Commuter Booster - <1kg Friction Drive

Motor was the 6374-170kv AerodriveXP, but I think they got the wind wrong because it spins faster than it should on 5s.
http://www.hobbyking.com/hobbyking/store/%5F%5F7870%5F%5FTGY%5FAerodriveXp%5FSK%5FSeries%5F63%5F74%5F170Kv%5F3250W.html

I always pedal, I can't help t once I throw my leg over a bike.

There was also a killer hill on this trip on the way home, I must zoom in on the stats some time and look into it. It was the sort of hill I would normally have to walk up. When I got to the top there were a couple of lycras that were resting. I just kept on riding. :D

- Adrian
 
Motor was the 6374-170kv AerodriveXP, but I think they got the wind wrong because it spins faster than it should on 5s.
http://www.hobbyking.com/hobbyking/store/%5F%5F7870%5F%5FTGY%5FAerodriveXp%5FSK%5FSeries%5F63%5F74%5F170Kv%5F3250W.html

I always pedal, I can't help to once I throw my leg over a bike.

There was also a killer hill on this trip on the way home, I must zoom in on the stats some time and look into it. It was the sort of hill I would normally have to walk up. When I got to the top there were a couple of lycras that were resting. I just kept on riding. :D

- Adrian
 
adrian_sm said:
Motor was the 6374-170kv AerodriveXP, but I think they got the wind wrong because it spins faster than it should on 5s.
http://www.hobbyking.com/hobbyking/store/%5F%5F7870%5F%5FTGY%5FAerodriveXp%5FSK%5FSeries%5F63%5F74%5F170Kv%5F3250W.html
I always pedal, I can't help to once I throw my leg over a bike.

Hi,

Thanks for your reply. I have the same problem with this motor on my electric stand up scooter : I expected 28km/h and the real speed with 5S is 37 km/h ! :eek:

Eric
 
Yeah, annoy isn't it. I was actually wanting to try out a slower wind.

Just checked the no load speed, 57kph on 20.1V, so about 240-250kv, just a little bit off. Thanks mister HK.
 
That's annoying for you. I have a spare Aeolian 63-74 170kv motor that is definitely 170kv. You are welcome to borrow it for a while to test with your setup. I had a few sync issues with it. It would be interesting to see how your setup handled it.
 
Thanks John, I may take you up on that. But for now I have enough to keep me busy. :D

What was causing the sync issues, I just don't seem to be getting them. I guess it may be the way I control the throttle...
Might be interesting to trying it out just to test that theory.
 
The Aeolian motors just seem to be more susceptible to sync problems. Not having any issues with the Turnigy motors unless I get really aggressive with the power delivery at low speed. If I keep it down around the 500W mark, it is usually fine but will have the occasional out of sync event at low speed up a hill. Had the same problem using the CA interface too.

Would be a good test to see if your current control and ramp routines keep this motor in check.

Anyway, the offers there. :)
 
Well, I'm going to change the subject here for just a moment. I took the plunge and ordered 30AH at 22.2V of lipos for $200 shipped. Going near wide open throttle I get just over 1km for every AH... so that gives me around 30km range @50km/h... enough that running out of juice will not be a problem.

I took a chance and ordered 10C continuous 15C max cells, but even at 60amps that's only 2C, which seems very gentle for these cells. I'll be charging them to 4.13 V per cell and will likely never fully discharge to 3.7... most rides will be 10km or less at a time. Charging will take place at about .1C in my garage, not in my house, so I fail to see where the risk is in getting these cheaper cells. I know it's too late now, cause I already made the order, but is there any risk in what I am doing? I feel as though I will be running the cells very gently and charging them safely, so I'm not too concerned about exposing any dangers in a high IR (cheap) cell.

Am I wrong in my thinking here?
 
Sounds like you have it all covered to me. The first cycle or two are probably the most critical when it comes to finding a possible bad cell. Just keep this in mind also.

Do you plan to parallel charge the whole pack at once and link up the balance leads?
 
yes I plan to bulk charge in parallel... with balance leads in parallel also. A separate balance charge will happen every so often, maybe every week or so.
 
