TSDZ2 mid drive with 860C, 850C or SW102 displays only -- Flexible OpenSource firmware (Casainho code only)

[e3s]

For me it's better to have more assist climbing and less in the flat road.

Nm*rpm*x +speed*k with k negative. It Can be positive. The power is related to the speed so having the speed in the formula is important.
Power=2/3 speed^2.
To maintain higher speed you need higher motor power, not higher human power. Power assist is also safer for the internal gears.
It's better to include the last formula but buba said that the exponential can't be included

Nm*rpm*x+2/3 speed ^2 *k
Playing with x and k you can have more power assist at different speed
Hibrid torque and power assist could be
Nm*rpm*x+nm*y+speed*k letting the user to choose x, y and k.
If you multiply for 0 y is a power assist, for 0 x is a torque assist.
Simpler formula could be
Rpm*x+nm*y+speed*k

 

[rippledabs]

Dumb question. How do I compile the code to generate the hex/bin file? I have very little experience other than audruino scripts.

The cadence sensor is on the motor controller, it is a small piece of black plastic but it has some wires and a connector on the board. Maybe you can try to remove and connect again.

Or maybe take the v0.19.0 sources, on ebike_app.c, change the calculation of cadence:

static void read_pas_cadence(void)
{
  // cadence in RPM =  60 / (ui16_pas_timer2_ticks * PAS_NUMBER_MAGNETS * 0.000064)
  if((ui16_pas_pwm_cycles_ticks >= ((uint16_t) PAS_ABSOLUTE_MIN_CADENCE_PWM_CYCLE_TICKS)) ||
      (ui8_g_pedaling_direction != 1)) // if not rotating pedals forward
  { 
    ui8_pas_cadence_rpm = 0; 
  }
  else
  {
    ui8_pas_cadence_rpm = (uint8_t) (60 / (((float) ui16_pas_pwm_cycles_ticks) * ((float) PAS_NUMBER_MAGNETS) * 0.000064));
  }
}

To this:

static void read_pas_cadence(void)
{
  ui8_pas_cadence_rpm = 60; 
}

That will force a fixed value. Should make your system work using the torque sensor input only. But be careful, something may start up the motor, keep brakes pressed when not pedaling.

 

[mctubster]

Ok... For me the best option is power assist because the torque assist is already in the other levels.
Maybe the level can be changed at a certain rpm. Example my ideal rpm is 60. I go to 90 and the level changes automatically.
About the speed in the formula is only an idea to insert this variable.. Then if it's an exponential or a multiplication factor it's the same result.
Nice idea to consider the speed also in the torque assist formula!

The problem with using speed is that there is no context for it. Are you going up a hill? Down a hill? Do you have a headwind?

Torque provides context. If you want to speed up, you provide more torque, and then over time power increases.

I think there is a fundamental disconnect in usage. Two types of users

1 / User wants assistance for various scenarios when they cannot generate enough power themselves
2 / User wants continuous assistance to go faster than they are physically capable of

I am firmly user type 1. I want assistance to accelerate, to climb a steep hill, to fight a headwind. In this case the user is easily able to signal the desire through pushing harder on the pedals - ie torque.

User type 2 wants the motor to magnify their power output, at all times, even at steady state. Personally I don't agree with the philosophy and I would recommend a pedelec Bafang style motor for this type of usage. This is where a cadence sensor is good enough and the Bafang motors are superior in continuous power ratings. The TSDZ2 motor is very small and light. Its greatest attribute is the torque sensor. Unfortunately the mechanics are lightly engineered!

Anyway I will try Alpha8. I found eMTB level 3 slightly undercooked at peak power but eMTB level 4 overpowered. So will try around 5 on the new scale

 

[casainho]

Ok... For me the best option is power assist because the torque assist is already in the other levels.
Maybe the level can be changed at a certain rpm. Example my ideal rpm is 60. I go to 90 and the level changes automatically.
About the speed in the formula is only an idea to insert this variable.. Then if it's an exponential or a multiplication factor it's the same result.
Nice idea to consider the speed also in the torque assist formula!

