2-speed systems (R-D, dog-clutch or derailleur changing)

[John in CR]

BTW, do you have any of the designs you've shared in the build stage yet, or just sharing?

No, I haven't actually built anything much, lately....

I understand. Once you have a good ride you spend more time riding and thinking about how to do the "perfect one" next, instead of building another compromise. Plus yours are so highly developed already that the improvements are harder to come by.

John

 

[Thud]

After reading through this thread I am still curious.
Has any one has built a small 2 speed transmision & tested it yet?
Is the idea still viable for a "legal" bicycle? It seems Gary's direct drive w/a 3 speed hub may have brushed this aside.
Over powered (the really fun ones)Bikes & those building insurable/licenced electric motorcycles may still seek an option.

 

[Miles]

I'm still working on retro-direct designs. No prototypes yet, though..

I don't think the ideas are any less relevant - I've had a direct drive bike with a 3 speed hub for 2 years and I'm still interested in alternatives

A separate automatic 2 speed gear, for the motor alone, has its attractions...

 

[John in CR]

I'm still working on retro-direct designs. No prototypes yet, though..

I don't think the ideas are any less relevant - I've had a direct drive bike with a 3 speed hub for 2 years and I'm still interested in alternatives

A separate automatic 2 speed gear, for the motor alone, has its attractions...

Miles,

What is your biggest issue with the direct drive going into a 3 speed hub? Is it the narrow gearing range, or operational problems like slipping out of gear, or ? I ask because I'm about to try a couple of variations of that route, and avoid any know pitfalls.

Also, though it's OT, are the 90%+ efficiency and high power that Randy claims for alternator/brushless motor conversion realistic. Those PMA's are pretty cheap, and installing hall sensors in them seems fairly straight forward, so 10-12lbs for a durable motor capable of 5-10kw is appealing.

John

 

[Miles]

John,

No particular issues with it. I certainly don't need a wider gearing range. You have to ease off on the throttle a bit, when changing gear.

I like the idea of an automatic gear change for the motor, but not for pedalling. Hence the need for separate gearing systems.
With separate gearing systems, there is some continuity of power through each individual changeover.
A retro-direct 2 speed could have the gear changing handled completely by the motor controller.

The only data I have is for the motor that I used (see below).

If you mean the Hydrogen Appliances PMAs, there is no information, yet, on the quality of material used for the stator core or the thickness of the laminations used.

 

[Thud]

Wasn't sure where to post this, so I guess I'll start here. I scaled down a 2-speed dog transmission I originally designed for a Mars e-motorcycle project.

Question-Does any one have data (or an off the cuff idea) of the ideal ratio spacing for an RC build?
(I have only experienced 22mm motors on my rc planes) I have a couple motors for testing. An 1800watt 380kV & a pair of 2100 Watt 270kV (both Turnigy, I assume those are peak ratings) I am limited to 8 cells by the CC-Ice controller I am testing with (28V nominal).

Here is what I am doing:
primary reduction is GT2 5mm pitch, 15mm wide, 3-1reduction
low gear = 2:1
high gear=1:1
final drive reduction #35 chain & sprockets

I am using #25 chain & sprockets in the transmision so I may re-configure the ratio's to get the motor happy

Fruit of tonights toil

still need to cut splines in the secondary shaft

The engagement dogs will be pressed into the center hub.The completed transmision will fit between the rear tire & seatpost on a salvaged bmx aprox 2.5"w x 3"h x6.5"l with motor mounted.

(I didn't take the time to model it 3D but I have a very jumbled 3-view in AutoCad)
Here is a photo in my hand for scale (thats the large 26T sprocket) & the bike its going on 1st

More pics of this transmissions guts:
https://endless-sphere.com/forums/viewtopic.php?f=28&t=13731#p204269

[moderator edit with Thuds permission, added link]

 

[Miles]

Question-Does any one have data (or an off the cuff idea) of the ideal raitio spacing for an rc build?

Hi Thud,

For a given motor power, it varies with the speed range wanted and the steepest hills to be climbed.

I think your 2:1 ratio difference is probably about right for most peoples needs - it's what I'm intending to use for my first prototype, too.

I think you should go for two reduction ratios, though - it seems a waste to have a 1:1 stage.....

I'm looking forward to following your project

 

[Thud]

project update:
Found a couple of hours today to complete the shifting hub for the 2 speed build.
pictures speak far better than my typing skills. I thought I would try a video with a free camera I picked up a while back.

[youtube]C0aiUBf1yGo[/youtube]
The goal is to finish the case to house the unit & get it mounted this weekend. The weather forcast is rain tomorrow so I have have an excuse to play in the shop all day.
Miles, the ratio's are 1.15-1=high gear,1.86-1= low gear
your statment regarding 1-1 being a waste made me smile.
Given its in the middle of 2 other reductions I never considered it anything more than than a switching mechanism to modify ratio's
The only way I could reason calling it a waste would be engineering in a redundancy. Just to be clear, the statment seemed to be an emotional responce & I enjoyed the moment I read it.
I am looking foward to my 1st "E-bike grin" I have read so much about here the last few weeks.
Any comments or suggestions are welcomed. I am impossable to offend.

