HAL9000v2.0
10 kW
+1 but i just want to see how it will go.
2-speed systems (R-D, dog-clutch or derailleur changing)
- Thread starter Miles
- Start date
GGoodrum
1 MW
Miles said:
You know, the first time I looked at these Schlumpf drives, I didn't realize that you could actually shift them between ratios. That's very cool indeed.
I also like the Evo concept, especially for motors, as it is a very simple, but clever system. You could actually implement something like this using Matt's existing drive, but with two sets of pulleys. All you'd need is a longer motor shaft, to fit two small pulleys instead of one, and a longer output shaft to hold both large pulleys. The larger of the big pulleys would get a freewheel, like Matt did for D's drive, and the smaller large pulley would have some sort of engagable clutch. With the clutch engaged, the smaller pulley is driven, which would be "high gear" and the larger pulley would simply freewheel. Dis-engaging the clutch stops the larger pulley from freewheeling and it engages, but with a larger ratio, so "low gear". Dooesn't get much simpler than that, and it would be very quiet. I thought about using a slipper clutch, like Matt custom-built for his recumbent drive, which in theory any way, would allow automatic operation. At rest you would always be in "high gear". If you start out nice and slow, you might just stay in high gear. If you mash the throttle, however, dramatically increasing the load/torque required, the clutch would slip and the big pulley would stop freewheeling and engage, in effect downshifting. When the load subsides, it "shifts" back into high gear.
Another idea would be to use an electric clutch, attached to the big pulley. Still don't know enough about clutches to go much farther with this, but it would be fun to try the slipper clutch idea.
-- Gary
Here's another interesting 2 speed transmission:
http://members.tele2.nl/s_weggeman/CT2.htm
Here's one that uses a solenoid to shift:
http://members.tele2.nl/s_weggeman/CT2.htm
http://members.tele2.nl/s_weggeman/CT2.htm
Here's one that uses a solenoid to shift:
http://members.tele2.nl/s_weggeman/CT2.htm
etard
100 kW
Nice find Fechter, This seems to be what we need, I wonder if you could add that servo output to your throttle board to shift at certain speeds. It could be seemless and automatic I would think. Very exciting stuff happening right here. I just wonder if its not all gonna come back to using a CVT transmission with V-belts because the losses will be the same after we add all the extra complexity that is required. I hope not. :?
liveforphysics
100 TW
I just played with the dewalt 36v cordless drill at work.
It appears the space for the transmission is a cylinder roughly 2" wide by 2" long. It's a 3spd tranny. I'm not measureing the ratios or anything, but it seems to have a wide range. The brushed motor is noisey, but the tranny and gears seem quiet enough to perhaps be helical, and the arangement is definately planetarty with dog-hub shift collars.
I can't imagine you could make a smaller or more compact solution than this little 2" x 2" metal cylinder's guts. Planetary's are so much stronger than anything else in the same size range. Maybe moving these guts into a machined case setup to support all parts on ball bearings, and provide bearing supports for lateral loading off the output for mounting a sprocket would be a hot ticket.
I could just be my plastic-gear-phobia kicking in, but I think something like this would be a more compact and reliable solution.
It appears the space for the transmission is a cylinder roughly 2" wide by 2" long. It's a 3spd tranny. I'm not measureing the ratios or anything, but it seems to have a wide range. The brushed motor is noisey, but the tranny and gears seem quiet enough to perhaps be helical, and the arangement is definately planetarty with dog-hub shift collars.
I can't imagine you could make a smaller or more compact solution than this little 2" x 2" metal cylinder's guts. Planetary's are so much stronger than anything else in the same size range. Maybe moving these guts into a machined case setup to support all parts on ball bearings, and provide bearing supports for lateral loading off the output for mounting a sprocket would be a hot ticket.
I could just be my plastic-gear-phobia kicking in, but I think something like this would be a more compact and reliable solution.