The problem with using speed is that there is no context for it. Are you going up a hill? Down a hill? Do you have a headwind?

I also think is a bad idea to add wheel speed to emtb calculation if the motivation is to determine if we are going up a hill and so increase the assistance level at low speeds.

I usually find myself on flat terrain with other riders in front of me and I need to go slow, than I absolutely do not want increase of assist level on this situation. What I need is a good torque sensor calibration so I can pedal just slow cadence and with little torque, to have little motor assistance.
The same happens on critical sections where I need to go slow and with balance, otherwise If fall, I can be dead because of the cliff on one side or both sides - I absolutely do not want increase of assist level on this situation!!

Maybe in future, we can connect by Bluetooth to SW102 display, an external board with Arduino + IMU that tells the grade of ascending or descending and then will increase or decrease the assist level.

I think there is a fundamental disconnect in usage. Two types of users

1 / User wants assistance for various scenarios when they cannot generate enough power themselves
2 / User wants continuous assistance to go faster than they are physically capable of

Well, mode 1 is surely different from anything we have now!!

I am firmly user type 1. I want assistance to accelerate, to climb a steep hill, to fight a headwind. In this case the user is easily able to signal the desire through pushing harder on the pedals - ie torque.

I don't think this makes much sense as user will not "easily able to signal the desire through pushing harder on the pedals - ie torque" as he can't apply the same torque at different cadences so the result of motor assistance will vary, should be very hard for user to predict what torque it must do for every different cadence.

User type 2 wants the motor to magnify their power output, at all times, even at steady state. Personally I don't agree with the philosophy and I would recommend a pedelec Bafang style motor for this type of usage. This is where a cadence sensor is good enough and the Bafang motors are superior in continuous power ratings. The TSDZ2 motor is very small and light. Its greatest attribute is the torque sensor. Unfortunately the mechanics are lightly engineered!

On Bafang with PAS sensor only, on acceleration, user will need to provide much more power itself then on TSDZ2, because on Bafang the assistance level will only increase when the cadence increase and for that, will be solely the user energy used to accelerate. On TSDZ2, when user apply more torque even at constant cadence, the motor will immediately react and assist so the acceleration energy is shared between TSDZ2 and the user -- and this is a BIG difference, as user can be providing only 50 or 100 W and TSDZ2 providing 800 W, which is 16x times more!!!

See that I implemented startup BOOST in a way that it can use the max motor of 18 amps at startup and on next seconds, so we can take advantage of motor max power that is 2x more than regular power at 8 or 10 amps:

And to help user expectations, I wrote the bellow notes on wiki page: How to install the Flexible OpenSource firmware.

NOTES:
- Motor configurations: TSDZ2 motor configurations found on online shops (36V, 48V, 52V, whichever power in watts, with throttle or without) the motor controller is just the same. The only relevant difference is brushless motor inside, that there are only 2 different: 36V and 48V motor (52V systems use the 48V motor).
- Motor max power without damage by heating: the motor can handle only about 8 or 10 amps continuously (10 amps with a 52V battery is 520 watts) otherwise it will heat up fast and damage permanently the motor. Peaks of 18 amps can be used, like at startup or for short time but it is advised to install the motor temperature sensor so the firmware will be able to protect the motor, by reducing the motor current as soon the starts heating after a defined threshold.

 

[e3s]

With torque assist you'll never can go outside the saddle if the assist is too much. Many people has this need.
You don't feel anymore that the gradient changes, so if you have the highest gear the motor helps you with the maximum power,
pedalling is not natural anymore, you don't feel the road.
Taking a bit less in consideration the torque and more the rpm and the speed is safer for the motor

 

[casainho]

With torque assist you'll never can go outside the saddle if the assist is too much. Many people has this need.
You don't feel anymore that the gradient changes, so if you have the highest gear the motor helps you with the maximum power,
pedalling is not natural anymore, you don't feel the road.
Taking a bit less in consideration the torque and more the rpm and the speed is safer for the motor

Maybe I miss experience here.