Sounds good. That is pretty much how I use my batteries.

Just keep an eye on the cell voltages every so often, especially:
- when you first get the batteries
- the first few cycles, and
- whenever you use close to full capacity.

This should tell you if you have any cells that may be a problem.

But the interesting point you bring up inadvertantly is the low C-rate needed for friction drives, if people keep the power levels reasonable. My preference at the moment is around 500w battery side power, on 5or6s LiPo , which is <30 Amps.
So people could use a 10Ah 3C battery, and steer clear of LiPo if it worries them.

That is the next challenge I want to tackle, finding a good battery solution for my Commuter Booster. I really want a safe, plug and play option. RC LiPo is great for those that take care, but there is a reason not many commercial products are using them. Take the Stealth Bombers for example. We can see from builds like Keplers & full-throttle's how much weight can be saved, and range/performance improved with LiPo, but John @ Stealth still steers clear of them.

- Adrian
 
gfulton said:
and ordered 30AH at 22.2V of lipos for $200 shipped.

Hey that sounds even cheaper than HobbyKing Lipos. Where did you get them?

Also, shipping must have been cheap, where did you have them shipped to?

Thanks
 
Grinhill said:
Hey that sounds even cheaper than HobbyKing Lipos. Where did you get them?

Also, shipping must have been cheap, where did you have them shipped to?

Thanks

Yes, I found it to be much cheaper. I got it from a place called Leaders Hobby... (http://www.leaderhobby.com)

They have pretty good lipo prices, but rated for poor/fair discharge rates. These seem to be older generation technology lipo batteries with presumably higher than normal IR and limited C ratings by today's standards. However, in my application (usually max 60amps) these batteries are still exceeding my discharge needs by a factor of 5. They seem perfect for the job.

I did not just go and buy the batteries though, you have to be patient. Every day the site runs a one day only promotion on a particular battery. They carry several brands of 11.1 10,000mAh batteries rated between 10-15C. They usually sell for around $35-$40 each. I waited until one of those went on the one day promotion (took about 3 weeks of watching) and bought 6 11.1 10000mAh bullistorm name brand lipos at $23.99US each. Shipping was with standard parcel for about $50, but they just emailed me back yesterday saying that were not allowed to ship parcel for that many batteries... too dangerous. They required $6 more to ship DHL. Honestly that's even better for me, it is actually much faster, and I trust that company to handle the cells more carefully anyway. So now my order is closer to $210 but I get 1 week shipping via DHL which is super. I live in Canada and that company is a Chinese company, of course.

The reason I got 3s: This will allow me to charge them on my tiny little portable 3s balance charger on the go, if I ever require it. The charger is no larger than a small pack of cigarettes and plugs directly into the wall, no power supply needed. The intent is to have the portable 3s charger stowed in the battery case with the bike at all times. I will make two 11.1 30Ah packs and run these in series on the bike. I can also bulk charge as 6s without balancing using the big charger. Every so often I'll unplug them from the pack and give them each a slow discharge and balance charge on their own. I can't wait to get them.
 
adrian_sm said:
That is the next challenge I want to tackle, finding a good battery solution for my Commuter Booster. I really want a safe, plug and play option. RC LiPo is great for those that take care, but there is a reason not many commercial products are using them. Take the Stealth Bombers for example. We can see from builds like Keplers & full-throttle's how much weight can be saved, and range/performance improved with LiPo, but John @ Stealth still steers clear of them.

- Adrian

Yes, I understand many folks are a little put-off by lipos. I have come to build a good working relationship with them... in that I have a few RC helicopters that use lipos for many years now. I've seen my cells over-charged, over-discharged, dropped, involved in crashes (yes, I've crashed a few times). I've had older cells that just completely lost their capacity over time. I've never witnessed any explosive incidents, no fire for that matter.

That said, I'm still careful in my storage of lipos, i.e. they are outside the home.
 