 

[spinningmagnets]

Thud, you are my new hero. This is almost exactly the way a 2-speed moped transmission from many years ago works, so it is a well-proven idea.

It was discussed for a 48V E-motorcycle that spent most of its time at 40-mph, but needed to travel a for a short distance at freeway speeds. I if you use only a single gear that tops out at 65-mph, the lower-speed accelleration would be poor and would draw massive amps and heat. A two-speed transmission would be a huge improvement.

If you make the dog-gears a much larger diameter, they will have a lower relative RPM, and the dogs will be much easier to engage and dis-engage. The sliding "two-faced" dog is in-between gear #57 and #59.

http://endless-sphere.com/forums/viewtopic.php?f=10&t=9695&start=30#p167538

 

[johnrobholmes]

That is also the same type of system in an Emaxx RC car tranny. It would be easy to get large gear changes with just two speeds as well.

 

[Miles]

Miles, the ratio's are 1.15-1=high gear,1.86-1= low gear

That's better

 

[liveforphysics]

Great work on cleverly making a dog shifted 2 speed.

It's built the same way the gears in my racecar (and all proper racecars) are shifted, with a small exception. Rather than having the dogs fit snugly into the recesses in the gear face, the recesses are cut with at least double the width of the dog to give a much wider time interval for the dog hub to get slid up against the gear before it begins to carry load. This helps prevent the tips of the engagement dogs from being mushed over at the expense of having an additional 20-30deg of backlash in the tranny. It also enables consistent shifting even in situations with a large speed difference between the gear face and hub.

 

[Thud]

Thanks guys,
LFP, (you are one of my heros ) I considered the clearance issues you mention, there is aprox. 2mm of angular clearance but I suspect shifting will be a dellicate exeresize.

off topic:
I also suspect Matt, yourself,I & most of the regulars in this non-hub section would have a blast at an e-bike ralley...make that an e-bike performance contest. I am a 20+ year vet from the motocross scene here in mi.(racing is life) I am fighting the urge every day to build a twin,tri..HELL! quad motored power unit that fits into a carbon fiber framed moto crosser. 3 speeds (need a creeper gear to get to the starting line) I could go on but I am heading to the shop.

Back on topic:
I have many concerns about the ultimat longevity of this prototype given I am making no effort to properly harden any components that should be.

Spin, you flater me
The 2 design criteria for this unit is to: #1 shift & #2 fit in a tiny space of my salvaged (kids) bmx bike.
Once (if) this toy works I intend to build a much refined version for a daily comuter. (46 mile daily round trip)I have a ton of ideas (none original so far )& I am getting to the age where I actually have time to putter with this kind of stuff.

 

[Thud]

Got the case built today, seems to work as planned. I have the shift-fork actuating rod left long for now, I plan to move the torque arm inside the case to keep it streamlined. Last items to fab are the chain tensioners from some parts I have in the scrap bin. A couple lexan covers to keep from flinging any lube about & the cable stays for the shifting.

The current plan is to add another set of brake levers to the bars in a pull/pull senario to select high or low on the fly.
Then I'll tear it down & anodize the cases red/the pulleys black. This has been a super relaxing project to build I can't remember having this much fun designing & fabbing in a while. Here are some photos to stop my rambling.

[moderator edit for spelling, with thuds permission]

 

[spinningmagnets]

Its looking good! "IF" you wanted to have a freewheeling Bottom Bracket (BB), there a link in the resources stickie about an inexpensive adapter kit to covert your pedal-axle to the three-piece, and then there are links for a freewheeling crank thread there too.

If you lean into a turn, and the pedals are driven, it can cause a bad wipeout when a pedal strikes the ground (if I'm reading the pics right). I'm liking the gearbox, looks like it can be easily scaled up for E-moto size...

PS, You should probably start a new thread just for your bike build (it deserves one!), and put a link in this thread to make it easy to find it.

 

[Miles]

Nice work, Thud.

I wonder how feasible it would be to incorporate an electric actuator for the gear change?

 

[johnrobholmes]

It would be pretty easy I bet. I have a few small linear actuators here that would fit the bill, and allelectronics probably has a ton.

 

[etard]

Thud,
you have got to start a new thread along with a video of the tranny on action!! That thing is sweet and I imagine there will be much interest here in the non hub section (aka, the future)

Question: from looking at the early moped trans it looks like the sprockets have more holes for the three dogs to engage into, is there any advantage either way? Also, it seems this would be just as easy to do with belt gears, is there just not any off the shelf gears for this, or is chain the only way to go? Won't chain at this rpm be kinda load too?

I am so happy to see physical progress on this thread, thanks for your noble effort. Please detail your build more for us Thud.

 

[gwhy!]

Thud,
you have got to start a new thread along with a video of the tranny on action!! That thing is sweet and I imagine there will be much interest here in the non hub section (aka, the future)

Question: from looking at the early moped trans it looks like the sprockets have more holes for the three dogs to engage into, is there any advantage either way? Also, it seems this would be just as easy to do with belt gears, is there just not any off the shelf gears for this, or is chain the only way to go? Won't chain at this rpm be kinda load too?