Would be nice if would could somehow normalize the pedal human power and pedal torque output values and then we could insert any of them on the emtb curve. User could then change on display his preference for one or other.

 

[casainho]

Finally I finished the wiki update on How to install the Flexible OpenSource firmware: https://github.com/OpenSource-EBike-firmware/TSDZ2_wiki/wiki/How-to-install-the-Flexible-OpenSource-firmware

and now the recommended displays are 850C and SW102:

Pictures of 850C color LCD main screen and configurations screen:

Pictures of SW102 Bluetooth display main screen and configurations screen:

 

[buba]

In Power Assist, over a certain RPM, you only need to rotate the pedals with slight force to get assistance.

But the human power assist mode is pedal torque * pedal cadence, so if your pedaling very fast at 1NM x 100 RPM or slow at 10NM x 10 RPM, you are providing the same energy of 100 Watts and get the same workout!! Because we use gears and we are always changing them often on mtb, the best way for us to keep our pace of heart rate / workout with the different gears, is to use human power mode and not the torque mode. In torque mode we would change our heart rate / workout pace every different gear.

Actually there is a difference in the workout between Power Assist and Torque Assist. The energy you spend on workout is not an indication of what kind of workout you are doing. So:

1 Nm * 100 RPM is not the same workout as 100 Nm * 1 RPM

Even though it is the same power and calories the workout and effects of the workout are different.

The difference in experience is why some will prefer Power Assist and some Torque Assist.

Also, what is the motivation of adding as input the wheel speed to the emtb mode?

I did not indicate to add that. Just wanted to hear suggestions and try to get different viewpoints of riding mode philosophies. Was suggested to add speed in the algorithm.

Wanted to see if there was something I missed that would make eMTB really great for all users and basically have 100 % satisfaction.

 

[buba]

Ok... For me the best option is power assist because the torque assist is already in the other levels.
Maybe the level can be changed at a certain rpm. Example my ideal rpm is 60. I go to 90 and the level changes automatically.
About the speed in the formula is only an idea to insert this variable.. Then if it's an exponential or a multiplication factor it's the same result.
Nice idea to consider the speed also in the torque assist formula!

The problem with using speed is that there is no context for it. Are you going up a hill? Down a hill? Do you have a headwind?

Torque provides context. If you want to speed up, you provide more torque, and then over time power increases.

I think there is a fundamental disconnect in usage. Two types of users

1 / User wants assistance for various scenarios when they cannot generate enough power themselves
2 / User wants continuous assistance to go faster than they are physically capable of

I am firmly user type 1. I want assistance to accelerate, to climb a steep hill, to fight a headwind. In this case the user is easily able to signal the desire through pushing harder on the pedals - ie torque.

User type 2 wants the motor to magnify their power output, at all times, even at steady state. Personally I don't agree with the philosophy and I would recommend a pedelec Bafang style motor for this type of usage. This is where a cadence sensor is good enough and the Bafang motors are superior in continuous power ratings. The TSDZ2 motor is very small and light. Its greatest attribute is the torque sensor. Unfortunately the mechanics are lightly engineered!

Anyway I will try Alpha8. I found eMTB level 3 slightly undercooked at peak power but eMTB level 4 overpowered. So will try around 5 on the new scale

I think you are golden with that explanation! And with that, I think we will have to accept that we need different riding modes for different cases.

So I originally asked input from the community to see what riding mode eMTB should be based on. And seems like there should be two different eMTB modes to satisfy all users. Or at least a switch where users can choose between either Power or Torque logic. I do not want to add more complexity and "steal" more space so I think it is best to leave eMTB as is for now. In the future I can imagine it will be simple to add a switch where users can switch between the two different modes of logic.

 

[buba]

I am firmly user type 1. I want assistance to accelerate, to climb a steep hill, to fight a headwind. In this case the user is easily able to signal the desire through pushing harder on the pedals - ie torque.

I don't think this makes much sense as user will not "easily able to signal the desire through pushing harder on the pedals - ie torque" as he can't apply the same torque at different cadences so the result of motor assistance will vary, should be very hard for user to predict what torque it must do for every different cadence.

