41south
10 mW
Splendid work Adrian, that C80100 is a mean looking motor 
Commuter Booster - <1kg Friction Drive
- Thread starter adrian_sm
- Start date
Fietsbel
10 mW
Hi Adrian,
That 80-100 motor is going to be massive, in all aspects! Im curious how you are going to power that, considering you would want to keep the battery pack compact, while maintaining some range?
Its definately going to work, im not afraid of that! Keep us posted.
About the double swingarm... im not sure about the mechanical workings of it. I tried it and learned the hard way that 0-load current is way higher due to the extra bearing, that is actually putting more strain on the setup than without the second swingarm.
It "feels" stronger indeed , better supported, though mechanically speaking, the extra swingarm and bearing are overconstraining the setup. Just some thoughts for discussion
That 80-100 motor is going to be massive, in all aspects! Im curious how you are going to power that, considering you would want to keep the battery pack compact, while maintaining some range?
Its definately going to work, im not afraid of that! Keep us posted.
About the double swingarm... im not sure about the mechanical workings of it. I tried it and learned the hard way that 0-load current is way higher due to the extra bearing, that is actually putting more strain on the setup than without the second swingarm.
It "feels" stronger indeed , better supported, though mechanically speaking, the extra swingarm and bearing are overconstraining the setup. Just some thoughts for discussion
gtadmin
10 kW
Could use a roller bearing on one sideFietsbel said:...It "feels" stronger indeed , better supported, though mechanically speaking, the extra swingarm and bearing are overconstraining the setup. Just some thoughts for discussion
adrian_sm
1 MW
See this thread for the history.
http://endless-sphere.com/forums/viewtopic.php?f=28&t=28634&start=45#p413947
I am only building it as an experiment so i can make a video of me doing wheelies and burnouts.
Then i will return the motor and controller to full-throttle before i do something stupid.
Batteries will just be some cheap lipos. 10s2p of the 20c 5000mah turnigy cells will be good for 100amps easy.
As for the double swing arm designs, if it is machined well they run just as smooth as a single swing arm. Just need to make sure the centre to centre distance is the same on both arms. My little cb50 dual swing arm still runs less than 2 amps at noload and that is before the bearing grease has warmed up.
http://endless-sphere.com/forums/viewtopic.php?f=28&t=28634&start=45#p413947
I am only building it as an experiment so i can make a video of me doing wheelies and burnouts.
Then i will return the motor and controller to full-throttle before i do something stupid. Batteries will just be some cheap lipos. 10s2p of the 20c 5000mah turnigy cells will be good for 100amps easy.
As for the double swing arm designs, if it is machined well they run just as smooth as a single swing arm. Just need to make sure the centre to centre distance is the same on both arms. My little cb50 dual swing arm still runs less than 2 amps at noload and that is before the bearing grease has warmed up.
Fietsbel
10 mW
gtadmin said:
Gtadmin, what forces will that bearing take? The way I see it, the only force/stress it can support is the bending force on the axle after the last bearing in the motor its core?
That would mean, speaking of pure mechanic purpose other than "between the ears" and aesthetic, point of view, that in the best case, the second swingarm and bearing would only prevent bending of the axle as a result of the tire pressing against the motor can. naturally, the stress on the pivot point will be lower when the pivot axle is stressed from two sides, though leaving that out of consideration as this seems not to be a problem at all.
English is not my first language though I hope I am making my point clear here..
The main point is, an axle needs only 2 ball bearings to establish its place in space, adding a third will need to be (theoretically) endlessly precisely positioned and therefore in practice not the right thing to do.
In practice, it is done if loads are so high that the 2 bearings will not support the shaft, in that case, the internal freedom (the bending of the shaft) can be constrained by the third bearing. (imho this particular application does not require that third bearing)
any thoughts, agree/disagree..
Edit: http://memagazine.asme.org/Articles/2009/september/Exact_Construction.cfm might help to explain the degrees of freedom and constraining of them..