John in CR
100 TW
liveforphysics said:
It's been on my long list. I just don't have any high speed motors, so hadn't looked into it to hack out most of the reduction. The hammerdrill tranny is probably what you want, as the robot war guys use it, so it must be more durable. Efficiency is about the only question, since the gear range seems about right with design output (if I recall correctly) is 400, 1200, 1800rpms. Can pick them up pretty cheap on ebay from people just getting the tool for the batt packs.
John
Looks like you can buy one here:
http://www.robotcombat.com/products/BP388974-00.html
http://www.robotcombat.com/products/BP388974-00.html
John in CR
100 TW
1. If electromagnetic tooth clutches can be found that are spring loaded in a way so their unpowered state is engaged, then that becomes the easy way to go for that EVO type tranny with the overrunning clutch bearings. Then you only use the extra juice for low gear to engage the clutch on the low speed input pulley, while at the same time disengaging the clutch on the high gear driven pulley. Then only in low gear are both belts (or chains) driven, and you've got extremely solid shifting and engagement in small and high torque clutches. Plus only using the extra juice in low gear is no big deal, since it's typically short duration and motor is way more inefficient without the low gear.
2. At the machine shop today I saw an industrial size corded drill with a compact 2 speed tranny, definitely up to the task (if the cordless versions aren't). Probably just a couple of pounds, and most brand name tool parts are available as replacement parts, so price should be reasonable even if you can't find a used tool for cheap. This one was a Makita.
3. How well do these cogged belts retain their shape? I ask for an idea that borrows the over-running clutch bearing concept, but to engage the high gear you flip a lever that puts the belt tensioner in place to engage the high speed pulleys, much like engaging the belt driven mowing decks on typical riding lawnmowers. The belt would need to keep it's oval shape to clear both pulleys when not engaged, and I you couldn't shift under load since it would likely tear up the belt cogs.
I really like those emag toothed clutches, since they're small, can handle big torque, and have an engagement that you can depend on. I'd bet one can be found that only needs power to disengage for design uses where the typical status is engaged. It's just a matter of finding it.
John
2. At the machine shop today I saw an industrial size corded drill with a compact 2 speed tranny, definitely up to the task (if the cordless versions aren't). Probably just a couple of pounds, and most brand name tool parts are available as replacement parts, so price should be reasonable even if you can't find a used tool for cheap. This one was a Makita.
3. How well do these cogged belts retain their shape? I ask for an idea that borrows the over-running clutch bearing concept, but to engage the high gear you flip a lever that puts the belt tensioner in place to engage the high speed pulleys, much like engaging the belt driven mowing decks on typical riding lawnmowers. The belt would need to keep it's oval shape to clear both pulleys when not engaged, and I you couldn't shift under load since it would likely tear up the belt cogs.
I really like those emag toothed clutches, since they're small, can handle big torque, and have an engagement that you can depend on. I'd bet one can be found that only needs power to disengage for design uses where the typical status is engaged. It's just a matter of finding it.
John
John in CR
100 TW
fechter said:
and the 3sp version for the 36V tool is even cheaper here:
http://www.robotcombat.com/products/BP629059-00.html
That, a grinder 90° gearbox, and one of those 2kw inrunners, and you've got a complete unit that you could attach to the seat stay, or like a USPD but with multiple speeds.
John
liveforphysics
100 TW
They make all metal gear grinder gear boxes with 2:1 reductions and helical cut bevel gears.
I think makita makes them.
All helical gears and shaft drive could make for a very slim and compact silent running unit.
I think makita makes them.
All helical gears and shaft drive could make for a very slim and compact silent running unit.
John in CR
100 TW
liveforphysics said:
I've already have some angle grinder gearboxes. A small one with about 3.5:1 reduction and a couple of bigger ones from professional grinders with 4:1. The small McCullough is the only one I've had apart, and I'm certain it's up to the task with those nice helical gears and design usage. That's another thing on my project list. All the grinders I've found spin the same direction (no 2 speeds unfortunately), and they'd have to be a left side drive due to the 1 direction usage and no retro-direct.