Hello there,

I am newbie, but how come this friction drive is able to reach 40kph for 30km with a total weight including battery no more than few kg?

All other kits I found (Golden motor, conhismotor, Ezee, etc) weight much more and add almost 10 kg to do the same thing.

It almost like the other kit are outdated compared to the friction drive. Am I missing something here?
 
My design just uses all the lightest weight technologies available, and I bolt it on to an efficient bike. But it comes with some downsides.

The downsides are:
- not good for high power/torque for extended periods due to low thermal mass, and limitted ability to shed waste heat.
- not able to apply power from a standstill, due to type of controller used
- not good in the wet (reduces the amount of torque able to be applied)
- can wear out tires when used at high power
- more expensive, due to low production volumes

Hub motors are designed cheap and robust from the start. This pushes the weight up.
eBike controllers are designed to handle starting from standstill, and limit power, this pushes their weight up.
Alternative batteries weigh more.
Finally all the above extra weight, and a possibly a crappy bike can lead to poor efficiency.

Hope that sums it up for you.

- Adrian
 
Update: Brain Box

Okay time for an update. I have been working very hard at getting this all running, with all the features I would like and it is turning out better than I had hoped for. I am seriously running into memory limits of the hardware, but the feature set is very rich.

For the Alpha prototypes I will only be making a version with an 8x2 LCD designed for my friction drive. Later I might do other varients, with hall/pot throttle input, 16x2 display, etc.

Here is the description of the Alpha BrainBox:
Main Features:
- LCD display
- button throttle
- fuel gauge: battery used, distance remaining, voltage etc
- normal bike computer stats: trip time, dist, speed max/avg, total dist, time etc.
- eBike computer stats: wh, average watts, wh/km etc.
- limits power via PID control loop,
- controls friction drive engagement/disengagement
- enforces minimum speed
- can enforce maximum speed, to stay legal, or improve efficiency
- can perform cruise control (PID speed control)
- datalogging to microSD card, trip summary and live data (~20Hz or less)
- speed, volt, current, and temp sensors
- user programmable on the bike via single button

Inputs:
- button throttle
- wheel sensor
- voltage/current sensor
- temperature sensor
- mini USB
Outputs:
- 8x2 LCD display in a compact enclosure
- ESC throttle signal
- SD datalogging

Button
- all programming and throttle functions performed by the single button
- throttle button allows selection of two user settable power levels, and cruise control while riding
- (ie. normal power, hold this speed, and turbo boost)

Display
- battery fuel gauge, speed & distance displayed when in motion
- when at rest can scroll through all other stats: dist travelled, dist remaining, Ride time, wh used, Batttery Voltage, Average Speed, Max Speed, wh/km, average power [watts], Temp [Celcius], Trip #, Total Dist, Total Time, Total kWh.

SD Datalogging - two different log files are created for each trip
(1) trip summary stats
--- automatically recorded to SD
--- same stats as those displayed on LCD
(2) trip live data readings automatically recorded at user settable data rate, ie. 20Hz or less
--- data file includes header with all user settings, software version, file name
--- data fields recorded are currrently: Time (secs),Watts,Throttle PWM,Throttle Mode,Speed,Distance Travelled, Calculated Distance Left, Ride Time (Mins),Battery (wh),Batttery (Volts), Temp (Celcius)
--- filename is automatically incremented for each trip

Menus:
- entered by holding down the button during startup
- allows adjusting power levels, speed limits, PID control loop parameters for power and speed, battery settings, log rates, wheel size, ESC throttle min/max, sensor calibration.
- allows resetting to default values if you stuff it up when tweaking

Wiring:
- currently I have hard wired everything I can.
- button throttle runs into the display enclosure mounted on the handlebars
- a single cable runs back to the shrink-wrapped uC that then has connection to ESC, battery sensors, wheel sensor, temp sensor
- this allows the uC, battery sensor, SD card, and USB connection to all be safely housed in the battery bag out of the elements.
- in the future I may change this to make quick removal of the battery possible.

