MatthiasBr said:
I am designing a cargo bike / commuter bike.
I am still in the ‘napkin sketch’ fase as I am checking if everything is possible as I want it before going all out in Solidworks.
150 kg cargo max, own weight around 50kg is my goal + rider at +- 80kg
Well, with that kind of weight for cargo, the extra weight of the hubmotor won't make much difference to the system.
Cost, that's a separate thing; probably at least $200-$300 for a common cheap DD hubmotor kit (since you don't need much motor); more if you want to build it into a good wheel (instead of the maybe-it's-good one it will come in), and more if you want a better controller with more control over the regen. Might be $500+ just for the front hub system (motor, controller, etc), depending on your specific needs.
You don't need a powerful DD motor or controller (500-1000w is likely more than sufficient, since the innotrace will be doing any heavy lifting thru the bike gearing), but you probably do need a controller that lets you have fine control over the regen braking.
Most true-FOC controllers can, as a side-effect of how they work, also finely control the amount of braking. The smallest one I know of that is also waterproof is the Phaserunner (or Baserunner if you are installing it inside something else like a battery cradle) by Grin Tech
https://ebikes.ca/product-info/grin-products/phaserunner.html
but they're a lot more expensive than the average controller. (more versatile and probably much better supported, but kind of a big budget hit).
One thing about more versatile controllers with more features is that they are not plug-and-play; they all require you to set them up for your specific system, motor, and needs. But at least you *can* do this, which isnt' typically possible with the easier-to-use plug-and-play kind.
Use the Innotrace midmotor to do most of the work as a torque machine. Get the bike up to speed and use it for riding around normally. If possible I would also like a thumb throttle on the innotrace to sometimes give it an extra burst of power when riding around.
The Innotrace should be more then enough to fulfil those needs I guess, even in the normal 1000W trim.
Depends on your gearing, and how (or if!) you shift while using it.
If you gear the pedal drivetrain for low enough speeds and high enough torque, you could probably climb trees with it.
If you never shift gears, you'd want to gear it low, which means your pedals are also geared low, so you won't be able to do much if any human power at higher speeds.
If it is geared low enough so that the top speed of the innotrace is at the point you want the hubmotor to take over, then the innotrace will be using very little power at that point anyway, and it could be a smooth transition.
On my normal riding (with my current Bosch powered cargo bike, an Urban Arrow for those interested) I notice I brake a lot. It is light braking, just to slow down a bit for a pedestrian, or slow down to a stop for traffic light. It is the type of braking that wears the brake pads out in an inefficient way I feel. So that got me thinking that a heavy cargo bike might be good to use variable regen braking with a thumb throttle, to eliminate that light braking with the callipers.
It can certainly be done that way. That's how I eventually will be using my SB Cruiser heavy cargo trike,
https://endless-sphere.com/forums/viewtopic.php?f=2&t=67833
using the Lebowski controllers on the dual rear hubmotors (they have variable regen too).
https://endless-sphere.com/forums/viewtopic.php?f=30&t=105711
Presently I have a Grinfineon controller version that has variable regen (often called proportional regen, if you are searching for pages about it), but it is very weak, so I still use the front (regular wheel) Avid BB7 MTN with 203mm rotor for most of my braking; it does wear the pads pretty quickly, but works great in most situations.
Because teh Grinfineon doesn't have a direct input for regen amount, but does a wierd thing where the bottom 0.8v controls it (below the normal throttle level), then I have to use the Cycle Analyst v3 to do this (it's designed partly for this). So mine is a little complicated, but yours doesn't have to be, and mine wont' be later when I finish the Lebowski controllers and use direct control over their regen. So my regen control setup is this, right now:
https://endless-sphere.com/forums/viewtopic.php?f=2&t=105460
It sounds complicated but it isn't really, and it is much less complicated if the controller you use has an analog input specifically for the braking control input.
For the future I will be using a dual-pull ATV lever on the right side to also allow all-wheel braking control from one lever, where one cable pulls the actual front disc brake cable, and the other cable pulls a cable-operated throttle to control rear regen braking. Ideally I'll also modify my frame to accomodate rear disc brakes, and use another dual-pull lever on the left side for those plus the regen, but that is farther away in time.
Most basic controllers dont' have variable / proportional regen, just on/off (though some have a programmable strength for when it does engage).
Some of the controllers that do use two separate analog inputs for it, some use just one and the brake lever input just switches between modes for throttle or brake control with your throttle.
The cargo bike will also be my commuter as I mentioned and I have about 80% towpath and bike-highway on my commute with hardly any stops on that. So I don’t need a high powered hub motor to maintain the speed of 45 kph.
DD hubs are more efficient than mid drives in Wh/km so I was thinking this was a good extra reason to look into the regen bit.
As long as the DD hub is setup (via winding choice, wheel size, and battery voltage) to be most efficient at the speed you are riding at most often, and most of your riding is done at that speed, then yes, it can be more efficient. The Grin Motor Simulator may help you determine that, if they happen to have the motor you want to use listed (otherwise you may be able to get specs for the motor you want and set it up as custom in the simulator, to check it out). http://ebikes.ca/tools/simulator.html
But if you have to change speeds a lot, or your terrain / motor load changes a lot, the middrive can be more efficient if you shift gears for it correctly.
So, it depends.
- When up to speed switch the main drive to the hub motor and feed the innotec with a small current to account for internal losses. This doesn’t have to be automatic, would be perfect if I could do this with a switch.
Is this in any way possible?
Yes.
IF the innotrace doesn't have a freewheel / clutch built into it so that it has little to no drag on the pedals when not in use (it should), and doesn't have the built in ability to "virtual freewheel" (probably doesn't), and has a throttle input (don't know), then you can determine the amount of throttle it takes to just barely provide the power needed to "freewheel", and setup a simple resistive voltage divider to provide that amount of throttle to it at all times when you are not actually using higher amounts of throttle.
You can either set it up as a separate divider and input to the same throttle signal, switched in by a relay or other system at specific demand points manually or automatically, or you can just add a resistor to the ground line of the throttle to "lift" the ground of the throttle enough to force it to always output a minimum amount of voltage. The latter may not work if the innotrace has a "throttle fault" check, which controllers do to check if the throttle is already "stuck on" when the bike is turned on, to prevent runaway operation without a rider.
The same thing can be done for the front hubmotor system.
Both of them would be "disabled" as motors whenever you are braking as long as the brake lever(s) send the ebrake signal to both controllers to turn them off like they normally would.
The Cycle Analyst v3 can be used to do the throttle offset, too, as well as a number of other useful things, for the front hubmotor system (the innotrace already has "smarts" in it that are likely to try to override thigns the CA would be doing, so I wouldn't use the CA on it). Then you can use a "dumb" controller for the front hubmotor, and use the CA to tune that system to work the way you want it, including disabling that motor below a certain speed, etc. You can even set it up with a cadence PAS sensor on one of the cranks (wherever there is room around the innotrace system) and control the hubmotor speed or power with the pedals and not need a throttle (but still be able to use one in addition if you like; I do this on SB Cruiser).