John
John in CR
100 TW
This might be the easiest other than plug and play ability of the emag clutches. My issue with the different gearboxes has been the linkage and shifting mechanism, so if it's not off the shelf then it's unlikely I could do it, other than something with a chain and shifting similar to a bike in some way. Sorry I can't draw.
Start with a toothed clutch. It's natural state is disengaged from the pulley due to a lightweight spring. Attach a strong doughnut shaped magnet to the clutch. Then your engagement mechanism is simply another strong magnet with same polarity facing the clutch on some kind of lever. Push the magnets close to each other and the repelling force compresses the spring and engages the teeth. What makes it so easy is the parts don't touch, so the issues of spinning things needing a reliable interface to stationary shifting mechanisms goes away.
John
Start with a toothed clutch. It's natural state is disengaged from the pulley due to a lightweight spring. Attach a strong doughnut shaped magnet to the clutch. Then your engagement mechanism is simply another strong magnet with same polarity facing the clutch on some kind of lever. Push the magnets close to each other and the repelling force compresses the spring and engages the teeth. What makes it so easy is the parts don't touch, so the issues of spinning things needing a reliable interface to stationary shifting mechanisms goes away.
John
John in CR
100 TW
Miles,
Yes, that's great! Nice follow through with belts and gears naturally engaging in opposite directions. Gear for low and belt for high speed/low torque...best of both worlds and quiet in high gear. Give me simple every time.
John
Yes, that's great! Nice follow through with belts and gears naturally engaging in opposite directions. Gear for low and belt for high speed/low torque...best of both worlds and quiet in high gear. Give me simple every time.
John
TylerDurden
100 GW
Yes, Miles... that is elegant.
GGoodrum
1 MW
Using gears for low is fine, but they'd probably have to be metal-on-metal, wouldn't they? If the belt drive was low, like we're doing on Matt's drive, you might be able to use a metal/composite plastic gear for high, as it would be less torque required.
John, I like the elctro-magnetic clutch idea for the Evo-style 2-speed belt idea as well, but where do you find something off-the-shelf?
-- Gary
John, I like the elctro-magnetic clutch idea for the Evo-style 2-speed belt idea as well, but where do you find something off-the-shelf?
-- Gary
John in CR
100 TW
GGoodrum said:
When there's a service whose company mission statement is "to organize the world's information and make it universally accessible and useful.", finding anything is so easy. I love Google. Here are some emag clutches, which I'm sure are commonly used in automated systems, and Miki Pulley probably has lots of other stuff we can use http://www.mikipulley.co.jp/en/product/model.php?id=pcbts&code=001 I've submitted an inquiry to locate distributors and get a current price list.
John
John in CR
100 TW
Miles said:
A little more digging has me thinking to isolate the drives for high gear, put an emag clutch on the low speed input pulley and a spring loaded emag brake on the high speed output pulley, but have that brake acting at the clutch. I think some of those brakes act in the manner I want...engagement without power. That way you only have a power drain in low gear, and on the 2nd gear side the sprung brake/clutch uses electricity only to hold the disengagement for high speed. I think the efficiency hit for both belts spinning in low gear is okay, especially since the always turning high side gives you regen and electric brakes potential in high gear, where I think you want it anyway. Maybe those friction clutches are ok on the EVO, but my gokarting days as a kid makes me want to shy away from friction clutches...too quick to wear out.
John
John in CR
100 TW
Miles said:
Here's the complete assembly instructions with pics, that really helped me understand exactly how the clutch mechanism works. http://scooter.wikia.com/wiki/Evo_two_speed_transmission_assembly
John
GGoodrum
1 MW
The pictures/instructions help, but I'm still having trouble understanding how the pulley clutch is engaged. Does the cable somehow pull that arm so that it puts pressure on the end of the clutch/pulley assembly?
The emag clutch seems much easier to understand, but I think you would want it so that the clutch is energized when in low gear, which will be a lot less time than high.
-- Gary
The emag clutch seems much easier to understand, but I think you would want it so that the clutch is energized when in low gear, which will be a lot less time than high.
-- Gary
