Thuds 1st build

Mike,
Really not as much as you would think.
surplus center parts;
6 sprockets,chain,master & 1/2 links. rounded up $40
vxb bearings $12
sdpi (gates sprockets & belt) $47 (most costly part not considering labor)
misc screws & bolts $10
value of stock on hand $15
add shipping charges $30
Thats $155.00 ish in parts
Estimated 30 hours of actual design & build time on the prototypes.
So the suggested retail price for a prototye unit would come in near 1K.
I could whole sale them for way less. Not that I need any additional projects ATM :p
 
Come'on thud... you need another project = )_

Seriously though... I'll bet there would be a good amount of demand for those, how complicated is the machining process? How many axis mill required or can it be routed?

Honestly I wish I could say one of my prototype drive systems came in that low of a cost... (my prototypes always end up right around where your retail would be - multiple screwups=usual cause) and youve acheived a dual speed transmission!...

Just a thought... with the lever design, how much force/torque is required to actually "shift" if it is not too much a small linear actuator or servo motor could be used to control this from the bars? That would be slick and convenient = )

-Mike
 
Sorry but as I review your parts list and cost breakdown... if the belt and pulleys are the most expensive part, would it be feasible to design first stage using gear drive or would there be too much torque or speed for gear design? Just thinking about the cost for gears vs those belts/pulley, prob. 1/5th the cost?

Ive been thinking about your design and really 2 speeds is plenty at the reduction level normal cruising speed and a bailout gear for massive hill climbs = )

Combined with controller supporting delta/wye and a manual delta/wye switch - and you have an effective 4 speed transmission = )_

2 speeds, light weight > 1 speed (geared or direct, light weight or heavy)
4 speeds, light weight > 2 speed (geared or direct, light weight or heavy)

Finally an inarguable reason for going with a more complex RC/BLDC design!

Regards,
Mike
 
Mike,
Idealy, I think the whole transmission could/should be built using gears. Once ratio spreads are settled upon, the unit would be smaller/lighter,& more efficiant.

As an add on to the existing unit, a simple re-arangment of components would allow a primary gear reduction with catolog parts (greater reduction flexability also)
I didn't consider gears for the tranny build, dismissing them early on for complexity issues in the twin shafts. but I confess I didn't reaserch too deeply once I found sprockets for less than $4. I will have to revisit the concept if anyone shows sincere interest in haveing a batch produced.

Has anyone bread boarded a D-Y switching controller yet? (I know some hub motors are using it)
 
thud,

I think about a dozen of us or so have some form of bread boarded delta wye, some relay or contactor based and some using FETs or so forth. It's a shame we don't have the infineon source, we could incorporate a secondary addressable AND gate and simply switch between Delta FET output or Wye FET output. This would be the most efficient method I have been able to dream up for performing Delta/Wye switching.

The problem is efficiency... for delta/wye to work properly requires either 3pdt relay or the electronic equivalent. No matter how many times I revisit this, a FET based design will require minimum of 12 fets to function the same as a 3pdt relay. Which is why I always return to incorporating it into a controller as a second driver and use the same PWM lines to drive it, just address the gates via the AND gate to select if Delta/Wye and also to create dead time and over voltage/speed/current traps.

Gary is working on one - may be working, fechter did this a while ago (I think)...

The basic circuit is nothing more than a 6 Phase Full Bridge which operates as 2 independent outputs of 3 phase full bridge.

Using gears would rock and really could decrease the form factor and weight... noise would be an issue as Miles points out but we could look into some of the more exotic materials, plastics, fabric laminate bakelite, etc style gears - no need to go metal but no desire to use nylon = )_ I believe Reid is our resident expert on old gear tech...

Would noise really be an issue... wouldn't we want to enclose the gearbox and run them in oil/grease? Wouldn't that drastically reduce the noise of even metal gears? (between the encasement and the oil)?

-Mike
 
I know running helical cuts in a box dramaticly reduces the noise of a gear trane. (there is that pesky money problem again Miles) I do beleive that fiber re-enforced polymer gears could be designd with the correct strenght for application.(tooling costs soaring) At minimum, a shock absorpton feature or tourque limiting clutch would be needed to protect it from Hackers & over volters who shall remain namless :twisted: well.... here we are again. Deep into parameters that need to be identifyed & prioritized to continue on with a design.
these are questions & desisions for a large production run......I don't see that Happening any time soon

On the other hand for a short run of really quiet 2 range transmisions, (as that is a requisite that will make these steath)I see no reason to look further than the syncro belts. The size will grow a bit. but anyone willing to mount one on an Extracycle, behind the seat ala GrinHill,or fab a custom frame (go Jester GO!) will not be constrained by any of the down falls the unit presents.

