Has anyone heard of or tried Toseven mid drive?

[Bogmonster666]

I'm not one to spin at a high cadence but as I understand it the OEM firmware cuts assistance around 90rpm but the OSF takes it to around 120rpm.

Jbalatutube on YT has a vid demonstrating this.

Indeed, my understanding is that OSF supports 'field weakening' to allow a motor to operate at a wider speed range without changing its voltage. Not really looked at the physics of it but it is a real thing. I believe the OSF hybrid mode works really well to give you the benefits of torque sensing at lower cadence and more of PAS at high cadence where it uses field weakening. You kind of get to have your cake and eat it.

I'm not particularly good at judging my cadence but looking at my speed, wheel size and gearing, I was comfortable with 90 rpm ( and on my exercise bike it tells me I'm doing 90 to 100 rpm when I'm performing at my optimum efficiency). Unfortunately the motor with stock firmware is delivering little at this cadence - and it felt considerably worse with voltage sag as the battery charge depleted. I found myself needing to drop to high 60s rpm on steeper sections as the battery depleted and this was causing unwanted knee strain. Obviously I can install lower gearing and accept lower speed which is what I will do.

It feels like some really clever folks have worked wonders with osf. It's just a shame that the stock firmware / kit does not deliver this. I have little doubt I will end up flashing the motor in due course, but will leave stock whilst plenty of warranty period left.

And that was my point, the flashing has some risks, requires IT skills (which I fortunately have), voided warranty etc. A factory delivered unit that delivered this would be a big bonus (I know eco cycles supply flashed motors).

C

 

[guerney]

A factory delivered unit that delivered this would be a big bonus (I know eco cycles supply flashed motors

Eco Cycles – Opening Soon

cycles.eco

"I agree to assume any and all liability for risks, damages, and claims associated with the use of Open Source Firmware, releasing and indemnifying Eco Cycles LLC from any and all causes of action. "

"While Eco Cycles does offer our Warranty on this Motor Kit, note that the Warranty does NOT cover Firmware Issues, or Overheating. That being said, OSF is very functional and stable nowadays, and as long as you mind your gearing and don't max out the power (we send pre-programmed to a generally reliable level), you should not have any issues. "

 

[Bogmonster666]

Do they offer warranty?

Eco Cycles – Opening Soon

cycles.eco

"I agree to assume any and all liability for risks, damages, and claims associated with the use of Open Source Firmware, releasing and indemnifying Eco Cycles LLC from any and all causes of action. "

Sorry, I wasn't trying to suggest eco cycles offered warranty, just avoids the need to flash the motor yourself. And even if there was a warranty, all the import pain and then the pain of returns to the US would fry my head....

 

[guerney]

Sorry, I wasn't trying to suggest eco cycles offered warranty, just avoids the need to flash the motor yourself. And even if there was a warranty, all the import pain and then the pain of returns to the US would fry my head....

Their programming cable looks convenient:

Eco Cycles – Opening Soon

cycles.eco

Eco Cycles – Opening Soon

cycles.eco

Pre-flashed is available too:

Eco Cycles – Opening Soon

cycles.eco

 

[alexfnoble]

Early days but I think the biggest flaw with the tsdz2 for many users is the assistance cutoff / winding that tends to favour lower cadence. I know some folks use a 36v winding motor at 48v for this reason. This is particularly noticeable going up hills where I want to spin faster but find the assistance kicks in at lower cadence. This is where I would be looking for an improvement.

C

We've improved that with cadence of up to 120rpm out of the box and no need to flash. Our FOC Controller also allows for much improved motor control. Paired with improvements like our new strain based torque sensor the overall user experience is much better.

I know some of those who like to tinker are really excited about the FOC controller as they have been reaching out for more details.

 

[Woosh]

the TSDZ2 has also FOC controller. Can you give us more details about your strain gauge torque sensor? Drawings will be useful. is it fitted to both the DM01 and DM02 and available as spares?

