Cadence sensor driven rear hubs.. adding a torque sensor to existing systems?

[Ken Shopken]

I know that there are newer Bafang systems that now use both sensors, but has anyone modified an existing cadence sensor system to include a torque sensor?

I have read here and there about dropout mounted torque sensors.

It would be great to modify my existing Rad bike or any other new hub motor bikes to use both sensors.

I like to pedal the bike and earn the power as a torque sensor supplies.

 

[harryS]

ebikes.ca sells a torque sensor that replaces your pedal BB, but you need their cycle analyst display to make it work with a typical cadence controller. Its output feeds the throttle input on the controller. A couple of hundred bucks by the time you're done. For the best feel, you may have to buy one of their controllers.

I'm not convinced that an add-on system will work like an integrated system where they've spend a lot of time tuning the firmware to use both T/S and cadence in concert, Maybe a Grin customer will comment.


.

 

[Ken Shopken]

Thank you- good input.

I was visiting the ebikes.ca / Grin website a few days ago.

It would be neat to hear from someone that has converted the Bafang rear hub to torque sensing.

Adding the regen capability would be cool too, as I have no intention of riding the machine without some electrical assist.

 

[Jeremy McCreary]

When I first heard of regenerative braking in ebikes, it sounded pretty cool to me, too. But if you dig into the engineering, you find a lot of credible sources saying that the total mass (bike+rider) to be slowed is generally too small to yield a meaningful return.

As I recall, regen would also necessitate a change to a direct-drive hub motor. And that would carry its own set of disadvantages WRT the geared hub motor you already have.

 

[scrambler]

I thought that too, but My experience with the Boxxbike changed my mind.

The bike came with a regen on throttle mode that would trigger a certain level of regen when you release the throttle. It was meant to create an engine brake (this is an 8kw bike), and it did a good job for that.
I then added a second regen on the brake lever and programmed the controller to give me a much higher regen in that mode.

As a result, when going on rides with a lot of elevation, almost all my braking going downhill is done with that regen on brake, and it does recover a significant amount of battery.

So when riding on flats, I would agree that the regen is probably not that significant. But when going downhill, it can be.

 

[PCeBiker]

I then added a second regen on the brake lever and programmed the controller to give me a much higher regen in that mode.

Apparently, GRIN offers variable regeneration.
I don't remember if you twist the throttle backwards to turn on and increase regeneration, or if the regeneration is varied by how far you pull the brake lever.

It should be relatively easy to adapt a hal sensor to either the brake lever or throttle that will allow a variable input to a controller to adjust the amount of regeneration/braking.

My KT Controller and Display allows for 5 levels of regeneration/braking power.

It would be cool if you could vary your braking power from 0-8000 Watts.

 

[scrambler]

While variable regen is the ultimate tool (usually on a lever), a good controller gives ramping curve adjustments so that you can adjust how it will ramp up when you trigger regen and ramp down when you release it.
In my experience this is sufficient for a great regen experience and much easier to implement with regular throttle and brake levers.

 

[PCeBiker]

While variable regen is the ultimate tool (usually on a lever),

You could potentially delete your rear brake system all together and have the the rear brake lever operate regeneration only.

The only issue that I'm aware of is that you don't want to have a fully charged battery at the top of a long mountain descent, or you could dangerously overcharge the battery.

 

[scrambler]

No need to do that because the magura brake lever switch is at the very beginning of the lever action. During that early pull, it does not apply any mechanical brake.
So if I feather the brake lever I only get the regen braking, and if I do need to add mechanical braking (extreme descent situation), I can keep pulling the lever.

Best of both world with a single system

 

[PCeBiker]

My old e-bike has a direct drive 250 Watt 36 Volt motor and three 12 Volt, 10ah SLA batteries.


It also has regeneration on each brake lever and the regeneration also turns on at 25 kph to be legal in Europe.

I found one for sale in England,..

I got it used for $400 about 8 years ago.
The seller told me to NEVER ride the e-bike without the battery because the power has to go somewhere when you hit the brakes and I'd blow up my controller without a battery being installed.