Erogo
100 W
adrian_sm said:
Cool, you're going to do something stupid after you return the motor and controller? Looking forward to hearing about that!
gtadmin
10 kW
Fietsbel said:...It "feels" stronger indeed , better supported, though mechanically speaking, the extra swingarm and bearing are overconstraining the setup. Just some thoughts for discussion
Me: Could use a roller bearing on one side
Yes, the roller bearing would support the bending force (and twisting of the side arm and it's mounting point) but not constrain the axle laterally which should overcome high your 0-load current, however ...
Of course, it (a roller bearing) will add width at it's mounting point, the second swing arm. Your english is fine BTWadrian_sm said:...As for the double swing arm designs, if it is machined well they run just as smooth as a single swing arm. Just need to make sure the centre to centre distance is the same on both arms. My little cb50 dual swing arm still runs less than 2 amps at noload and that is before the bearing grease has warmed up.
Cheers,
GT
adrian_sm
1 MW
Update
Finally got a decent fail from my testing. Resulting in a long walk.
The short of the story is that I now recommend dimpling the axle on the swing arm, and loctiting the securing grub screws.
The long story is I had the main swing arm on my dual arm CB50 part ways from the axle. This was due to a number of factors:
- sliding fit between axle and swing arm, rather than new interference fit :?
- having the grub screws poorly located relative to the flat on teh axle shaft
- This first prototype resulting in placing some axial force via the torsion spring, that tries to pull the shaft from the swing arms. The force is relatively low, but with the other factors was enough.
- and I didn't RTFM and check that the swing arm grub screws were still secure every so often, during the first few hundred kms.
So for all the Alpha testers, I recommend that once you have assembled the shaft to the swing arm, you should remove the securing grub screws, and use a small drill bit to dimple the shaft, then replace the grub screws and loctite them in place.
In the future I will only ship the kits with the shaft installed to ensure I have control of this crucial step.
- Adrian
Finally got a decent fail from my testing. Resulting in a long walk.
The short of the story is that I now recommend dimpling the axle on the swing arm, and loctiting the securing grub screws.
The long story is I had the main swing arm on my dual arm CB50 part ways from the axle. This was due to a number of factors:
- sliding fit between axle and swing arm, rather than new interference fit :?
- having the grub screws poorly located relative to the flat on teh axle shaft
- This first prototype resulting in placing some axial force via the torsion spring, that tries to pull the shaft from the swing arms. The force is relatively low, but with the other factors was enough.
- and I didn't RTFM and check that the swing arm grub screws were still secure every so often, during the first few hundred kms.
So for all the Alpha testers, I recommend that once you have assembled the shaft to the swing arm, you should remove the securing grub screws, and use a small drill bit to dimple the shaft, then replace the grub screws and loctite them in place.
In the future I will only ship the kits with the shaft installed to ensure I have control of this crucial step.
- Adrian
deVries
100 kW
I don't know if an interim solution might work like using an led or alarm that would come on once the bike got enough speed & then the rider would know it's ok to apply throttle. Human brain dependent. :lol:
Some ESC have a break function. Is there some way to have that "turn on" using the throttle break signal coupled to a minimum speed sensor, which would then release the throttle "break" at minimum speed & turn off the break on the ESC to let it ramp up?
Ok, I'm just rambling late at night, because I'm excited to see you advancing the FD-tech beyond what we've have available before. 8)
Great to play catch-up now on ES. Have been offline for several months, so I'm having an ESpecially good time reading about everyone's updated projects.
gtadmin
10 kW
Adrian, that's not really a failure of the design, just of the operatoradrian_sm said:Update
Finally got a decent fail from my testing. Resulting in a long walk.
The short of the story is that I now recommend dimpling the axle on the swing arm, and loctiting the securing grub screws.
The long story is I had the main swing arm on my dual arm CB50 part ways from the axle. This was due to a number of factors:
- sliding fit between axle and swing arm, rather than new interference fit :?
- having the grub screws poorly located relative to the flat on teh axle shaft
- This first prototype resulting in placing some axial force via the torsion spring, that tries to pull the shaft from the swing arms. The force is relatively low, but with the other factors was enough.
- and I didn't RTFM and check that the swing arm grub screws were still secure every so often, during the first few hundred kms.
