Motors, torque, windings Scale Speed

[victor miranda]

Hi Bill,

I once tried to wrap a 5 pole motor...
if you think 3 poles is normal for effort
a 5 pole is past 5 times that effort..

the thinner wire is tough to handle.
so believe me I do not plan to build a 5 pole
unless we have a lot of data to show there is an area
we can't reach any other way.

BTW double and triple winds do have interesting qualities in a race motor...
they seem as powerful, only a little less fast, some what cooler
and far easier to control.

that may be what we want...

lets beat the bushes fo find one someone has built first.

victor

[victor miranda]

static measure  after a spin to clear the commutator

atlas grey end bell 26 ohms

ll diesel sw1200 114 ohms

a green paint Kato 50 ohms

these three are remarkably close in size

looking at the windings, the LL has the finest looking wire
and it looks the least filled on the armature

the kato and the atlas look like the same wire
and the kato looks more filled.

some one else look measure and decide if they get different...

victor

[mmagliaro]

Watch this!
http://hackaday.com/2011/10/06/video-everything-you-wanted-to-know-about-dc-motors/

It gets very interesting at about the 9:00 mark, where he starts to rewind the motor with heavier wire
and discusses what effect it will have.

That Atlas gray (Scale Speed, I presume)
motor spec, Victor, says something I hadn't considered before.  I guess I was assuming that
a motor wound with thinner wire would still FILL the armature up.  I never thought they would
wind with thinner wire AND use the same, or fewer, turns.   I suppose they were picking
the spot where they could use enough wire to present enough resistance to not burn the wire out
at stall, but no more.   This would draw as much current as possible without burning
the wire, I suppose.   Perhaps they found the spot where they got the best compromise between
torque and rpms.   Adding more wire would make for a stronger magnetic field, but perhaps
the motor runs too slowly.   Boy... it would be nice to be able to wind several of those gray bell armatures
with more and fewer turns and see what happens.

[victor miranda]

keep in mind that the cost of magnet wire goes up as the wire gets thinner

I paid more per pound for 44 gauge wire than I did for 36 gauge.
from a motor building perspective thinner wire breaks way easier.
so more frequest failures add to the price.  or the fuss in building.

it is easy to see that Atlas is not making a more expensive motor.

from what I see you are chasing, I am a little nervous.
Atlas' gray motor seem to have taken the slotcar road,
Compared to the kato anyway, to get the static resistance down.
The appearent volume of wire implies a shorter length of wire
and I am thinking they are very likely to have use a slightly thicker wire,
like from 40 to 39. mainly it does not look like it lost a lot of volume.

at this point I have to say this... I have always thought of the number
of wraps of wire in the coil as 'efficiency' or 'magnetic traction'

so Atlas gray is half the static resistance and only lost a quarter of the wraps.
(this is guessing.  I ain't cut nutin' up here) the two together keep some effiency
that would be lost if they just cut the wraps by half.

we have a problem...
a vohm meter is easy to use  for static motors...
how do you want to test for torque/efficiency?

victor

[victor miranda]

here is an idea.

I have a pair of re-motored A/RR picifics
the both have the same motor and gearing...
and they have differnet features loco One has ball bearings on the worm and is pulling
a cc hudson tender.

I set them each on the track and ran them at 45 smh (30 sec a lap)
and got
loco 1 110ma and 5.3v
loco 2 90 ma and 4.3v

and then towing a 7 car train
loco 1 130ma and 5.7v
loco 2 110ma and 4.8v

now to process it a bit to explain my thinking for the idea.

loco one runs by itself are 583 mW and two at 387 mW.
towing cars loco one 741 mW and two is 528 mW
 
the power to pull these 7 cars is 158 for loco One
and the same 7 is 141 mW for loco Two.

this gives us a known load ( approx 150 mW)

If you assume this load is the same at all speeds (a big if)
you can test the motor against the load at various speeds
and then run the loco without the load and any time
the above varies you are looking at a place where the
efficiency of the motor (or the loco) has changed.


victor
 

[mmagliaro]

You've confused me as to why this helps understand whether thinner windings or more of them
is making more torque at a good speed for a loco.   Whether the motor is more or less efficient
doesn't really matter?  (Unless one motor is drawing an unmanageable amount of current).

I would be more interested in setting two identical engines (except for the motor) at the same speed with, say, 10 cars, and then keep piling on cars until one loco noticeably pulls ahead of the other.
The rotational speeds of the motors started out the same, so therefore the power output is the same.
As you pile on cars, you are demanding more power from the motor.   If one of the motors can't deliver any more, it will start slowing down.   Power = rpm x torque   If you add a car and the power demand goes up by 10%,
then the motor has to deliver 10% more torque to maintain the speed.    Once the torque maxes out,
the rpms will drop.

In fact... why didn't we think of this BEFORE????   Anybody got two identical Atlas locos?  One with the original
high-speed motor and one in which they replaced it with a Scale Speed motor?
And they have to be running straight DC.  We can't have the decoder trying to compensate
the motor to keep the speed up.
Run each one separately with 10 cars, at a low speed, say, 20 scale mph.  Tack on 5 cars,
note the speed.  Tack on 5 more, note the speed.  Etc.

Whichever one slows down more has lower torque.
 

[victor miranda]

back emf will rear its ugly head...

