeping it under the rug" is your only goal, an electronic flywheel for dealing with imperfect rotational motion is also effective.  But it is not hand built, and I totally get your motivation there.  I'm not trying to convert you by any stretch.

[peteski]

Peteski, never argue with a man's religion; his Nirvana may not be your Nirvana
And I can respect that...in life and in modeling.
Otto K.

I'm just pulling Max's leg - I know he is a DC-man!   

[nkalanaga]

Mmagliaro:  It looks like I misunderstood your question.  I thought you had a relatively slow motor connected directly to the drivetrain, and wanted to use the gearhead to turn the flywheel faster.  I've heard of that being done, but it was on a motor-generator, and the idea was to store larger amounts of energy, not to smooth mechanical performance.

[mmagliaro]

I'm just pulling Max's leg - I know he is a DC-man!   

The funny thing is, I really don't have any religious attachment to DC.  I use it because:

I pretty much, 99% of the time:

a) never "operate"
b) never run more than one train at a time
c) never run multi-unit lash-ups, and on the rare occasions when I do, they are perfectly-matched engines
(like 2 or 4 identical Kato F units, and even though they may not be "identical", they are awfully darn close)

So to me, DCC would just be another thing I have to do that would add very little to my enjoyment.
It's always been a low priority thing.

Now... if and when the day ever comes when there is true, portable, battery power in N Scale locos,
I would switch in a heartbeat.  Getting rid of the nuisance of keeping track clean would be worth it to me.

[delamaize]

i like trains.

[narrowminded]

But...
(there's always a but)...

It just has occasional points where it slows down or speeds up ever-so-slightly.  By 5 or 7 mph, this side-effect completely goes away.  So I am being very fussy here.
This is something that, most likely, could be impossible to improve or "fix".  We have a hand-made mechanism here, and there are so many potential sources of tiny errors that could cause such a small effect that I would go mad trying to "fix" it.

And I do not intend to do that.

But if a flywheel could "sweep it under the rug", I'd be all for it.  That's what got me thinking this way.

When you say it speeds up or slows, is it rhythmic with wheel rotation?  Or is it a few inches of travel, then goes away for some distance, then reappears?  I ask because if it's rhythmic with wheel rotation it could be a slight bind at the axle or an out of round wheel or something at axle speed.  If it's rhythmic but at a faster rate than the wheel, maybe a bind of some sort one step back in the train.  Anyway, you get the point and probably already had that covered but if that's the case and subtle as you described, it may just be a break in issue that a few hours running then a good dismantle, cleaning, and fresh lube, will fix.

Then again, if it's rhythmic and determined to be caused by a wheel running slightly eccentric there's probably no fix other than truing the wheel.   A flywheel nor BEMF will fix it because it's not a speed change at the motor, the only place those two things can help.  The motor's doing fine, it's just the eccentric rotation that's causing the issue.

Then again, is it pickup?  That might occur in the non-rhythmic scenario.  This is where a milliamp meter of a suitable range can be a godsend.  In my testing that has been the handiest tool of all proving pickup and electrical issues but also improvements or backwards steps with mechanical parts.  Bearing fits, lubrication, gear mesh, etc, all show up as effects on the amp draw.  The meter I'm using is a 100 mA range as the biggest motors I'm testing have a stalled current draw of around 80 mA.  The useful readings for me are in the 2 to about 30 mA range and almost never exceed 40.  My original 1 amp gauge wouldn't read fine enough for that so I picked up an analog 100 mA gauge which affords instant and usable, visible feedback.  That's been pretty ideal for these purposes.  Sometimes digital with it's decimal points for its other decimal points just isn't a very useful tool.   This is one of those times.

[mmagliaro]

It is most likely a minute error in quartering, or the spacing of my rod holes, or even a little error in the distance between the crank bosses and the center axle.  Remember, all this stuff was done by me, by hand.  And even though I meticulously cut and measured everything along the way, it's still not 100.000% dead on.  A thousandth off here and a thousandth off there, and it all adds up.

And the fact that it's so slight only makes it more impossible to try to "fix".  I did a lot of work over the past couple of weeks experimenting with sets of rods, ever-so-slightly varying the rod hole spacing, and so on.  It's at a point now where most likely if I mess around with it, I would only make it worse.  So I have no intention of messing with it.  I've hit the "diminishing returns" point.

[narrowminded]

That sounds good and running it in will probably smooth it out even more.  I'm anxious to see the final result as it's one of the coolest projects ever.