What ratio do planetaries rotate at in a gearhead?

[mmagliaro]

The video gives the formula

1 + ring/sun

So, you must have 36 teeth in the ring (there are 12 on the "sun", the motor pinion).

1 + 36/12 = 4

4:1

[mmagliaro]

Great find, Tim.
I wish we could see what the ratio is.

Now that we know the relationship, if we could just count the teeth, we could know.  But that picture in the eBay
ad isn't close enough for me to tell.

[timwatson]

I just bought some - I'll dismantle it and tell ya when I get it. It's slow boating from China.

[timwatson]

So the little guy (the 6mm one) I posted has ~28 teeth in the ring and 8 teeth in the sun gear.

So I guess it's 4.5:1 in one set of planets? It has 3 sets.

So is it ~92:1 or 411:1 exponential math clogs my brain? I need one of my kids to do this for me. 

[mmagliaro]

I just bought some - I'll dismantle it and tell ya when I get it. It's slow boating from China.

Yay!  Research!
And by the way, I found those gearheads, or almost exactly the same gearheads, in other eBay listings for
under $6.00, free shipping from China.   The key will be how the input shaft works.  It looks like
any ol' motor shaft could be made to shove in there.

Something else everyone should be wary of.  These cheaper plastic gearheads (like the Gizmos one that I worked
with a few years ago) work well, but they are noisy and they do not have the efficiency of
the high-end ones that Faulhaber makes.  When you turn a Faulhaber gearmotor OUTPUT shaft, you are
reverse-driving the planetaries and the motor, and it STILL feels like you are turning NOTHING.  It is so super smooth
and efficient that it feels like there's no friction at all (because there almost isn't any).   They are rated at
90% mechanical efficiency, which is phenomenal.  With the plastic ones, you are lucky to get 50%.
So you've lost half your motor power to frictional losses.  You get a 4x multiplier, so the torque does
go up by 2x, and the speed comes down by 4x, but it's not as good as the higher-end gearheads that give you
almost 4x the torque.

(You have to remember, I've become spoiled by drivelines that have so much power that they pull a train up any hill or around any curve, and the engine does not slow down one whit, and the ammeter hardly registers any increase).

[mmagliaro]

So the little guy (the 6mm one) I posted has ~28 teeth in the ring and 8 teeth in the sun gear.

So I guess it's 4.5:1 in one set of planets? It has 3 sets.

So is it ~92:1 or 411:1 exponential math clogs my brain? I need one of my kids to do this for me. 

How'd ya find that out?  Can you actually see the teeth in that photo?
1 + 28/8 = 4.5:1  yes.

The reduction multiply together geometryically:

So 2 stages are 4.5 x 4.5 = 20.25 : 1
3 stages 4.5 x 4.5 x 4.5 = 91.125 : 1

And by the way, THIS is why there are no planetaries that have the oh-so-desirable 3:1 or 2:1 ratio.
You can't increase the diameter of the case if you want a 10mm motor.
You can't make the ring teeth any smaller because they are already 0.2 module (or smaller!) and going smaller
than that is just not practical.
You would therefore have to make the planetaries larger to have more teeth

In the 10mm with 36, you would need 18 teeth on the planet gears instead of 12.
They would become so big that you couldn't fit the motor pinion in the center of them.

4:1 is about the limit until somebody wants to make ever-smaller gear teeth.  And if Faulhaber ain't doing it
in stainless steel, I don't think anybody else can or wants to do it.

[timwatson]

LOL for you man I counted the teeth on the 6mm one with a pin.  ...

[mmagliaro]

I just discovered something quite surprising and important about gearheads.

Here is the list of facts.

1. I remotored an engine with a Faulhaber 1219 with a plastic Maxon 4:1 gearhead on it.  The engine could only
reach 20 mph at 12 volts.  Current draw was about 60 mA.
2. I replaced the gearhead with a stainless one.  Because it ran faster and had a funny looking nose on the front of it,
I assumed it had been customized by motorman to be a lower ratio.  But NO.  I conferred with him.  It is just a
Maxon stainless gearhead
3. The new gearhead has ball bearings in it, not sleeve bearings.
4. The engine reaches 65 mph.  It still draws about 60 mA
5. In both tests, I was pulling the same 17-car train.