 

[alexfnoble]

the TSDZ2 has also FOC controller. Can you give us more details about your strain gauge torque sensor? Drawings will be useful. is it fitted to both the DM01 and DM02 and available as spares?

I was under the impression it didnt have a true foc controller. Stand corrected if wrong, i will look into that further

Optimized TSDZ2 Firmware

I start this new thread to share the results of my TSDZ2 Motor Firmware optimization work. This work started out as a low priority activity, but along the way I realized that there were many things that could be improved, and in the end, I have to say that the result went beyond my...

endless-sphere.com

Yes we have a stain based torque sensor vs a electromagnetic induction torque sensor and its fitted on both the DM-01 and DM-02.


We will have a full range of spares available. I understand that customers on forums like this might be happy to service themselves but we intend to set up service centers for the majority of our customers to ensure a long lifespan.

 

[Woosh]

Yes we have a stain based torque sensor vs a electromagnetic induction torque sensor and its fitted on both the DM-01 and DM-02.

so what's different compared to the TSDZ2's torque sensor? The TSDZ2 uses two coils, the induction current in the induced side gives the torque measurement:

 

[Bogmonster666]

The DM02 looks to be the more logical choice for the UK market with a 250w version and somewhat lighter. I see the rpm is different on both incarnations. If can genuinely handle a higher chance out of the box, has spares, is as or more reliable than the tsdz2 then if sounds promising. Lots of ifs there and time will tell I suppose.

C

 

[Bogmonster666]

Incidentally, all the hub drives I see on the website are not UK legal for use in public places as all over 250w nominal. I'm not up with eu and Australian law but I suspect that a few market opportunities will be dented without a 250w labelled hub motor.

C

 

[alexfnoble]

Incidentally, all the hub drives I see on the website are not UK legal for use in public places as all over 250w nominal. I'm not up with eu and Australian law but I suspect that a few market opportunities will be dented without a 250w labelled hub motor.

C

Sorry about that we do have 250w hubs. Just not currently listed. Its no problem for us to produce hubs in any wattage needed by a partner business/company

 

[guerney]

@alexfnoble : If the new axle is similar, needs to be as flexible as the TSDZ2's, I fear the axle snapping problem will remain - why not use piezoelectric torque sensors instead?

 

[Woosh]

I can't see piezo-electric effect being used as torque sensor in the context of a crank drive motor though. The pedalling torque is usually translated into an angular displacement of the chainring versus the crank. That has been used in countless designs. To measure the displaced angle, a Hall sensor and a magnet or an IR LED and a photocell (Suntour HESC) or an induction coil in the case of the TSDZ2 are the most straightforward methods. Recently, you can get Arduino with BLE 5 and wifi for under £5 so that opens up the possibility of building the torque sensors directly into the pedals using 10 pence a piece strain gauges. I have seen strain gauges used in the cassette and chain tensioners but the results are not very good.

 

[Blacklite]

Hello Alex, you worry me a little. If for example your company is sued for patent infringement and loses, the victor can ask me to deliver up all unsold products and list of all those who bought infringing products. I will have to refund my customers. It's important to know that your products are covered by your own patents.

What patents do you think would cover a motor driving reduction gearing driving a chainwheel?
It’s hardly something new and novel that would be able to be patented.

 

[Woosh]

there are several aspects that need protection in a new crank drive motor:
1. the shapes of the casing, the dimensions of the motor, commonly known as the design elements
2. the main sensors.
There are 3 types of sensors in a crank drive motors: a torque sensor, a movement detector and a cadence sensor.
The torque sensor gives you a measure of pedalling effort, a movement detector tells you that the bike is moving and when it is safe to start the motor, a cadence sensor lets you know how much magnetic field strength is optimal so the controller can adjust the current angle versus the motor's Hall sensors.
Those sensors are not obvious so subject to patent protection.