So for all the Alpha testers, I recommend that once you have assembled the shaft to the swing arm, you should remove the securing grub screws, and use a small drill bit to dimple the shaft, then replace the grub screws and loctite them in place.
In the future I will only ship the kits with the shaft installed to ensure I have control of this crucial step.
- Adrian
However, it has highlighted an opportunity to improve the design, reducing the dependency on the operator to perform some basic maintenance. And, everyone only RTFM when all else fails
adrian_sm
1 MW
@ gtadmin
It is definitely a design failure. Luckily one that has already been resolved in my current CB63 design. I was just being lazy and used old CB50 parts.
But I agree it is good to learn from these to ensure they don't happen to others. Especially since you can't rely on people doing maintenance checks. I have to make this a bulletproof as possible before I release it to a wider audience.
@deVries
All those suggestions sound just as hard if not harder than what I currently have cooking. So I might just stick to the plan. But thanks for throwing some ideas out there, I love getting a fresh opinion now and then.
It is definitely a design failure. Luckily one that has already been resolved in my current CB63 design. I was just being lazy and used old CB50 parts.
But I agree it is good to learn from these to ensure they don't happen to others. Especially since you can't rely on people doing maintenance checks. I have to make this a bulletproof as possible before I release it to a wider audience.
@deVries
All those suggestions sound just as hard if not harder than what I currently have cooking. So I might just stick to the plan. But thanks for throwing some ideas out there, I love getting a fresh opinion now and then.
full-throttle
1 MW
Bummer..
I still don't understand why you had to walk. Did the drive destroy the tyre? I did notice the back tyre on your bike was flat when I was leaving tonight, with no apparent rips.
I still don't understand why you had to walk. Did the drive destroy the tyre? I did notice the back tyre on your bike was flat when I was leaving tonight, with no apparent rips.
adrian_sm
1 MW
Update: Modified Watt Meter
The reason I had been running the CB50, was I have been testing my modified watt meter while I what for the next gen hardware to arrive for future throttle interface development. It has been happily limiting my current to user selectable limit (currently 30amps nominal). As well as replacing the servo tester, and ramp control do-hicky. This means I just plug in a button as the throttle, connect up to the battery and esc throttle, and I am ready to rock.
It has kept my tiny little 50mm motor happy, while still providing 500-800w of assist. The power level is really usable, so I can cruise at over 40kph, go up hills at 30+ kph, and get the benefits of riding an efficient road bike. Meaning I am actually getting better wh/km than my hub motor bike when riding the same route at very similar speeds.
The improvement in efficiency I put down to a number of things:
- road bike is more efficient, so you end up using the assist a lot less
- you don't have the energy sapping resistance of a hub motor to over come when just pedaling
- the lower power limit means you are operating the motor more efficiently
- I have reduced the tire engagement to loss less power to the motor/tire rolling resistance.
It has also given me a greater appreciation of some of the subtleties of how to implement my next generation throttle interface.
- Adrian
The reason I had been running the CB50, was I have been testing my modified watt meter while I what for the next gen hardware to arrive for future throttle interface development. It has been happily limiting my current to user selectable limit (currently 30amps nominal). As well as replacing the servo tester, and ramp control do-hicky. This means I just plug in a button as the throttle, connect up to the battery and esc throttle, and I am ready to rock.
It has kept my tiny little 50mm motor happy, while still providing 500-800w of assist. The power level is really usable, so I can cruise at over 40kph, go up hills at 30+ kph, and get the benefits of riding an efficient road bike. Meaning I am actually getting better wh/km than my hub motor bike when riding the same route at very similar speeds.
The improvement in efficiency I put down to a number of things:
- road bike is more efficient, so you end up using the assist a lot less
- you don't have the energy sapping resistance of a hub motor to over come when just pedaling
- the lower power limit means you are operating the motor more efficiently
- I have reduced the tire engagement to loss less power to the motor/tire rolling resistance.
It has also given me a greater appreciation of some of the subtleties of how to implement my next generation throttle interface.
