2 speed transmision evolution----

Thud

1 MW
Joined
Aug 20, 2009
Messages
2,690
Location
West Michigan,USA
Hello from the shop of Thud.
I have made some desisions regarding the design of my 2 speed transmision & its application to my light EV comuter. I thought I would show some progress & hear any comments.
The "To Gear or not to gear" thread showed me that there are more than a few guys looking for a way to optimise their set up & extend the operational envelope
That said, I guess it is no secret that the majority of the builds I see around here (all of mine so far & in the future) are kinda illegal. so I can only recomend that you submit to your local ordinaces :p

Here are some design goals:
#1 maintance free
#2 quiet operation
#3 ease of integration

#3 is the puzzler-The bigest issue with any of the RC reduction units & this unit also is getting it to fit somewhere. This is really not an issue for myself, But I have had several inquiries asking for units for bicycles & still am looking for a better solution. There is no simple "bolt on" given the miriad of bike frames out there. I apreciate the input about size, but in reality this is the lower limit I am comfortable building & offering.

As of now the design has evolved into a closed case to allow an oil bath for all components. #1 acomplished! This should also reduce the whine inhearant in the rc reductions.#2 addressed! We the option of mounting the motor anywhere about the case with a simple bolt on mount. Orentation can be vertical or horizontal.
after optimizing sizes & reduction senarios I am right back where I started with the 1st prototype. roughly 2-1 reduction in low gear & 1-1 in high. 87% reduction to be precise. The rest of the gearing is acomplished in the primary & final reduction.

Several have asked to add reduction in the unit but the size grows exponentialy just getting to 3-1 in low range. Gears would solve the size issue but raise more design challenges regarding shock loads & set up manufacturing costs.
I plan to use this current version on my comuter Bike this summer. I will also be testing it with a shimano 3 speed hub & 6kwatts. A true EV comuter. The pedals (should I include them) will be for emergancy only. I am very close to starting the production of the units. Like Mud, I am not looking to get into full time production, Just off setting the devlopment costs & looking to make the e-bike experiance that much better for ever one.

Here is a foam pattern I will cast in aluminum. the finished tranny will weigh apox. 3 lbs not including the motor.
the bulges are bearing pockets. it will have a pr of .5 x 1.125 x .312" ball bearings & a pr of

I am certain the design will get another round of refinements before I am satisfied. the built up foam pattern is still a bit time consuming. Thats all for tonight. I need to switch gears & get My motor build back on track.
Todd
 
Very cool Todd!
 
8)
very

thud can you make me a light sabre ?
promise i wont go on a jedi killing spree . just a few politicians maybe a corporate lawyer or 2
 
It is good to see someone with the capability to do casting. That part would b estupid expensive to machine from billet. ;)

Great, so far..........

I completely agree about the size when adding reduction. I went through the same thing while looking into this. Multi speed transmissions are not as easy to make as it may seem. I am glad others have knowledge in that area.

Looking froward to the finished product!

Matt
 
johnrobholmes said:
Show us the casting procedure please! Looks cool, and just about the right size.

ditto. It looks really small, can't think of a bike frame it won't fit. Your the man Thud, I enjoy the methods to your madness. :wink:
 
Todd

Several have asked to add reduction in the unit but the size grows exponentialy just getting to 3-1 in low range. Gears would solve the size issue but raise more design challenges regarding shock loads & set up manufacturing costs.

Very impressive build!

There might be a way to get more reduction out of this box without increasing it size. I read your other post about this transmission and studied the pictures. It appears the sprockets on the input shaft are fixed and the ones on the output shaft float on bushings. High range uses one set of sprockets and low range uses the other set. What if you switched this set up around so the input shaft and and the output shaft would be the same. The pulley would drive the fist sprocket. These two components would be connected together and float on the output shaft. To select high gear your gear selector now located on the input /output shaft would engage the sprocket and this would give you a 1:1 ratio. to select low range the gear selector would move over to engage the other sprocket. The power is driven from the first sprocket connected to the input pulley to a second sprocket located on a jack shaft. This jack shaft is in the same location of the original output shaft. The driven sprocket is keyed to the jack shaft. The sprockets would be the same ratio as you original low range. The power then is transferred from the jack shaft to another sprocket which is also keyed to it. This sprocket is connected to the final sprocket located on the output shaft also the same ratio as the original low range. Your original 2:1 is now 4:1 without increasing the size of the case.