You actually do not need to predict anything. You just push the pedals harder and you get more assistance.

The difference lies in if you want constant assistance or assistance when there is a resistance. When there is resistance you will be able to apply torque on the pedals and get more assistance. When in Power Assist you will always get assistance even if there is minimal resistance because the cadence can be high.

I do not need the temperature sensor when using Torque Assist. It helps me when I need it the most and I can easily get full power when necessary. Speeds of 50 km/h are reached without a problem so there are no issues with assistance. As soon as I ride normally the assistance is minimal and proportional to my force on the pedals.

I actually like both modes for different situations but would love to be able to combine them into one single mode! I recommend everyone to try out and evaluate all the riding modes. Pick whatever mode feels best and enjoy! I was just hoping that eMTB would be a great complement to the default modes but it will certainly be better with the torque sensor calibration.

 

[elem]

Hello, just finishing testing V.20.A8, small MTB ride full eMTb mode .... fantastic

set up standard with power mode 5 lvl
startupboost disabled
acceleration 25%
eMTB 7 sensibility ( tested firstly 5= too hard for me, for powerfull performer only, 10 default mode =too esay for me soo i tested 9 and after 8 and finaly 7 witch is perfect for my sport use, 8 will be my "trail" mode )

noticed : motor 48V with standard mode, assiting too 96 rpm fully charged ( 50v as i am using actually 12S battery ) 92 rpm at 47V .

mean : don t need experimental mode ( don t match with 48V motor only usefull with 36V motor )
but using 14S battery is the best way .

Yesterday was my last test with my son'bike ebosch powered ( witch the bike itself is a "war machine " ) full ride with eMTb mode
some MTb climb impossible with my bike using v0.19 was easy with this bike .
today with V0.20A8 i tested those climb i wasn t able to climb with 0.19, now using 0.20A8 it's ok

what i like with OSF : i'm able to desactivat start up boost and adjust acceleration witch isn't possible with eBosch system ( may be possible but reselers don't have acces to this kind of parameters exepted maybe with ebosh team in world cup ) the only things possible with ebosch is to change the default power mode ( need purion display=550$ or going to a reselers with the ebosch program : from 0€ if you' r a regular client to more than 25€ each time ) .

Soo big "BRAVO", it' s a HIT .

PS : using eMTB assist was fluctuating from 30W to max avaiable .
my set up : 14A and 550W max, 12s battery 590Wh .

test was 14.5 km 285 climb, 92 wh used ( from display )

 

[Rafe]

I am making an alpha where you can set the button bounce time. In your case your would want it to be 0 so it deactivates immediately. I hope that is okay with you.




I want that smile back so please test eMTB and let me know if you are happy with it after changing the sensitivity so it fits your old sensitivity value. I will await your feedback.


--------------------------------


I have set more values and tried to improve the eMTB response so all users can find a perfect value. If anyone feels the old eMTB is better in any way please let me know. But before doing so. Configure and find your new sensitivity value as it is not transferable from the Alpha versions before Alpha 8.

If you previously had 2 in sensitivity try using 3 - 6 in Alpha 8
If you previously had 4 in sensitivity try using 8 - 10 in Alpha 8
If you previously had 6 in sensitivity try using 10 - 12 in Alpha 8

Hi Buba, I had eMTB sensitivity set on 3 in A7 and 6 now seems closest in A8

I think you may need to look at the debounce again. As you asked I have tried A8 out on the route that I use walk assist. Very steep steps in flights of about 6 or 7 steps cut into a cliff. I set assist level to 9 at the bottom as I always do and lifting the front wheel up to start each flight and pushing the saddle to help get some traction I also always use panniers and run a heavy bike. Then I noticed that the assist was getting weaker and weaker and this was because the assist level had dropped down to 5 so I upped the assist level back to 9 and it was not long before the assist level had fallen down again. When I got to the top I discovered that this was due to fast finger bounce on the walk assist button (the down button) manually lowering the assist level.