I want to say I think you are on the right path with that test
but you are allowing the motor speed to change and that will also
change efficiency ratings.

motors have efficiency curves... they change with motor speeds.

measure at a given speed with no load  and against a known load at the same speed

if you plot this for about 5 speeds you can see the torque curve ( I think....)
same loads against another motor should allow comparison.

the Kato motor should display more torque than the Atlas slow speed motor.

victor

[strummer]

Hi Strummer,

what conclusions did the O guys reach?

the differences may be in different units, and our options may get limited by size constraints.
I suspect the end points may be close.

victor

I'm afraid I can only recall that a certain type of Pittman can motor (the series with a "2" or "3" after the first digit) was generally thought to better than others,although it pains me to admit I can't say why, offhand.

Wish I could be of more help. You are correct in assuming that O scalers of course don't have to worry quite so much about interior space; I remember putting a very large Central Locomotive Works open-framed motor in an Atlas diesel. That motor itself probably weighed more than a half-dozen (or more!) N scale engines...fun stuff,that. And of course the sheer mass of models that big comes into play as well...

Mark in Oregon

[mmagliaro]

back emf will rear its ugly head...

I want to say I think you are on the right path with that test
but you are allowing the motor speed to change and that will also
change efficiency ratings.

motors have efficiency curves... they change with motor speeds.

measure at a given speed with no load  and against a known load at the same speed

if you plot this for about 5 speeds you can see the torque curve ( I think....)
same loads against another motor should allow comparison.

the Kato motor should display more torque than the Atlas slow speed motor.

victor

And what if one motor displays more torque up to 30 mph, and less after that?   Ugh.... 
I guess you would choose one or the other depending on how fast you like to run your trains!

[victor miranda]

Hi Max,

I kinda brought that testing idea because I am not sure how we will know
that we have accomplished something  I like atlas slow speed motors
in that they do not seem to overheat.

has anyone tried one on a Kato mikado?
... I've never overheated a mike... so I may not be a good tester?

Max, please give it some thought, I know you want more torque
I am not sure I have seen a 'more torque' motor

the motor in the IM ac12 is famous for being 'high voltage'
and from all I have read about how to get more torque
I am pretty darn sure that motor is on the 'go check it' list...
I'll see if I can find it.

I have another post I want to write....
back soon I hope.

victor



 

[mmagliaro]

Victor,
I suppose I should narrow my search.  What I really want is more torque at lower rpm.
And I am will to sacrifice high end speed and high end power to get it.

So a motor that produces more torque up to 30 mph is what I'm really after.   
Now, I do like your idea of testing at varied speeds.  If I could test two Atlas engines,
one with each type of motor, at speed steps as you suggest, that would really show us the
torque picture.

Unfortunately, I do not have such beasts.  And I'm not eager to spend the $150 + to get them.
I guess I could plunk down my 28 bucks and just put an Atlas Scale Speed motor in my Kato E8, and
test it against the Kato motor at various speeds.

[victor miranda]

Hi Max,

right now I feel like I ran off with your thread...

I did some hunting in my shop and found a kato green paint motor I marked as bad.
I opend it up and checked things...
to be sure we may want to check resistance at the comm plates
cause I tested the motor from coil to coil and found the resistance about 9-10 ohms
for the motor to run, two coils will be the minimum so it should be 18 ohms across the brushes.
so much for the earlier resistance readings.

but it is a start.

I had to cut the first coil off and I counted it as I went.
I destroyed the second coil in attempting to unwrap it and I lost my count.
the kato arm has 95 wraps on the top layer and the inner layers are shorter so I don't know it the wraps in the interior
are increased or not.  I unwound the third and fourth coils.
I will try to get the count right on the last coil.
I measured the wire with insulation as .004 inches diameter.
and one wrap at the top was 38mm and  one interior was 30mm
total guesstimated length is 38*95 --> 3.6 meters approx 10 ohms...
all that leads me to 40 gauge wire. so thinner wire will be required.

I found the AC12 motor.
I ran it at the same speed are the earlier locos, it runs at 112ma and 8.4 volts (941mW)

from that info, I am thinking this is the motor you seek.
the test bed is similar to an 0-8-0

the motor seems like it uses slightly thinner wire than the kato
and the armature is full like that atlas gray motor.

the current draw and voltage point to a higher static resistance
the thinner wires point the same way.
this motor has to have more wraps than the atlas and I am thinking than the Kato.
so it may have the higher torque you want to try out.
were I you, I'd try to get one from Intermountain to test.

victor

[victor miranda]

one last kick...

I have a kato 5 pole armature I have free of coils.
and If I can locate it, I have a spool of 43 guage magnet wire.
Two 43 awg strands are about the same resistance at one 40 awg wire

if the same 95 wraps of two strand will fit, the resultant motor should have more torque
at the same voltage/current.

shall we give that a try?

victor

[rodsup9000]

Victor, Thanks for doing all this research. I'm learning a lot from you.

[Bill H]

one last kick...

I have a kato 5 pole armature I have free of coils.
and If I can locate it, I have a spool of 43 guage magnet wire.
Two 43 awg strands are about the same resistance at one 40 awg wire

if the same 95 wraps of two strand will fit, the resultant motor should have more torque
at the same voltage/current.

shall we give that a try?

victor

Victor:
Try it, but I think you will get more RPM and less torque- if my memory of my experiences winding hundreds of competition slot car armatures is correct. On the other hand, I might have that bass akwards. One last thing that might also be noted it to keep the commutator centered, in slot cars we would advance it to change the timing, as we only expected the motor to run in one direction. Moving the commutator in relation to the blank effects both the RPM and the power curve.

And if you really want to get confused, there used to be a style in slot cars when double winding of first winding one wire by itself, then winding the second wire on top of the already wound first wire rather than winding both wires simultaneously. Purportedly made a smoother (I assume less cogging) armature.

Kind regards,
Bill