Rear all axle v3 with torque sensor

[stancecoke]

Under no circumstances would I use a "channel" to do a wire's job.

The UART connection is a wire Lishui uses UART3 of the STM32F103 processor for the communication with the sensor.
Most manufacturers are using CAN bus with their bikes these days. Nothing bad, very reliable. And easy to rebuild for a tinkerer as long as the communication is not encrypted.

regards
stancecoke

 

[FuzzyWuzzy]

Looks like they figured out how to use CA3 with 72v and a torque sensor.

Grin Rear All-Axle 72V CA3 Kit

Grin Rear All-Axle 72V Conversion Kit with Lithium Battery, Phaserunner Controller, and built-in Torque Sensor. Supports 135x9 Quick Release, 142x12 and 148x12 Thru Axle dropouts. Full Product Information Page

ebikes.ca

 

[szkuba]

i do have TDCM HCA5-F motor, apart from phase wires it does have 7 wires coming out of the cable. I suspect it has the torque sensor in the cassete. what should be the power line voltage 5V or 12V? From the grintech website the A5 motor has the 12V supply to power halls and torque sensor. Is there a safe way to check it?

 

[stancecoke]

From the grintech website the A5 motor has the 12V supply to power halls and torque sensor.

Hm, the Kclamber cassette sensor needs only 5V for supply and one wire for communication. So it works with the common 9pin motor connectors. But maybe, grintech has built in a "translator" to make the Kclamber compatible to the CA, so one or two digital signal for PAS and one analogue signal for torque. Would be really not straight forward to convert the serial UART data from the sensor back to analog signals, but perhaps a quick and dirty solution for the CA

regards
stancecoke

 

[smitty2k1]

Hoping to buy an All Axle V3 with torque sensor in the next couple months to convert my Xtracycle cargo bike with 20" rear wheel. I'd been set on a mid drive but I think this will be cleaner and less wear on the drivetrain and brakes with the regen.

 

[neptronix]

20 inch wheel is very ideal for mid size hub motors like these. Should climb and accelerate like a beast given the correct amperage. Will be very efficient when climbing/accelerating too, versus putting it in a 26" wheel.

 

[szkuba]

Hm, the Kclamber cassette sensor needs only 5V for supply and one wire for communication. So it works with the common 9pin motor connectors. But maybe, grintech has built in a "translator" to make the Kclamber compatible to the CA, so one or two digital signal for PAS and one analogue signal for torque. Would be really not straight forward to convert the serial UART data from the sensor back to analog signals, but perhaps a quick and dirty solution for the CA

regards
stancecoke

So what you say it should be 9pin cable, but this one is 10pin . The thermistor is on separate wire to the possible torque signal. Or it should be torque and PAS combined?
any simplest way to check the signal - oscilloscope? Or kind of UART decode circuit? Can you recomend something?

 

[neptronix]

I have been running multiple scenarios with the rear all axle motor to see what the absolute continuous power is.

Peak power and efficiency occurs in a 20" wheel. The amount of power you can extract without overheating is pretty mad.
The high efficiency tells me it can take even more power than this.



On a 29" wheel, our performance is subpar compared, maybe we can sustain 36mph but that's it.



Performance is pretty good in a 26" wheel. We can probably sustain 38mph.



So with this motor, the smaller the wheel, the better.
20" is way too small on a 26" MTB. You will have pedals hitting the ground issues.
22" has a problem where the tire would collide with vbrake posts in the 2.3 inch range.
24" puts the vbrake posts inline with the rim, so the tire is above the vbrake posts, and also in an area where you have more chainstay width. This is the perfect opportunity to use a 2.5" tire or larger, and technically have a 24.5" wheel.

Looks like this small size reduction would allow us to go pretty high speeds continuously with the fast wind.

So here's our 24.5" wheel size:



We could possibly push it to 40mph for periods of time since it handles 38.2mph so well.

With the standard winding, we are in a nice part of the efficiency band around 30mph.


The motor definitely has hot rod potential. 82 ft-lbs of torque from a 11lbs motor is pretty good. It's not a torque beast like the MAC, but it can at least shed the heat well enough to hit and maintain high speeds.

Looks like i'll be building a bike with one of these and a 24.5" wheel. Losing 0.75 inches or ~18mm of ground clearance vs the previous 26" wheels is a pretty reasonable loss for the continuous power gain.

 

[smitty2k1]

I have been running multiple scenarios with the rear all axle motor to see what the absolute continuous power is.

Peak power and efficiency occurs in a 20" wheel. The amount of power you can extract without overheating is pretty mad.
The high efficiency tells me it can take even more power than this.

View attachment 348067

On a 29" wheel, our performance is subpar compared, maybe we can sustain 36mph but that's it.

View attachment 348068

Performance is pretty good in a 26" wheel. We can probably sustain 38mph.

