Epicyclic retro-direct 2 speed transmission

[mclovin]

McLovin,

2. Instead of changing rotation direction use similar and simpler mechanical steps to create a 2sp tranny.

John

Can you elaborate?

BTW I'm not advocating any particular method. My intention is to contribute to the body of knowlege via graphical means.

Cheers.

 

[Miles]

I think changing the motor input direction can be done rather simply by a mechanical system.

That's true but, I'd consider the fact that the retro-direct system in conjunction with an electric motor has the ability to change ratios by non-mechanical switching an advantage.

- Easier to route electrical cables than mechanical cables.

- More straightforward and flexible for the implementation and control of automatic operation.

- No need to synchronise any power input compromises during changeover.

The major disadvantage of the retro-direct system is the inability to back drive it for regeneration.

 

[Miles]

Show me a simpler 2 speed than this.....

 

[mclovin]

- More straightforward and flexible for the implementation and control of automatic operation.

- No need to synchronise any power input compromises during changeover.

I was unaware that motor reversing ESCs were available for RC motors. Then definately - reversing motor direction is preferable - but you guys alreadt knew that.

Where can I find one of these ESCs? Can you point me in the right direction?

Cheers.

 

[Miles]

http://www.castlecreations.com/products/hydra_hv_line.html
http://www.castlecreations.com/support/documents/hydra_hv_user_guide.pdf
http://www.castlespecialprojects.com/csp_reversing_controllers.html

The software isn't very well suited to our usage, though - not high enough PWM rates for higher pole count motors (unless they've changed recently).

 

[JS Tyro]

My understanding is that some motors can be reversed by switching the power cables at some location (on the controller? ). If that's the case, you could just put a DPDT relay in the path (crossing the wires on one side relative to the other) and get a simple electrical reversing switch on the handlebar. Much cheaper and easier than any mechanical device. Downside - the size of an xx ampacity relay (where xx >> 10).

 

[Miles]

You can change the direction of a brushless (& sensorless) motor by swapping over any two motor phase wires (not the power input wires to the ESC!), your idea would certainly work with that - better to have it handled by a suitable controller, though....

 

[12p3phPMDC]

Miles,

I take it that the blue sprocket is the output and the freewheel is embedded
in the large cog reduction..?

 

[Miles]

Miles,

I take it that the blue sprocket is the output and the freewheel is embedded
in the large cog reduction..?

Yes, the blue pulley is for output by 8M toothed belt (something happened to the tooth-profile...) - it could be chain..

It needs at least two opposing sprag clutches on the motor pulley and sprocket.

Nothing to do with the system which is the subject of this thread, of course.

See here: http://www.endless-sphere.com/forums/viewtopic.php?f=28&t=8919&start=0

 

[mclovin]

Show me a simpler 2 speed than this.....

I don't know of a simpler 2-speed design idea. I think that's pretty well reduced as far as it's going to be. I'd be interested to see/hear about your ideas for gear lubrication. I'm guessing your planning on a gear box with some shaft seals. Or a big glob of grease?

Are you thinking of leaving it as a single stage reduction?

I wonder what's involved is dissipating the rotational energy of the rotor as it transistions from CW to Full Stop (if momentarily) to CCW? Would you (or the controller) apply some kind of braking or just simply reverse the rotating field and wait? Either way I would suspect that braking/reversing the motor would cause a large voltage spike in the controller but I wouldn't know, really.

 

[Miles]

It'd probably be more interesting to hear your ideas on lubrication I might do a test version with a nylon gear and steel pinion, sometime. I'm not keen on using gears on the first stage - I hate things that sound like electric drills..... The other disadvantage of this type is the limited ratio difference.

I envisaged it as single stage plus another 6:1 to the wheel by 8mm pitch synchro.

We need to do some experiments on motor reversing, for sure. Boat controllers allow you to go from forward to reverse under load. There is a neutral position at half of the throttle travel, I think. Anyone have experience of these?

 

[Miles]

Meanwhile, back to the epicyclic... here's a schematic model to show the general layout:

 

[gogo]

I thought I read somewhere that the distance between two pulley centers is supposed to be 3X the radius of the large pulley. That would allow a sufficiently sized planetary.

 

[Miles]

I'm trying to keep it as compact as possible. The present shaft centre distance is at the close end of the recommended range for these pulley sizes, but it should be ok. I haven't really worked out the epicyclic spec., yet...

 

[gogo]

I was wrong anyway. Here is what I read from the wikipedia on v-belts:

The preferred center distance is larger than the largest pulley diameter, but less than three times the sum of both pulleys.

I would expect that toothed belts would have a different guideline.

 

[HAL9000v2.0]

That's it!!! Do you plan to go as you draw it 1:4 or as we before concluded 1:3 ?

http://www.maedler.de/de/Catalog/P/1643/1618/260/267.aspx

 

[Miles]

Thanks Hal. That's useful.

Mod. 1.5 was a bit over the top....

You mean the ratio difference?

 

[HAL9000v2.0]

Yes. Ratio inside planetary gears?

I think Modul 1 is enough even for the biggest RC motor.

 

[Miles]

The model above is 3.25:1, but not with any great intentionality

Even 3:1 is too much for me....

 

[WTC]

Miles,

There is one small issue with the epicyclic and double sprage clutch arrangement. Unless I am mistaken, it seems If the output shaft is rotated backwards, the transmission becomes a structure (ie it is locked)...your bike will not roll backwards, must be lifted.

Still maybe worth the inconvenience

Britt

 

[Miles]

Unless I am mistaken, it seems If the output shaft is rotated backwards, the transmission becomes a structure (ie it is locked)...your bike will not roll backwards, must be lifted.

Hi Britt,

Yes, that's true. It's the case with all retro-direct systems, I think.

I guess it could be an advantage in some circumstance....

 

[gogo]

You could put a freewheel on the rear wheel hub?

[EDIT] Guess I shouldn't be skipping my naps.

 

[Miles]

gogo,

That would be essential anyway, but wouldn't allow you to roll the bike backwards. A freewheel which overran in reverse wouldn't allow normal drive You'd need a positive clutch. With a retro-direct, the input is in both directions, but you can't backdrive in either direction...

 

[12p3phPMDC]

I think that you can make a variation of this drive and turn it into
an eCVT. Rather than make the planet carrier fixed, it becomes the output,
and the ring gear and the sun gear are both inputs driven by two motors.

If you put opposing spragues on the sun and rings, then when one motor
supplies torque, say to the ring gear, the sun sprague locks, and the carrier
is driven. Vice versa, If you drive the motor direct to the sun gear, the ring
sprague would lock, and the carrier would be driven. Now if you supply
torque from both motors, you get the sum of the two torques on the carrier
and dual redundant operation.

Solomons drive doesn't seem to have one-way bearings.
Niether does the Overhead lift/crane drive. But it seems that the stall torque/currents required to
hold the motors still would cause severe heating problems.
Therefore, putting spragues into the system for holding torque would simplify the system.
But, you become limited in what you can do. i.e. no reverse, no regen e.g. Prius.





I'm gonna start a new thread on this....no need to clog up Miles work. :wink:

 

[Miles]

I've decided on the spec. for the prototype.

1st stage 18t to 72t (4:1) 5mm pitch GT2 synchro belt.

Epicyclic: Sun 48t, Annulus 80t, (1.67:1) Planets 16t (4 off), Modulus 1 pitch.

Sprag clutches CSK 17 (3 off)