I am so happy to see physical progress on this thread, thanks for your noble effort. Please detail your build more for us Thud.

yes please, look like it will fit the bill for many peeps here.. well done.

 

[Thud]

Alright,
Being a noob around here, Any suggestions a where to put this new thread? seems to be some overlap & every one here is really pushing the limits as to what a bicycle is.... I have landed in the non hub section cause I can relate to the builds as a motocrosser & a lighter paltform of the bicycle seems a logical step in electrical transportation. Given the talent pool I am discovering reading back a year or so, I am a bit suprised that this hasn't been done alredy, I suspect it has & I haven't found it yet, that said I will answer some questions & attemp to steer this thread back on topic.

Miles, it would be no problem. An actuator would eliminate some of the parts I included (the detents for positioning the hub) given all the electronic prowess about, I speculate a load sensing auto shifting unit could be easily be designed & built.

etard,(love that name BTW) This is a plagerized version of a typicle constant mesh transmision lmited to 2 speeds. as far as noise from the chains, it will not be silent(which will be a goal on another build soon) I suspect the usual chain whine will be emited.As for more holes, I see no advantage over just an elongated hole LPF eluded to in his reply. (the female parts are 2mm over the dog dia. not clear in the photo's) any "more" or "bigger" is taking away from the integrity of the part. I used the C&S for costing considerations. They are way cheap compaired to an HTD belt system, easily modified & the #25 is the smallest I could buy surplus, as size was a major consideration on this build.

I am glad to see some interest in the transmision.....but I almost threw the idea of fabbing this thing out the window after watching & hearing Garry jamming gears on his direct drive S-A 3 speed (my e-bay unit just arived this week ) seems like a hub could be fabricated to accept the 3 speed mechanism & offer disc brakes & a dual freewheel. Easyer for some I suppose but well within the skill set lurking about.

My thoughts on a final design:
Use real gears (they would be way more compact,& gearbox design is very flexable)
As a K.I.S.S. devote' I would opt for a manual shift.(gives the operator a more conected feeling....specultive)

No such thing as a "universal" bike (was there ever?) so mounting is allways the bigest hurdle. I would think a bolt on design would require a thourough discovery of the bottom bracket. as this seems to be the most universal part between bikes these days. This should get some conversation going.
Apoligies for the mini novel, I await the replies.

 

[Miles]

My thoughts on a final design:
Use real gears (they would be way more compact,& gearbox design is very flexable)
As a K.I.S.S. devote' I would opt for a manual shift.(gives the operator a more conected feeling....specultive)
.

Ok....

This is a gear system for the motor only. Presumably there'll be another gear system for pedalling? I don't see how having two separate systems to control keeps things simple for the operator? Or, it's just a token pedal input?

The reason I'm pursuing a retro-direct solution is that it makes automation a piece of cake, well mechanically at least

 

[Thud]

You got me Miles,(too easy..no points)
Yes, for the motor only. The peadling side is to remain isolated in my thought process......unless you often use them in concert, the human side of this is pretty well solved already. I did buy the largest chainring I could at the moment when I refurbed the bicycle intending to help the motor out as much as posible. (48-16 on the pedals) But me thinks that is not the point.

I really like your retro direct design also Miles, I think a centrifical (spelling?) shift could be utilized. not as sophisticated as an electro-mechanicle device but near bullet proof. I have seen some incredible performance from centrifical clutches also in the last 2 years. I like this design & I am sure you see its aplication potential.
http://www.rekluse.com/zstartpro.shtml
but for size (small) & strength it is hard to compete with a gear to gear transmision.

 

[Miles]

I'm really grateful to you for stirring things up a bit, Thud.

As I see it, either the pedals and motor share at least some of the gears, in which case manual changing is the best option or, they have separate gearing, in which case it's better to have manual changing for pedals and automatic (with override) changing for the motor.

There are pros and cons for both systems....

 

[spinningmagnets]

Heres the "F&S Torpedo". Its a 2-speed hub that can be had as a kick-back shift, or an automatic shift based on cetrifugal weights working against springs. I'm sure the weights and spring-tension can be adjusted to change the shift-point. I think I read somewhere that it was developed as a cheap option for cargo bikes and tandems (or, at least, cheaper than a 3-speed cargo hub), so it may be very beefy. For bicycle levels of power, it might be possible to use it as a transmission for an electric motor...

http://hubstripping.wordpress.com/fs-torpedo-automatic/

 

[Thud]

Spin,
I missed your post on page 11 some how. I looked up that hub in your last post & they seem to be out of production for a while now. please don't tease
I do think for a bicycle this will be the hot set up. I havn't tore down the 3 speed shimano e-bay hub I picked up yet, but I am thinking it would be realitivly easy to mod the unit for a double freewheel & still fit standard frame spacing.
The Skyway wheels used on this build were fairly simple to add an additional drive sprocket on the right side inside of the existing free wheel position. The #35 does not freewheel but I included a netral position in the gear box. The motor drive & pedal drives are totaly independent. chain clearance is an issue more because of the floppy bicycle chain. I need to stop by the bike shop & look at heavy duty bmx chain to see if it is better suited.