Motors, torque, windings Scale Speed

[victor miranda]

Hi Max,

I kept the voltage the same, I added 5 cars
the 15 is close the the loco's limits,
If I twist the throttle a bunch the loco will grind rail.

I got 140ma, 4.5 volts and 55 seconds.

and after a half dozen laps
the loco has settled to
137 ma and 4.6 volts 47 seconds.

on the loop 30 sec/lap is about 45 smph and 40 is about 30 smph

and to give you a comparison to the fixed
30sec/lap
155ma and 5.7v

victor

[mmagliaro]

Okay, so now, what happens to the other engines, with the other motors?
That's the key.  When you start them out at 30 sec/lap with 10 cars, and add 5 more,
how much do they slow down?

That's the key bit of data that will say whether your motor has more torque than the others,
at least at that speed.

[victor miranda]

that test was all I had time to do yesterday.

we already have 10 cars
I'll add the 5 and keep the voltage.

I'll try tonight.

[mmagliaro]

Okay, here comes a big dump of text and graphs.   Ouch!

I made a series of measurements using my Kato E8 with the stock motor and after swapping in
an Atlas Scale Speed Motor.   The motor swap was about as darn near a "drop in" replacement
as you're going to get.   The Atlas flywheels have the same size hex holes machined into them
to accept the Kato drive shafts, the overall length is almost exactly the same, and it snaps
into the plastic motor saddle perfectly.  The only thing I had to do was trim off a little bit of the
plastic nubs on the ends of the Kato drive shafts, and steal the two bronze connection strips
from the Kato motor to put onto the Atlas motor (these are the strips that snap under the brush holders,
so you do have to pry out the holders, not lose the brushes or springs, put the strips in place, and then
reinsert the springs and brush holders).

The motor does indeed run considerably slower than the Kato, which I like.

Now...
Below is an eye-glazing "report" I put together with all the data I measured.
Feel free to read, criticize, point out flaws in my scientific method.  I am perfectly open to criticism
for what I may have overlooked or where I may have taken a short-cut that I can't get away with!

(You might also just look at it and say, "Ugh... Too much
data, next thread!   That's okay too!    )

If you can't bear to read all this, the "Reader's Digest" version is that the Atlas motor outperforms
the Kato.  It maintains speed better, can move the train more slowly, and draws less current doing it.
It is not nearly in the same class as a 4:1 Faulhaber gearmotor for achieving these goals, but it's a nice improvement
and I'm glad I switched it into my E8.

There are some graphs in the last pages of the report that may make my point more easily than all the
tables and numbers.

What follows are all the pages of the report.

[Mark5]

Max,

Fascinating stuff - can't wait to review this more thoroughly at home!

Mark

[Lemosteam]

Max, my only concern is that your test track may be throwing variables into the mix. If a loco has headway under torque, the mass might make it seem that the loco is climbing the grade more efficienlly, no?  Sit the loco start in the exact position as the other? Etc.

If say you had a sixteen foot board with straight unitrak on it, you could perform identical tests at various grades and compare the results.  The differences might even be visible.

Each locomotive would be facing the identical circumstances regardless of where you started.  Make the distance at grade the variable not the time.  You could create a specific time lapse for the transformer to cut out and then measure the distance traveled. Torque is power, right so the farther it goes up the grade  for the allotted time,could only mean one thing.  Besides a dyno, trucks that pull trailers have been measured this way and it is a way to compare pulling capability.

Just some thoughts of mine.

[mmagliaro]

John,
I'm not sure I understand.  In every case, I started the engine at a marked location on the layout, ran it around the entire mainline one time and stopped back at that exact spot, measuring the time with a stop watch.   That
starting line" spot was on a long stretch of level track, with feet of level track leading up to it.  So the train could always start several feet behind the starting line, while I got the track voltage up to the number I wanted for the test (i.e. 3.25 volts, say), so that
by the time the engine crossed the starting line, it had been running steadily at the prescribed voltage for several
feet.    I don't think there's any variability in this,  but if there is, take another whack at me and try to make me see it.