As for the controllers & delta-Y switching? how do they do it on the hub motors? All the design language used describing gating &such may as well be Chineese as I am just begining my foray into electronics. I did assemble the driver board for my bench top cnc (big deal) but it is all magic if I cant see a mechanical relationship of parts interacting. I am working on it slowly.
 
Thud said:
protect it from Hackers & over volters who shall remain namless :twisted:
Click to expand...
whoa... nevermind, i don't want an iZip = )_

Seriously though, tooling costs dont need to be that high... how would extron be for a gear material, it holds up on carts running 10K rpm with all the friction of metal chain to extron gear I would have thought they would wear more... nope, dont get warm either and the material is machinable with little force...

Am I wrong (probably) but aren't helical cut gears the most inefficient type? Or is it a matter of variance and clearances between mating surfaces?

-Mike
 
mwkeefer said:
Seriously though, tooling costs dont need to be that high... how would extron be for a gear material, it holds up on carts running 10K rpm with all the friction of metal chain to extron gear I would have thought they would wear more... nope, dont get warm either and the material is machinable with little force...
Click to expand...
Aren't they injection moulded nylon with long fiber fill? Kevlar and something else? Extron is just the brand name for the sprockets...
 
see edits in post above
 
Extron are a Kevlar composite... sorry.

thud,

Delya Wye is nothing more than three pole double throw switch with center off.

Even more mechanical...

On hub motors and RC motors with this mod now... you just seperate some of the wire inside the motor and run 3 more phase power wires (the opposite end of the existing, common 3 wires) the controller is connected to the switch which electrically makes the two termination schemes. There are a ton of simple diagrams (one pole motors) for easy understanding.

-Mike
 
Mike, I get the hook ups to make the switch.
performing the switch with ellectronics is where my confidence falters. Think of me as a 47 year old new born. :p I am prolly not stupid, I just never had a need to delve int such things before.
on a sensorless design the mechanical switch is NBD theroy wise, but dont you jumble the timing in a conventional sensored set up? & going to more than 3 phases, whoa...that just confuses my brain exponentialy :?: :!: remeber, its not the basics, its the details(again I would think adding sensors to match the switch & switching them at the same time would be in the realm of doable) Any one here riden a bike 2 speed with the delta wye switching? (other than the hub motor set ups) seems as good a solution a a gear box all things considered.
 
Thud said:
Any one here riden a bike 2 speed with the delta wye switching? (other than the hub motor set ups) seems as good a solution a a gear box all things considered.
Click to expand...
Delta / Wye changeover doesn't do that much to help the motor out, in the way that mechanical gearing does. It helps out the batteries and controller, more..... :arrow: :)
 
This hasn't been brought up in a while, so maybe some of you guys have not seen it but there exists a two speed transmission already on the market for scooters.

http://www.extremescooters.biz/evo-2-speed-transmission.aspx

I don't know the power handling of this unit, but I have been waiting for someone to put this on an ebike, so far no takers. :(
 
Thud said:
Mike, I get the hook ups to make the switch.
performing the switch with ellectronics is where my confidence falters. Think of me as a 47 year old new born. :p I am prolly not stupid, I just never had a need to delve int such things before.
on a sensorless design the mechanical switch is NBD theroy wise, but dont you jumble the timing in a conventional sensored set up? & going to more than 3 phases, whoa...that just confuses my brain exponentialy :?: :!: remeber, its not the basics, its the details(again I would think adding sensors to match the switch & switching them at the same time would be in the realm of doable) Any one here riden a bike 2 speed with the delta wye switching? (other than the hub motor set ups) seems as good a solution a a gear box all things considered.
Click to expand...

even on a sensored model, all the change does is throws one phase 30degrees out. This sounds horrible but the good news is infineon and others seem to handle this and adjust automatically (based on testing) so it becomes less important... if you want to detail it out, you only need to add one hall sensor to run both in perfect time (swap one hall out via relay type circuit).

It's not a gearbox and as miles points out... its more for economy, you startup in wye and when the current falls but speed remains constant...
interrupt throttle
switch off wye output
switch on delta output for 1.73 * wye termination
resume throttle

In industrial applications it's used to startup very high rpm motors at lower current...

it would more compliment the 2 speed reduction / transmission.

-Mike
 
Thanks Mike & Miles,
That clears up a few questions I had regarding the whole D/Y switching debate.

E-tard,
I have seen that tranny a thousand times & it looks good to me,every time. When I build another 2-speed (not if) I may just go with syncro belts. I haven't thought about adapting that paticular tranny given but I will do the math next time around, I may spend a few bucks to save some time. The stamped steel frame would be easy to modify & addapt to.
Are there any sizes listed anywhere on the web specific to it?
Any one around own a scooter with one on it?
Of course I would have to re-dye all the plastic parts to match my color schemes. :wink:
 
This build is so impressive.