 

[Blacklite]

there are several aspects that need protection in a new crank drive motor:
1. the shapes of the casing, the dimensions of the motor, commonly known as the design elements
2. the main sensors.
There are 3 types of sensors in a crank drive motors: a torque sensor, a movement detector and a cadence sensor.
The torque sensor gives you a measure of pedalling effort, a movement detector tells you that the bike is moving and when it is safe to start the motor, a cadence sensor lets you know how much magnetic field strength is optimal so the controller can adjust the current angle versus the motor's Hall sensors.
Those sensors are not obvious so subject to patent protection.

A cadence sensor, which measures the rate of rider pedal stroke, is definitely not used to determine optimal magnetic field strength. The two are not related at all. A BLDC motors optimal magnetic field angle in the stator winding is derived from the rotor position of the motor (so either hall sensors in the motor or by sensing minimums in back-emf in the motor phases in a sensorless controller), not from measuring crank cadence, particularly after a large level of reduction gearing and most likely a one way clutch between the crank and the BLDC motor.
There are also plenty of crank drive motors without torque sensors - the Bafang BBS series for example.

 

[Woosh]

The two are not related at all. A BLDC motors optimal magnetic field angle in the stator winding is derived from the rotor position of the motor (so either hall sensors in the motor or by sensing minimums in back-emf in the motor phases in a sensorless controller), not from measuring crank cadence, particularly after a large level of reduction gearing and most likely a one way clutch between the crank and the BLDC motor.

Google 'field weakening control'.
It's a technique used by OSF TSDZ2 firmware and others to extend the cut off cadence, beyond which the assistance disappears.
The technique advances the phase of the drive current, causing a change in the two motor constants (motor (Km) and voltage (Kv)) thus shifting the sweet zone to higher cadence.
Field Weakening - What, How, and Why to do It | Power Transmission Engineering Magazine
Motor constants - Wikipedia

 

[Blacklite]

Google 'field weakening control'.
It's a technique used by OSF TSDZ2 firmware and others to extend the cut off cadence, beyond which the assistance disappears.
The technique advances the phase of the drive current, causing a change in the two motor constants (motor (Km) and voltage (Kv)) thus shifting the sweet zone to higher cadence.
Field Weakening - What, How, and Why to do It | Power Transmission Engineering Magazine
Motor constants - Wikipedia

I understand what field weakening is, and how it is applied. But it doesn’t use any information from a cadence sensor, that as I said is sensing pedal cadence after a reduction drive and one way clutch. I’m very familiar with the TSDZ2 OSF code, and can happily point you to the relevant lines of code that are responsible for field weakening, and happy to point out how there is absolutely no reference to detected pedal cadence in any of those lines. In the OSF code the references in that section are motor ERPS, the PWM duty cycle (target and actual) and the actual and target motor current. Would you like the line numbers in motor.c for a few of the different forks of the OSF so you can check for yourself?

 

[Woosh]

Would you like the line numbers in motor.c for a few of the different forks of the OSF so you can check for yourself?

yes please.
My reply is in the context of a general crank motor which may be a Bosch (which I believe uses field weakening too) or others that may have a cadence sensor. The TSDZ2 does not have a cadence sensor.

 

[Blacklite]

yes please.
My reply is in the context of a general crank motor which may be a Bosch (which I believe uses field weakening too) or others that may have a cadence sensor. The TSDZ2 does not have a cadence sensor.

The TSDZ2 most certainly does have a cadence sensor. It's connected to GPIO_PIN_0 and GPIO_PIN_7.
The cadence drive mode in some of the OSF's, and the derivation of required assistance power in the power mode (that uses torque multiplied with cadence to arrive at a value) require a cadence sensor.
The relevant lines for field weakening in the Emmabrusa firmware (I'm choosing this as the most up to date and I think common fork) are lines 646 through to 701 in this file https://github.com/emmebrusa/TSDZ2-Smart-EBike-1/blob/master/src/controller/motor.c
Obviously the line numbers will change with different forks.