[Lemosteam]

My assumtion was that you did all of these things.  I think the variable that comes in is that the locomotives will get to the beginning of the grade at different times with different rates of speed therefore different amount of force to get up the grade, if that makes sense.

[mmagliaro]

Oh, I see.  Well, getting to the base of the hill at a different time, I don't see how that matters.
And as for speed, I did try to get the two different motors running the train at similar speeds
when I ran these tests.  That is, the Kato motor reaches the base of the hill at 9 mph, or 30, or whatever,
and so does the Atlas.   When there's more load on it and one slows down more than the other,  it reaches
the base of the hill going slower than the other, but that seems
like a real-world case to me.  That's what the train will do when you run it on your layout.

No, I'm not strictly measuring the performance on the hill.  I'm just getting a measurement for how the engine does
over a whole real-world lap of a mainline

[victor miranda]

....
micromo states
current and torque are related and they hold the voltage steady.

If you are hunting torque.... you need to current values.
they state the two are directly related...

if you are hunting power....voltage and current are good to have.

I am thinking I'll find time to try to match your numbers... over the weekend I hope

more soon
I have to shift gears to read your post again.

victor

[Lemosteam]

I'll leave that dead dog lie. No biggie. I just see many opportunities for variables that could/may affect your data. I guess that's why a dyno in a controlled environment is so effective.

[victor miranda]

Hi lemosteam,

there are a lot of ways to skin this cat
and the important part is to attempt to get anyone who want
to try this to self calibrate.

I think we can do that part
what we do not have is a reasonably thought out
set of tests to use to compare.

I happen to like two tests.
how do you like it in shunting?
how does it respond to the control knob?

max has another evaluation
he likes how steady the loco speed is under varying loads
... like a grade?

I happen to like a loco to slow up a grade
and maybe go a little fast down a grade.

sadly, we have motors that do that.
However the stout hearted motor Max seeks is not
a common beast.

what we are avoiding is creating the four line graph
the motor geeks make to sell the motors.

I don't think we need to be accurate to three decimal points
we can use most locos and three loads and three speeds to get us
useful comparisons.

I am still trying to get motor RPM out of my speed runs.

victor

[Lemosteam]

Victor,

That is quite fair enough. When testing anything it is always good to know that the only variable in a test of this kind is the motor itself.

[mmagliaro]

Edit for spelling errors...


Victor is probably at the nut of what I'm talking about.
We don't have a dynamometer, and although that would tell us which
motor produces the most torque and the most power at different rpms,
ironically that would still only tell me tangentially which motor I might like better.

Watching how it actually runs at low speed around the layout, up and down hills, and around
curves, really tells me if it's something I like or not.   And Victor hit upon one other point:
how well does it do at switching speeds?   I did note in my report that the
Atlas motor is capable of moving the engine slower than the Kato motor, including
up the hill, which is also a good indicator of better switching performance
because you can move loads in and out of a yard track at lower speed without
stalling.

I'm all for the train slowing somewhat up a hill and speeding up somewhat down the hill.
The key here is what "somewhat" means to different people.   I find the amount the variation
in this E8 with the Kato motor to be acceptable, but not great.  At 30 mph, it can climb the hill without
stalling, and it doesn't "run away" coming down the hill.    And it is a fair grade (1.7%) not some absurd
non-realistic 6% grade.

There are lots of variables at work here.  The particular engine I'm testing it in, how much friction is in its drivetrain,
how it is geared, to name a few.  What I can say with certainty is that in this engine, I like the Atlas
motor better.

[rodsup9000]

OK, Max and Victor,
  I think I want to get some of those Atlas motors and try them in a ABBA set of Kato's F units.  I do not have any of the newer (last 20 years) Atlas locos, so what part # do I need to get them.

Thanks