I haven't opened the motor yet but there was a discernable rumble that was not there before I climbed the steps so I would guess the blue gear I fitted last week had a lot of shear force from all this, remember I run a heavy bike.

 

[buba]

Hello, just finishing testing V.20.A8, small MTB ride full eMTb mode .... fantastic

set up standard with power mode 5 lvl
startupboost disabled
acceleration 25%
eMTB 7 sensibility ( tested firstly 5= too hard for me, for powerfull performer only, 10 default mode =too esay for me soo i tested 9 and after 8 and finaly 7 witch is perfect for my sport use, 8 will be my "trail" mode )

noticed : motor 48V with standard mode, assiting too 96 rpm fully charged ( 50v as i am using actually 12S battery ) 92 rpm at 47V .

mean : don t need experimental mode ( don t match with 48V motor only usefull with 36V motor )
but using 14S battery is the best way .

Yesterday was my last test with my son'bike ebosch powered ( witch the bike itself is a "war machine " ) full ride with eMTb mode
some MTb climb impossible with my bike using v0.19 was easy with this bike .
today with V0.20A8 i tested those climb i wasn t able to climb with 0.19, now using 0.20A8 it's ok

what i like with OSF : i'm able to desactivat start up boost and adjust acceleration witch isn't possible with eBosch system ( may be possible but reselers don't have acces to this kind of parameters exepted maybe with ebosh team in world cup ) the only things possible with ebosch is to change the default power mode ( need purion display=550$ or going to a reselers with the ebosch program : from 0€ if you' r a regular client to more than 25€ each time ) .

Soo big "BRAVO", it' s a HIT .

PS : using eMTB assist was fluctuating from 30W to max avaiable .
my set up : 14A and 550W max, 12s battery 590Wh .

test was 14.5 km 285 climb, 92 wh used ( from display )

test eMTB.jpg

Elem, absolutely wonderful to read about that! Thank you for sharing your experience with us and providing very interesting comparisons! Am very happy that you are happy!

 

[buba]

I am making an alpha where you can set the button bounce time. In your case your would want it to be 0 so it deactivates immediately. I hope that is okay with you.




I want that smile back so please test eMTB and let me know if you are happy with it after changing the sensitivity so it fits your old sensitivity value. I will await your feedback.


--------------------------------


I have set more values and tried to improve the eMTB response so all users can find a perfect value. If anyone feels the old eMTB is better in any way please let me know. But before doing so. Configure and find your new sensitivity value as it is not transferable from the Alpha versions before Alpha 8.

If you previously had 2 in sensitivity try using 3 - 6 in Alpha 8
If you previously had 4 in sensitivity try using 8 - 10 in Alpha 8
If you previously had 6 in sensitivity try using 10 - 12 in Alpha 8

Hi Buba, I had eMTB sensitivity set on 3 in A7 and 6 now seems closest in A8

Hi Rafe, does it still put a smile on your face? :wink:

I think you may need to look at the debounce again. As you asked I have tried A8 out on the route that I use walk assist. Very steep steps in flights of about 6 or 7 steps cut into a cliff. I set assist level to 9 at the bottom as I always do and lifting the front wheel up to start each flight and pushing the saddle to help get some traction I also always use panniers and run a heavy bike. Then I noticed that the assist was getting weaker and weaker and this was because the assist level had dropped down to 5 so I upped the assist level back to 9 and it was not long before the assist level had fallen down again. When I got to the top I discovered that this was due to fast finger bounce on the walk assist button (the down button) manually lowering the assist level.

Good points! Will fix and deactivate the button events during Walk Assist and Cruise!

I haven't opened the motor yet but there was a discernable rumble that was not there before I climbed the steps so I would guess the blue gear I fitted last week had a lot of shear force from all this, remember I run a heavy bike.

Yes, it is very vital to use Walk Assist with as low gear as possible to reduce the strain on the blue gear. But even then it could cause problems at very low speeds. The slower the motor rotates the more of a problem this will cause. Low gear -> motor rotates faster -> blue gear rotates faster -> less problems.

Will add this to the wiki!

