2 speed gearbox/jackshaft

Miles

100 TW
Joined
Mar 16, 2007
Messages
11,031
Location
London UK
There's been a lot of discussion about the use of IGH type gearboxes, recently.

I thought I'd throw this out, to get some feedback:

-It's a retro-direct gearbox with a ratio of 2:1.

-Second gear is direct drive (minimal additional losses).

-It has integral freewheeling, as do most retro-direct type drives.

-In first gear only, it has the ability to switch between freewheeling mode and regen. mode

-Gear change is by reversal of motor direction (no need to ease off the power...).

-Maximum input torque is 40 Nm.

-Weight is about 1 kg (as are most 3 speed IGHs).

-Diameter 100mm. Length 72mm (not incl. shafts).

-To make it clear - this is not a gearbox for combined motor/pedal drive, it's for motor only.
 
Interesting.
But maximum input torque is 40 N*m, so maximum output should be 80 N*m. Is it capable for about 50 N*m input and 100 N*m output?
How to link a motor to the gearbox?
 
Torque limit is set (very conservatively) by the one way clutches, which are on the input side. Nominal torque of these is 85 Nm with a maximum of twice that....

The number of planet gears can easily be added or subtracted to suit the actual torque requirements (up to 8 planets will fit).

I've designed it with a view to its use as a jack shaft so, synchronous belt. This way, some of the weight of the gearbox is offset against its function as a jackshaft.

You could have it inline with the motor (direct coupling).
 
Its lovely.
I am curious about the ratio spread.(current design & the min/max potentials)
also looking forward to some section & cut-away renderings.

thanks Miles.
 
Thud said:
Its lovely.
I am curious about the ratio spread.(current design & the min/max potentials)

Thanks Thud,

This one is 2:1 - direct drive and 50%.

For the given ring gear. Using a smaller sun gear makes the coupling with the clutches more difficult. Use a larger sun gear and the planets start to get a bit small.

I haven't incorporated anything for reversing (rollback). This is one of the drawbacks of retro-direct designs. Sure, there are ways around this but not easily and Xiongda had problems with this.
 
Hey Miles....WOW.....amazing.
My feedback is that I would have one..... :mrgreen:
The freewheeling/regen option is a very interesting feature....It allows regen with systems that are actually out of this game.
The input torque looks good enough to cover a wide power range, I would say perfectly suited for a 5:1 belt reduction with a 3220 ^_^ (7Nm*5=>35Nm), what about the RPM limits though?.

The rollback problem could be a minor issue for the 2 wheelers, but not for trikes and 4wheels, otherwise potential fields for it.....

It's amazing, I've been away from the forum for months, but was trying to move and push forward one other puzzle's piece just the last week.... :shock:

A sort of actuated compact&light standalone dry clutch unit could be the next needed piece of the puzzle?
 
Hi Jules,

It's a bit ancient history, for me :) I did a proof of concept version, 3 years ago and then put it aside to concentrate on motors.

This is what it looks like with the motor I'm developing.
 

Attachments

  • Gearbox-&-Motor.jpg
    Gearbox-&-Motor.jpg
    51 KB · Views: 3,615
How about availability? Mostly interested in availability of the gears since I can manufacture the other parts.
 
Miles said:
panurge said:
what about the RPM limits though?.
With straight cut gears, there's a noise tolerance factor, in first gear - not second, though, as it's direct drive. I envisaged running it at around 1200 rpm. Maybe 1500 rpm?

sounds good....and mated with your motor looks even better.... :)
 
nice thing. I dont think that this motor needs a 2-speed gearbox for most purposes though. I thought the motor was to be run at 3000rpm or so, which asks for single reduction. at 4 times this speed, like shown in the pic with belt drive, this will be a monster :twisted: a tiny 3220 is just a joke with its <10Nm torque abilities. Sure a high speed motor is needed here for ratios like 4*2*4 = 32 in first gear. This is 12800 rpm at 50 kph for a 26" wheel

regen with two speed retrodirect is somthing i also thought about a lot about. The thing that killed it always was actuation. How do you wanna do it? Shifter cable or servo drive?
 
It's a retro-direct gearbox with a ratio of 2:1

I used to think that maybe a slightly closer ratio would be more ideal, but now...If my standard cruising speed on a street commuter is 28-ish MPH (45 km/h), then I am now convinced that a max top speed that is higher is not a bad thing. In fact, max efficiency is often near 80% of max top speed (IIRC).