- Adrian
adrian_sm
1 MW
full-throttle said:
When the drive parted ways it fell down and jammed between the wheel and seat tube. Instantly locking up the rear wheel (going down hill at about 55kph). This pinch flatted the tube, bent my rim, and damaged my seat tube.Not a great start to the day.
I just want to make sure this doesn't happen to me or anyone else. Hence the updated method of securing the swing arm.
full-throttle
1 MW
Ouch..
Glad you're ok
Glad you're ok
gtadmin
10 kW
Ditto
adrian_sm
1 MW
This is the exact reason I wanted to limit numbers of the alpha build. Since I just don't have a huge number of kms of testing under my belt yet.
Good thing is it has already got me thinking through any other possible failure modes, and what I can do to make it all failsafe. In fact I was already thinking about it before I stopped skidding.
And just to make it clear, the CB63 alpha kits should all have an interference fit for the axle, and no axial load from the spring. So they should not have the drive trying to disassemble itself like my thrown together prototype CB50 unit. But I would prefer precautions are taken by dimpling the shaft, and loctiting the grubscrews.
Good thing is it has already got me thinking through any other possible failure modes, and what I can do to make it all failsafe. In fact I was already thinking about it before I stopped skidding.
And just to make it clear, the CB63 alpha kits should all have an interference fit for the axle, and no axial load from the spring. So they should not have the drive trying to disassemble itself like my thrown together prototype CB50 unit. But I would prefer precautions are taken by dimpling the shaft, and loctiting the grubscrews.
deVries
100 kW
One other thought I had about using the ESC's built-in "break feature" is you could at least "hack the signal" or "copy the input" for the brake on/off switching that the ESC uses, since most ESC are already programmed to receive this brake input signal by RC. Then that info could be copied & used/inputted somehow in your possible designs. No need to reply. You seem to be already into a solution you understand that will work & be less complicated. Look forward to your implementation.adrian_sm said:
One thing I was confused about on your google-website's shopping list is what parts do *you* supply that are separate from the Hobby King list (or need to be acquired elsewhere or local). Can you list just the custom parts *you provide* w/pricing that are not acquired from HK or from other sourcing. When you get ready to offer the next version of kits, I hope to be able to try one at some point, but I'd like to know what custom parts only you can provide that I will need from you directly w/pricing. (I'm hoping I can use my current stash of HK & other parts, etc. & just buy the essentials I would need from you. That would save on shipping costs too.)
Thanks!
gtadmin
10 kW
I think it was Edison, on making MANY prototypes before he finally got it right, said "I didn't fail 2000 times, I just discovered 2000 ways on how NOT to do it", or words to that effect :lol:adrian_sm said:
adrian_sm
1 MW
deVries said:One thing I was confused about on your google-website's shopping list is what parts do *you* supply that are separate from the Hobby King list (or need to be acquired elsewhere or local). Can you list just the custom parts *you provide* w/pricing that are not acquired from HK or from other sourcing. When you get ready to offer the next version of kits, I hope to be able to try one at some point, but I'd like to know what custom parts only you can provide that I will need from you directly w/pricing. (I'm hoping I can use my current stash of HK & other parts, etc. & just buy the essentials I would need from you. That would save on shipping costs too.)
Thanks!
The price I settled on for the Alpha kits was AU$175, and this only includes the custom parts that I manufacture plus the necessary fasteners.. For the Alpha kits I also threw in a few extras just to make life easier like quality fasteners to mount the motor, motor grip material, cable ties, heat shrink. The remaining parts you will have to purchase directly as per the shopping list, unless you already have the parts from your current stash.
I am not buying sufficient quantities to get any better prices than you can, so there is no point in shipping the prts to me, just for me to then ship them off again to you.So the main custom parts in the kit are:
- main clamp pivot block
- swing arm
- torsion spring
- and assorted fasteners for securing the clamp block, grub screws that adjust of dead stops etc.
Here is a picture of the main parts from the Alpha kits I shipped out.
For a few people I included the main axle, and axle collar.
In the future I will likely only sell the units with the axle installed.
The once you buy the motor, and bolt things together you will end up with something that looks a little bit like this.