What do you think?
 
I don't think a bigger ratio difference is what's needed...

You could have a dual concentric input/output shaft and include another reduction stage common to both, though (doubling back).... That would be cool.
 
I think verbal/written comunication regarding these things are confusing :lol:

I have tried to visualise your idea ( something I am normaly good at) here is where i get hung up:
What if you switched this set up around so the input shaft and and the output shaft would be the same.
Are you suggesting same sized sprockets, mirrored on the shafts? (I ran several iterations of that set up, it increases the overall size of the unit, as I need a minimum 26t for the dog hub to engage)

Unless I totaly miss understand your description (a picture is worth a k- :p )
The ratio spacing (hi vs low) of the unit can be wide'nd simply by pairing like sized sprockets on the shafts.(actually solves a lot of head aches on center spacing of the shafts) example: 12-25 on low & 25-12 on the other thats -2.08 in low & +2.08 in high. for a total of 4x variance in the system. (actually stepping up verses reduction from my way of looking at it)

The issue is keeping the primary & final reductions as reasonable as posible. The above senario would work if there is abundant power. I see this being utilized by the Higher rpm motors the general poupulation here is gravitating toward.
By more reduction I refer to the unit as being 4-1 in low & 2-1 in hi, efectivly still a 50% reduction step in the drive line.

From my testing with the tiny 50mm motors, the 90% increase feels ideal.

edit: re-reading makes me think you are on a differnt plane form my original though process:
The pulley would drive the fist sprocket. These two components would be connected together and float on the output shaft.
I think I see where you are going but really need a drawing to better visualize the marige of all that verbage. I could start asking questions, but I will wait to see a picture instead of trailing off on a tangent. I am all for making it as simple as it can be & no simpler!
 
Yes, I was suggesting a mirrored pair of sprockets, but if you limited to a min. of 26 teeth then it probably will not work. The smallest pair of sprockets you could use would be 26 / 37. This would give you your original 2:1 low range, 1.4 x 1.4. The 37 tooth sprocket would be just over 3" in dia. Probably too big to fit inside.

What if you switched this set up around so the input shaft and and the output shaft would be the same.

The input and output shafts would be in the same location, not the same, sorry for the confusion.
 
Up date:
My original plan was to show some castings tonight......best laid plans of mice & blabla bla...
Truth be told, I wasn't happy with the foam pattern assembly. Too many parts & PIA of assembly After re-evaluating, I have re-designed the unit & lowerd the parts count & cycle time for the hot wie cnc I cut the patterns with. this also gives me an option to use extruded metal on the covers so I can anodise them if I so desire :mrgreen: (I will be casting these & paint or polish will be the norm for the first batch) I would have made the casting goal but I had a controller issue on the cnc & had to service the baby for th 1st time in 4 odd years of trouble free service ( just a bad resistor it turned out)

This is a whiped together super ghetto cnc operated hot wire bow saw I use for cutting patterens with. capacity is 6"x17"x 4" (stepper's & drive kit from hobycnc,controller is the free EMC2)
PC270009.jpg


parts released after a cut. (super smooth & shiny :p )
PC270011.jpg


this photo looks daffy & really deep (its only 2" deep)
PC270012.jpg


I have 5 cases assembled, will be preping for casting tomorrow night.
I am playing with some video editing software, I will try to put togeter a short series showing the actual casting & shake out of the parts this week. (assuming I can get a shot with my chepo camera & a tripod)
I will be riniging in the new year by working in the shop if all goes to plan!
Happy new year every one!
 
Thud said:
I have 5 cases assembled

If these are for selling and all aren't spoken for i would gladly put my name down for one
Thud?

Very keen to see these finished best of luck mate

KiM

p.s anodised Blue i think if poss?
 