So, if actual top speed is near 33-MPH, and my most-often used cruise speed is 28-MPH, a 2:1 gearbox would put my "hill climbing" gear at a max 16-ish MPH, with max efficiency near 13-MPH?
 
crossbreak said:
I dont think that this motor needs a 2-speed gearbox for most purposes though. I thought the motor was to be run at 3000rpm or so, which asks for single reduction. at 4 times this speed, like shown in the pic with belt drive, this will be a monster :twisted: a tiny 3220 is just a joke with its <10Nm torque abilities. Sure a high speed motor is needed here for ratios like 4*2*4 = 32 in first gear. This is 12800 rpm at 50 kph for a 26" wheel
I don't think it does, either. Just a few sport cases and cargo bikes etc. The motor is designed to be used between 1000 rpm and 4000 rpm. So, the 4:1 drive, illustrated, would give 1000 rpm and 500 rpm output from the gearbox/jackshaft.

There's been endless discussion on the merits of variable gearing. How to justify the extra weight and losses instead of using the weight differential for a larger motor, etc. That's why I wrote
Miles said:
I've designed it with a view to its use as a jack shaft so, synchronous belt. This way, some of the weight of the gearbox is offset against its function as a jackshaft.

crossbreak said:
regen with two speed retrodirect is somthing i also thought about a lot about. The thing that killed it always was actuation. How do you wanna do it? Shifter cable or servo drive?
Could be either. Shifter cable would be simple to do as the the casing is stationary.
 
Rouckie said:
How about availability? Mostly interested in availability of the gears since I can manufacture the other parts.
Nothing is going to happen about this until next year. If you want gears, you can get them from HPC: http://www.hpcgears.com/ So, there's nothing to stop you, now :)
 
With the stated rates and the target motor, it would last a life.
spinningmagnets said:
It's a retro-direct gearbox with a ratio of 2:1

I used to think that maybe a slightly closer ratio would be more ideal, but now...If my standard cruising speed on a street commuter is 28-ish MPH (45 km/h), then I am now convinced that a max top speed that is higher is not a bad thing. In fact, max efficiency is often near 80% of max top speed (IIRC).

So, if actual top speed is near 33-MPH, and my most-often used cruise speed is 28-MPH, a 2:1 gearbox would put my "hill climbing" gear at a max 16-ish MPH, with max efficiency near 13-MPH?
Yes, that's the main idea and objective of the gearbox I suppose; the 2:1 factor is a good thing for the motor range target stated by Miles...Plus, in the real world, a 2:1 mech reduction will result in slightly closer speed and torque values for obvious reasons.
On that topic, is important to note that the GB is a 50% in 1st and not a 2x overdrive in 2nd. This is a natural approach but probably means also a gearbox projected to extend, occasionally, or better: -often but not continuously- the max climb or load ability, of a motor that could otherwise stall or work on terribly inefficient spots.....and not to extend the upper speed band of a motor
crossbreak said:
a tiny 3220 is just a joke with its <10Nm torque abilities.
that would probably never stall at any grade, yet, with the right single speed ratio.

Having 1500rpm and 40Nm as limits in mind, a 3220, after a 5:1 belt reduction (7Nm peaks and 8000rpm=>35Nm and 1600rpm) is actually the most powerful kind of motor to mate the gearbox with. And you cannot couple them inline, because of the overrun....
It is perfect for Miles Motor or eventually for a 3210 with the right reduction....and would be very interesting for cargo applications, even with smaller motors, if not for the rollback problem.
The Ideal motor would be of course one capable of the 40Nm and 1500rpms limits, without any reduction.... :)
 
Not exactly sure how this works but it looks like the ring gear is attached to the output shaft and the planet gears are stationary. Correct?


This would be perfect for a heavy duty cycletruck that can handle 450lbs (and up) on a 9% grade with legal power limits (Washington state or Canada) a "hill climbing" gear set at max 16 MPH, with max efficiency near 13-MPH is perfect. And the high gear does not need to be faster than 20mph.
However in Canada you may need two motors as well as the two speed gear box; switching on the second one only to climb the steep hills at 1300 watts. Assuming that you keep both motors controlled down to legal power.

Maybe would a voltage multiplayer work better? But conspiring against the absurd laws is tricky. I would rather have a vehicle that is illegal only when climbing the really steep hills.

And if you want a lower RPM try using a hubmotor at 72 volts and a lower amperage. I think it may work at 500-1000 rpm...not sure about the efficiency compared to a smaller motor running at 7500 rpm.
 
Back
Top