The bare minimum extra parts to get you going would look something like this:
But if you bought the recommended shopping list you will end up with something that looks more like this:
If that price seems steep and you have more time than money, I suggest you make your own. A number of people on the forum have already done it so don't be too scared off.
At the moment I don't plan on selling any more units until I get more feedback from the current Alpha testers. Hopefully the next Commuter Booster version will not only include the mechanical improvements from testing, but I should also be able to provide a plug'n'play electronics interface. So you can just connect your throttle of choice, via this interface to the ESC, hook up your batteries, and commuter booster and away you go.
Hope that answers yoru questions.
- Adrian
adrian_sm
1 MW
Update: Brain Box
Here is the update on my throttle interface, or as I like to call it the Brain Box. It is based on Arduino so I can get hardware off the shelf, and speed up development. Later on who knows.
So finally got the bits on Friday, only had a couple of hours to play with them yet, but managed to get the following working:
- wheel sensor (using an interrupt)
- button throttle input (digital)
- battery current sensor (analog)
- battery voltage sensor (analog)
- basic PWM out for the throttle out to ESC (PWM)
This gives me most of the building blocks I need. Now I "just" have to get all the logic working. So far I am really happy with it. It looks like it will be able to do everything I want, and has been really quick and easy to get things working.
- Adrian
Here is the update on my throttle interface, or as I like to call it the Brain Box. It is based on Arduino so I can get hardware off the shelf, and speed up development. Later on who knows.
So finally got the bits on Friday, only had a couple of hours to play with them yet, but managed to get the following working:
- wheel sensor (using an interrupt)
- button throttle input (digital)
- battery current sensor (analog)
- battery voltage sensor (analog)
- basic PWM out for the throttle out to ESC (PWM)
This gives me most of the building blocks I need. Now I "just" have to get all the logic working. So far I am really happy with it. It looks like it will be able to do everything I want, and has been really quick and easy to get things working.
- Adrian
Powdersummit
10 mW
It sounds like you got most of it figured out. I had thought of Arduino after listening to a podcast about it. It should do everything that you want and be able to build it fairly easy. I know a lot of people are interested in not blowing up their esc. Keep up the good work
Carl
Carl
adrian_sm
1 MW
Thanks. Still a long way to go. The design is about protecting the motor predominantly via current limits, and the ESC by enforcing a minimum speed. Then I'll limit the rate of motor acceleration/deceleration to protect the mechanical bits, and give a more pleasant riding experience. It will also allow a very simple button throttle to be used. So it will be firmly targeted at my needs. Others are working on more universal solutions to keep ESCs alive for higher power setups. I'll leave that to them.
- Adrian
- Adrian
jbond
100 W
- Joined
- Oct 1, 2010
- Messages
- 204
Are going to try to limit phase currents as well as battery current and do you think you might need to? As I understand it the Infineon controllers don't directly limit phase current, but calculate what the phase current would probably be based on PWM, motor speed and battery current. This is then compared with a preset value (typically 2.5*battery current) and when it overshoots, PWM is further reduced. I guess your minimum speed limit partly serves the same sort of protection. I take it the ESC doesn't have any control outputs that could be used to get actual motor speed?
adrian_sm
1 MW
No plan on phase current limiting. The current sensor I selected is not suited to this task, and I don't understand the value of doing it. What is the advantage?
As for the ESC giving motor speed, I don't know, and don't really have a need for it at the moment. I still would need a wheel sensor as my motor is totally disengaged from the wheel before I spin it up. So to enforce a minimum speed it doesn't help. But I didn't want to rely on a specific ESC. I would much prefer to make the design generic enough that it doesn't care and people can use what ever ESC they like. Otherwise I would just get a custom ESC made that does all my "Brain Box" activities on board.
- Adrian
As for the ESC giving motor speed, I don't know, and don't really have a need for it at the moment. I still would need a wheel sensor as my motor is totally disengaged from the wheel before I spin it up. So to enforce a minimum speed it doesn't help. But I didn't want to rely on a specific ESC. I would much prefer to make the design generic enough that it doesn't care and people can use what ever ESC they like. Otherwise I would just get a custom ESC made that does all my "Brain Box" activities on board.
- Adrian