I figure it should be simple enough to install a 250 Watt ceramic resistor using the bike frame as a heat sink.

Some simple wiring and technology should be able to switch between charging the battery or dumping the energy into the resistor?

 

[scrambler]

A good controller should take care of that automatically and not create any regen if there is no way for it to go anywhere.

 

[PCeBiker]

No need to do that because the magura brake lever switch is at the very beginning of the lever action.

My switches do the same thing but it would be really cool, and relatively simple to have that first bit of lever movement adjust the amount of regeneration before the real brakes kick in.

 

[PCeBiker]

A good controller should take care of that automatically and not create any regen if there is no way for it to go anywhere.

I don't like AI.
I'd rather control it myself.
KISS.

You'd program the regeneration every time you activate the brake lever.

Let the programming take care of when to dump the energy or disable regeneration if your battery is fully charged.

I remember someone here on the forum had a front hub motor with regeneration.
He adjusted his programming to maximum regeneration and got tossed over the handlebars when it engaged.

A progressive regeneration brake lever would eliminate that.
It would be way more efficient too and you may rarely need to use your real brakes.
8000 Watts of regeneration may be enough to lock up your wheel?

 

[PCeBiker]

As I recall, regen would also necessitate a change to a direct-drive hub motor.

A geared hub motor will also work.
You just have to disable the internal clutch mechanism.

Having the regeneration set to maximum is hard on the plastic gears inside when it slams on.

A  progressive application of regeneration is much kinder to the gears.

 

[Ken Shopken]

When I first heard of regenerative braking in ebikes, it sounded pretty cool to me, too. But if you dig into the engineering, you find a lot of credible sources saying that the total mass (bike+rider) to be slowed is generally too small to yield a meaningful return.

As I recall, regen would also necessitate a change to a direct-drive hub motor. And that would carry its own set of disadvantages WRT the geared hub motor you already have.

I read somewhere that regen recovery with general traffic use is only 5-7% with all the associated losses. I do not know if this is accurate, but seems in range.

I am interested in this because I would be climbing continuous and fairly steep grades of 1-3 miles. Energy recovery on the downhills would be significant here even with recovery losses. I am also curious as to how much current is potentially generated or ground shunted by some form of current regulation, and what significant damage to the batteries occurs when subjected to high charging loads. Requires a super capacitor or high charge rate capable cell to ideally charge the main batteries?

 

[scrambler]

Regen has to be implemented properly by the controller and the BMS.
The controller and BMS need to know what max charge current the battery can take, and once it does, it will activate regen up to that amount limiting the amount of regen so it never damages the battery.
It will also make sure not to overcharge the battery.
Ideally both the BMS and the controller have these capabilities and they will supplement each other. But at the minimum one of the two must be able to exercise that control.

 

[PCeBiker]

Is the maximum charge current normally a lot less than the maximum discharge current?

I have a 30 amp BMS.
Could my battery not then charge at 30 amps?

 

[scrambler]

It depends on the cells used in your battery and their configuration (serie / parallel).
Each battery cell model has a max continuous discharge and a max charge amp number, they are not all the same.

 

[PCeBiker]

It depends on the cells used in your battery and their configuration (serie / parallel).
Each battery cell model has a max continuous discharge and a max charge amp number, they are not all the same.

I've got Samsung INR21700-50E cells in my 48V 25AH battery.

I looked up the specs, and it looks like the cells can charge at roughly half the continuous discharge rate at maximum.

 

[scrambler]

If your battery is a 48V 25Ah with these cells it is probably a 13S 5P configuration
5 cells in parallel provide 5 x 4.9Ah = 24.5Ah
These in series 13 times create 13 x 3.6V = 46.8 Volts (manufacturer often use 3.7V for their calculations... hence the 48V)

If so, then its max safe charge current would be 2.45 A x 5 = 12.25 Amps
With possibly a peak charge for brief moments of 4.9 x 5 = 24.5 Amp