It's a pleasure to have you share your work with us Thud!
 
etard said:
This hasn't been brought up in a while, so maybe some of you guys have not seen it but there exists a two speed transmission already on the market for scooters.

http://www.extremescooters.biz/evo-2-speed-transmission.aspx

I don't know the power handling of this unit, but I have been waiting for someone to put this on an ebike, so far no takers. :(
Click to expand...

DAMNit why was i not told of this etard i would of incorporated it into the crusier :-| Any more specs on this other than the pic?

evotransmission.jpg



KiM
 
There was a whole thread on this Evo-type transmissions awhile back.

I've still got the D-Y switcher on my todo list, but I must say, I like the simplicity of Thud's mechanical unit. I'd like it even better with belts, because of the noise factor. I am not happy with screaming gears/chains/etc. The problem with gears, other than the noise, is that metal-on-metal wears too much, and using composite material for one still won't last that long. I did a small reduction unit for an RC electric helicopter conversion years ago that used beveled helical/spiral-cut gears and they would last a longtime without lubrication and didn't make a lot of racket. That might be an option here, I don't know, but personally, I like the belt drives, even if it does increase the size slightly.

The D-Y switcher circuit that Fechter helped me with is shown below. It uses some 60A latching relays that simplifies the circuit, in that you can have a single SPDT switch mounted up on the handle bars, but the problem is that the latching relays are oly SPST so you need six of them to make this work. This means it takes a pretty good size box (4" x 6" x 1"?) to hold everything.

Delta-Wye_Switcher-v1 1.png

The 1.73:1 ratio change you get, going to delta mode, is more than enough for most of the RC setups we've been looking at. That really widens the torque range to just enough to not require a huge motor or any additional "gears". My 3-speed hub is an overkill for the motor. Two speeds, with a ratio change between 1.5 - 2.0:1 is enough.

I agree with Mike, that ideally it would be best to have a fully electronic version of the D-Y switcher, even if it is separate from the existing controllers, if for no other reason than it could be small. Until then, I think Thud's 2-speed unit is the best option I've seen. If it could be a bit quieter, and not too much bigger, so much the better. :)

-- Gary
 
I had also previously thought a belt-based two-speed transmission would be the ideal.
The scooter box may not have sufficient power-handling for some of us.

I had dismissed this as being too hard to custom build, particularly the sliding dog arrangement...and here Thud has come up with the goods!

Legendary work, Thud. :D
 
Gary...

Do you know the current draw of the relay based delta/wye switch, and the relay cost for say 50v or 54v and 60A (assuming 32A in wye * 1.73 for delta current = 55.36 A).

Assuming IRFB4110 in qty at 2.00-3.00 ea, methods has them for 2.00 ea I think. That would be as much as 36.00 for the FETs... then you need the AND gates to create the 3PDT switch from the FETs, oh and the charge pump for 6 high sides... I think these could be kitted up for about 50.00ea using the solid state version.

Possible advantages over relays:

Lower power consumption -
Lower power loss to heat - optimized switching (I need to check this but I'll wait till I get a better idea what the relay method would cost people)
Lighter weight
Higher voltage and / or Current handling (we can always start paralleling IRFB4110s to drop the on delay and resistance)
More compact
Can be driven by simple logic gate (switched between the 2 termination modes) with a short delay/counter (555) between for dead space
or
Manged by 4.00 MCU - allowing monitoring of wheel speed, current, voltage, back EMF (if needed for safety) and now we can not only automate but provide manual with tons of safety checks to prevent possible issues this could cause +
Could be combined with servo actuator and the 2 spd transmission to truly provide real automatic 4 speed transmission (could go completely silly and run to a internal geared hub still... how many gears do you thik we can automate?)

I'm kinda stuck in a loop right now because Im working on BLDC motor controller (so incorporating delta/wye seems best option) but it's going to be a while before the controller is near ready for others to build in the meantime if the relay board won't handle enough power or is too expensive, I may work a standalone delta/wye switch out (with safety - but then Im going for the 4.00 mcu to simplify monitoring for me, and now that ive added an MCU why not make it a full blown controller (it has 12 fets no matter what).

-Mike
 
I'm not trying to be funny... (for a change) but, Thud you need to name this damn thing... we can't call it a "Thud Transmission" - that's just wrong (I give up, I'm only funny when I try to be serious!)

-Mike
 
You must log in or register to reply here.

Similar threads

M
    • Like
Kawasaki eliminator 125cc cruiser build
Replies
10
Views
865
MX-215
M
M
    • Like
KLR250 Scrambler Build
Replies
5
Views
800
MDG
M
S
Mirroring hydraulic brakes and belt drive on frame runner
Replies
18
Views
448
slomobile
S
neptronix
    • Like
Mini Hiryuu - Mid power ebike
2
Replies
28
Views
1,061
neptronix
neptronix
Bush mechanic
1st battery build (20s2p) - collective advice?
Replies
6
Views
507
electric_nz
E
Back
Top