Pictures

Here is a quick look at one of the units all wired up, on and off the bike.
fr_223_size580.jpg
View attachment 2
Here are all the connections from left to right
wheel sensor, temp sensor plug, battery volt/amp sensor, main uC with ESC throttle connection poking out the top, button throttle
View attachment 1
This is a close up of the Brains Box, at the left of the picture you can see the ESC throttle connection and the micro SD card slot, at the right is the mini USB port. At the bottom is the voltage/current sensor with Anderson PP45s, but deans T connectors work even better.


The main thing remaining for me is to decide on how I handle the battery gauge. Currently it is set to automatically reset the battery meter if the voltage goes above a threshold (checked when you first turn it on), then just uses watt-hours to judge battery capacity. I am trying this as I wanted it to automatically reset itself, and not require the user to reset it like a CA. But this is not foolproof. Another method would be to use the battery voltage (assuming LiPos) as I am only really using it for a crude battery capacity display, all power/efficiency calculation are performed on the more accurate wh used.

I have purchased enough parts to make 20 of these, and will soon be making them available. I will finalise the price once I know how much time they take me to put together, it will be in the ball park of a CA, but obviously with a lot more features.

- Adrian
 
adrian_sm said:
I have purchased enough parts to make 20 of these, and will soon be making them available. I will finalise the price once I know how much time they take me to put together, it will be in the ball park of a CA, but obviously with a lot more features.
- Adrian

Well done Adrian !!!
I want one, if possible ! You can put me on your reservation list :D

Eric
 
adrian_sm said:
Update: Brain Box
For the Alpha prototypes I will only be making a version with an 8x2 LCD designed for my friction drive. Later I might do other varients, with hall/pot throttle input, 16x2 display, etc.

- Adrian

In your post, I understand that your interface is compatible only with a throttle button. Is it possible to simply adapt it for a hall throttle?

Eric
 
Please excuse my ignorance, but I have not read all the past discussion re: the brainBox. Is this some type of arduino type of project you are doing?
 
@ferias77

I am focusing on the button throttle for now, as it has been working so well I have not missed a twist throttle at all. It smoothly ramps up the power, and holds it for you. Then you get the option of switching to a higher power limit, or engaging a cruise control to maintain your current speed. All with a simple button. :D

But I am sure people will still want a normal twist throttle. So when I do get around to implementing it unfortunately I will need to rewrite a lot of the code due to the way I am using the button do so many different things. I may even have to defeature the interface to make it work. :(

If you want a hall throttle with power limitting, but don't care about the datalogging, temp sensors, or display size, go for the CA-LRC. If you also don't want power limitting you could just add some ramp controll using one of these, and use a modified servo tester to convert the hall signal to a RC-ESC throttle signal.

@gfulton

Yes the brain box is currently based on an arduino uC for development, and the first batch of prototypes I make. But if this goes much further I may abandon it for custom hardware. My lengthy post above should tell you what it is able to do.

- Adrian
 
Adrian:

First, congratulations on getting close to a production brainbox.
Second, a request: This thread now has 72 pages. For a casual guy with an interest (but not an intense one) in your development, it can be really confusing. If you have time, can you start a new sale thread detailing the complete product line (motor, mount, controller, brainbox...) along with price and ordering info? And can you reserve the first few posts for yourself so you can keep the info up to date. That way a potential customer only have to read the first few posts and have a good idea of what the whole thing is about.

Thanks,
Sam
 
Thanks Sam. Yeah this thread is waaaaaaaaay too long now. I'll definitely start a for-sale thread when I am ready.

In the mean time I did do a sort of technical summary site a while back when I sold my Alpha units, that includes a lot of useful info on the mechanical side of the drive.
https://sites.google.com/site/commuterbooster/

Again, thanks for the heads up on the confusion this long thread can create.

- Adrian
 
Thanks for your reply Adrian. I am interested by power limiting, and I am currently using a hall throttle with a modified servo tester. So I am interested by your brainbox, and as you describe it, the throttle button looks very attractive. To conclude : Well done, I want one !!! :D

Eric
 
Back
Top