OMG! Your DIY CNC foam cutter is so badass! Fantastic work Thud! You are the DIY guru!
 
liveforphysics said:
OMG! Your DIY CNC foam cutter is so badass! Fantastic work Thud! You are the DIY guru!

ditto.

im thinking of going to Michigan just to get a tour and soak up what ever knowledge i can . that and use his secret lab to build my lightsabre.
 
Thanks guys,
Its not all sunshine & roses for the diy'ers, Lots of failure goes along with this.
Example......the first 7 attempts at doing these lost foamie trany cases!!!!! :(
the good news is I finaly figured out my gating issues & got the pour temp correct to make what is undoubtly the most difficult single part casting I have attempted. the thin sections & dis-proportionet depths make this a tricky lil bugger.

Here is what it looks like just out of the sand & with the gates cut off. still totaly "in the rough"
P1040013.jpg

I was testing gating options so this one wasn't invested in plaster before the pour-just straight sand(stuck all over it)
P1040012.jpg


you can see the great detail you can get. this pattern was really rough as a test part. too many fails to put much love into test parts
P1040014.jpg

& this is the side that makes the unit such a bear to cast, A long thin flat section far from the thickest area of the metal flow.

the original 5 (RIP) I put a lot of time into making these perfect before wrecking them with 1/2 fill
P1020005.jpg



heres a variation of my intended motor mounting/cover plate
P1040015.jpg

I am toying withthe idea of machining these covers out of billet for the purpous of anodizing (casting alloys are crap for anodising) just to give a bling option.

Time to make some more patterns & re-fill the propane tank :twisted: till the next time. Thud
 
Thud said:
just to give a bling option.

:: puts up hand for one with CNCed alloy cover:: preferably in bluez with a couple of lightening holes...actually...i can add them myself :mrgreen:

FrankG is an OUTSTANDING lost foam metal caster also Thud, when you have spare time checkout his work on TheWorkshop.ca If you haven't caught it already. He may have some suggestions if your have any more issues?

I have a nice 20KG LPG container waiting to be converted to a furnace, my first furnace was a baby, melting metal and casting is something i intended to be back doing by now unfortunately things dont always pan out how you plan do they :: sigh ::

Best of luck anywayz mate

KiM
 
Thanks FrankG,
I post at BYMC under the name "Todd" for the last 6 years or so, I am a fan & really apreciate your site as a referance point &I direct folks there all the time.

Main problem - getting 100% fill.
The bulk of this case is only .187" thick to minimise weight. (I increased the wall thickness while tesing gate configs)
I think the main issue was pouring too cold, into loose sand that was also freezing cold at 18 deg.f. (effectivly pouring into a giant "chill") Once I minimised the foam gating (I am re-purposing toilet paper cardboard tubes now for the main gate/riser to maintain heat) & warmed the sand up indoors, it became much easyer. I am using loose sand for these patterns. & will coat the production parts ala' BYI technique.

I have a bunch of video clips I need to compile & edit before posting it on you tube for any one interested in looking at it.
(another FrankG insperation)
AJ, metal casting is an addictive activity in itself. Its funny how you look at everything differently once you know more methods of fabricating stuff. A guy with tools can have anything he wants!
 
ive had about 4 different things written to post to this thread for the last day or so and have not done so .

its just awsome .

i feel like the slackjawed wide eyed kid at the back of the class that keeps raising and lowering his hand with a million questions but realizing they are answered if i read a bit so i lower my hand and listen harder.

If i had found this kind of stuff when i was racing jetskis who knows where id be today. bottom plates, high compression heads, light weight total loss flywheels i could go on and on.
 
who let the dogs out?
P1120003.jpg

fruit of tonights labor.
The deeper i get into this project the more I remember why I hate production........the up side is: repitition builds speed.
progress continues.
 
So.. how much were these, again? I doubt I could afford it for a while, but a pair of them to run the front wheels of the trike would be a much better solution than what I have come up with so far. I don't have the machining capability yet to build one of my own, or I would try!
 
Thud,

Nice job man...!

I've been thinking about dog clutches alot lately.

Do you know how to flame harden the dogs or just harden them?

I imagine you would need an oxy-acetylene setup. About all I think I know is, is that you
hit the surface with the flame, probably a rich sooty flame, and then quench
the steel in oil or water..and that should surface harden them.
 
Back
Top