This means that the frictional losses in the plastic gearhead, even a very well-made plastic gearhead like a Maxon,
far outweigh the load of moving the engine and pulling 17 cars. 
The same setup with the stainless ball bearing gearhead allows the motor to rev 3 times faster.

Wow.  I knew that plastic gearheads, and especially cheaper ones, were not as efficient.  But I didn't expect such a dramatic
effect.

[victor miranda]

when chasing effiency in gearboxes, there are two standard places
where friction happens bearing surfaces and gear faces.

ball bearings are a whole lot less friction than journal, and
the shape of the gear faces can range from little friction to a lot of friction.
the example I was given is that flat gear faces have a lot of friction in them.
The curved gearfaces use a lot less power to operate (lowered friction)

If IF if, it is possible, doing all your reduction with one gear and pinion set
is the lowest potential friction because of the low bearing surface and gear contact area.
the practical sizes of things means that 10 to 1 is a common limit for gears and pinions.
Worms and wheels often go higher, and effiency gets very expensive.

effiency is not usually a high priority in little gearboxes,
low price is the decision maker.

if you want to improve the small plastic gearboxes I have seen,
use very light oil.

I'd like an oil that stays on the motor pinion.

victor

[mmagliaro]

All true, Victor.
(It's probably no accident that the Faulhaber stainless planet gears have a slight curved taper in them).

I like to use a heavier Labelle gear oil (I think it's the 102) in the Faulhaber heads because it quiets them down somewhat.
Even though they are pretty quiet. at higher revs they tend to sing like a sewing machine.

I've tried a variety of lubricants in the Maxon plastic heads.  Light oil, heavier oil, and even lithium grease.  As it turns out,
the grease seems to make them run the quietest, although I can't say it's the most efficient.   I don't ever see the
current draw change enough to even notice.  When screwing the gearhead on, one has to watch the ammeter and
find the optimium spot, which is not always when it is screwed firmly all the way down.  I see the motor current vary
from 10 to 60 mA on some of these, so when you find the spot where it only draws 10, you put some epoxy or LocTite on there
so it stays put in that position.

(Of course, you have to be careful not to let the Loctite get beyond the joint between the gearhead and the motor.

.... and find its way down to the motor hole
... and get inside the gearhead
...
... or you will be picking hardened LocTite off little planetaries into the wee hours of the morning before it will turn again.

... just saying.

[timwatson]

Hmm that is a very interesting find Max. I wonder if some type of hybrid could be made. Ball bearing on the shaft end of a plastic gearhead. Small ball bearings are easy to come by (relatively). Just thinking aloud here. Thats is a drastic difference.

[mmagliaro]

Yes, Tim, small self-contained ball bearing race assemblies are available at low prices - on eBay and from places
like Boca Bearings.   How you would fit one of those onto a plastic gearhead, however, I can't envision.
I suppose if I disassembled the gearhead, and pulled out the front-most piece that has the shaft on it it that comes out
the front, and then drilled out that hole bigger so the output shaft would not touch the housing, I could then put it
all back together and slip a ball bearing race over the output shaft on the front, stick it to the front face with a dab of Goo
or something, and run the motor.  Now, nudge the bearing this way and that, measuring the current, get the optimal position
(lowest current) and then apply some JB Weld around the bearing to permanently glue it in place.

This would only work if there were enough output shaft to sacrifice some to go through the new bearing.

The other surprising (and bad) fact is that in my particular case, this was a gear motor joined to a worm via a rubber neoprene tube.
The worm runs in its own bearings in a separate gearbox, so the gearmotor is not carrying any shaft loading at all, and still,
there was this big difference in efficiency.

[Chris333]

Maybe the plastic vs. steel thing is why the 0-8-0 goes sooo slow?

[mmagliaro]

Maybe the plastic vs. steel thing is why the 0-8-0 goes sooo slow?

You mean the Walthers/LL 0-8-0 with the Mashima + Gizmo gearhead?
 I think it more likely that

1. The 0-8-0 was already geared fairly low.  The stock ones only hit about 75 mph at full speed.
2. The plastic gearhead is 5.25:1, so now you are down to 75/5.25 = 14.28 mph

[timwatson]

Here is the homemade hybrid gearhead idea I was talking about. The benefits is it's a 4:1 ratio and it could be very tiny.