View attachment 348069

So with this motor, the smaller the wheel, the better.
20" is way too small on a 26" MTB. You will have pedals hitting the ground issues.
22" has a problem where the tire would collide with vbrake posts in the 2.3 inch range.
24" puts the vbrake posts inline with the rim, so the tire is above the vbrake posts, and also in an area where you have more chainstay width. This is the perfect opportunity to use a 2.5" tire or larger, and technically have a 24.5" wheel.

Looks like this small size reduction would allow us to go pretty high speeds continuously with the fast wind.

So here's our 24.5" wheel size:

View attachment 348070

We could possibly push it to 40mph for periods of time since it handles 38.2mph so well.

With the standard winding, we are in a nice part of the efficiency band around 30mph.
View attachment 348066

The motor definitely has hot rod potential. 82 ft-lbs of torque from a 11lbs motor is pretty good. It's not a torque beast like the MAC, but it can at least shed the heat well enough to hit and maintain high speeds.

Looks like i'll be building a bike with one of these and a 24.5" wheel. Losing 0.75 inches or ~18mm of ground clearance vs the previous 26" wheels is a pretty reasonable loss for the continuous power gain.

It's interesting the different use cases people have for ebikes. I have no desire to go that fast, especially on a MTB, or with just throttle. I just want to haul my kids around the city with some pedal assist.

 

[stancecoke]

any simplest way to check the signal - oscilloscope?

Yes, should be the easiest way.

regards
stancecoke

 

[neptronix]

It's interesting the different use cases people have for ebikes. I have no desire to go that fast, especially on a MTB, or with just throttle. I just want to haul my kids around the city with some pedal assist.

It's to keep up with traffic in the parts of my area that have no bike lanes, which is the majority of the road.

Consider their Shengyi motors if you need a lot less power. Would be my #1 choice if the target speed was around 25mph.

 

[smitty2k1]

It's to keep up with traffic in the parts of my area that have no bike lanes, which is the majority of the road.

Consider their Shengyi motors if you need a lot less power. Would be my #1 choice if the target speed was around 25mph.

Less speed but carrying a whole lot more weight as a cargo bike

 

[szkuba]

For the cargo bikes high speed above 20mph is not really recommended and even dangerous given all the mass that need to safely brake. I would optimise the motor choice to the reasonable max speed, in return you will have more optimal efficiency and higher torque that is needed to move and accelarate the mass. Btw when using motor Simulator i would insert real bike total mass that can be easily above 200kg

 

[nervagon]

For the cargo bikes high speed above 20mph is not really recommended and even dangerous given all the mass that need to safely brake. I would optimise the motor choice to the reasonable max speed, in return you will have more optimal efficiency and higher torque that is needed to move and accelarate the mass. Btw when using motor Simulator i would insert real bike total mass that can be easily above 200kg

I have a longtail cargo bike with 24" x 2.6" tires and it is the most stable high-speed bike that I have been on. When unloaded these can be fantastic high speed bikes. The quad-piston magura brakes will stop a motorcycle.

 

[nervagon]

I think the grin rear all-axle could replace the BBSHD on my cargo bike. And it looks like grin now supports smaller 3rd party displays with their phaserunner and baserunner.

 

[neptronix]

Try two 203mm discs and 500w of regen on this motor before you call it not enough braking power for >20mph.

If the bike has a taller than a 20" drive wheel, I would prefer the RH212 to provide maximum torque ( and braking power ) on a bike that is particularly heavy.

Of course, let the simulator be your guide to an ideal configuration.

 

[smitty2k1]

I think the grin rear all-axle could replace the BBSHD on my cargo bike. And it looks like grin now supports smaller 3rd party displays with their phaserunner and baserunner.

I'm about to buy an All Axle, and it will be my first ebikes conversion. Would you recommend a display not from Grin?

 

[amberwolf]

If you're referring to the CA as a display...it's not; it's a computer that does a bunch of things. If you want the features it has, you'll need that.

They are working on (but only for OEMs ATM) a cable to do some of the CA-specific types of things using the *runner controllers and a handful of specific "typical" displays, but it's not available for the general public yet (unless you want to do all the figuring out of how to make it work--the parts are there on the site to buy if you can find them but you have to get the version of *runner with the right firmware, the cable, and the specific display(s) noted there to work with it, and it'll be up to you to figure the whole thing out, until eventually they get it all worked out and make a user-end procedure for it).



If you need a controller-specific display to do controller setup / programming / etc with (like for KT, or Lishui, or some other brand controller than the Grin stuff), you'll need whatever display is specific to your controller (whatever comes with it, or whatever works with that specific controller and has the features you want).


Best recommendation is to state what controller, motor, and other parts you're getting, and what specific features, in complete detail, that you want the system to have, and how you want the whole system to work. Then we can probably tell if the controller has those features or if you need the CA to do it, or something else.

 

[nervagon]

 

[smitty2k1]

Thanks for the info and links. Definitely need to do more research, never realized the displays are actually the brains behind it all.

New comtroller options from Grin seem worth waiting on. The are sold out of the brake sensors I need for my hydros anyways.