 

[e3s]

I explain my point of view with some examples and numbers.
Torque assist is ok if you use the full power. You can reach the desired speed without problem.
And if I use a partial assist? If I want to push 10kg @ 60rpm, are more or less 90w and 1.5nm nearly. I need another 1.5nm. I want to go at 30kmh. If I have an higher torque assist I can't push 90w, but much less. I'm not pedaling, I'm going on a motorcycle. If I go at 40kmh I need 300w of power, but I have to push 150w, 17-18kg @ 60rpm. Too much. The motor gives another 17-18kg. In this condition how the torque assist can help me to push only 90w without the speed variable?
@mctubster: fighting with the wind and going uphill is the same. You go slow with lower gear and an high human power, going downhill the speed is high and the human power is low. In a wind condition, you have to stop the bike without higher assist. If I select a assist ratio 2:1 how can I have an higher assist in emtb mode? I have to change the level. The philosophy of the emtb is to not change the level. The need is have an assist at a different speed, not at a different torque. Speed is related to the power. Without this variable you can't know if you are going uphill, downhill or in a flat road. Bosch emtb gives more assist when climbing, because in my opinion it reads the speed and the power, low speed and high power means only two things: a climb, or upwind

 

[andrea_104kg]

pedalling is not natural anymore, you don't feel the road.

I partly agree. However the torque mode is oscillating, on my specialized brose it is not so. Help is constant during pedaling as in power mode. Perhaps in torque mode it would be enough to slow down the fall of the watts at each pedal stroke to end it rounder. A few tenths of a second would be enough.

 

[andrea_104kg]

hummm ALL the motor, bosch, brose ecc, they are a combination of cadence and torque, any only torque. I don't like the idea of speed, the magnet passes on the sensor only once per lap, in off-road measuring speed is a problem ...

 

[e3s]

@.7.2 kmh in a 26" it passes 1 time per second
@ 21.6kmh the speed is updated every 0.33 seconds
Someone has a square pedal and doesn't use the femoral biceps so there are more peaks of assist in the torque mode with low rpm. I think that power assist in this condition is better
Each one has their own human power. To maintain it it's possible to add a value: minimal human power, under this value, the motor doesn't help. Maybe also maximal human power, in this case there's no need of the speed variable. If you go over the maximal human power, the assist ratio goes high and remain until you go under the minimal human power. So you can have a throttle only going under or over these parameters.

 

[buba]

I previously asked the community for the ultimate riding mode. A riding mode that would please everyone and would be theoretically perfect. eMTB would preferably be based on this riding mode.

Power Assist is really great and shines when the cadence is above a certain level. In its optimal range it filters the torque sensor value and gives a nice response in the upper cadence range. But it is lacking in the lower cadence range where it needs Startup Boost to assist better and feel more responsive.

Torque Assist, on the other hand, works great during low cadence, high torque situations. It does not need Startup Boost but is lacking when the cadence is above a certain threshold.

So, one is better in the low cadence range and the other in the high cadence range. Both work in the entire operating range but they have their optimal range in different places.

Below is a very simplified graph trying to illustrate a possible difference in the operating range between the two riding modes. Imagine changing what mode is more dominant as the cadence is increasing. The blue curve is Torque Assist and the red curve is Power Assist.



If we were to implement a riding mode where Torque Assist is dominating when cadence is low but it slowly transitions to Power Assist with increased cadence we would have a very interesting riding mode. It would be designed with the best characteristics from both Torque and Power Assist.

With all that said, I am not declaring that I will be changing anything so I hope no one is worried. I am not even saying that this is the best implementation. Just wanted to throw this out there because I found the idea to be interesting.

 

[e3s]

Why the power assist curve goes up? The rpm is more multiplied in the formula?
One time it's reached the cruise rpm torque assist isn't necessary.
I have this style of ride: I need the assist on the climbs and to restart on them, to help me to have a higher cruise speed also on a flat road. At low speeds or restarting I don't need assist on the flat. For this reason on a flat road I don't like torque sensor. It's a waste of energy. Maybe if everyone says its style you can improve the algorithm.

 

[buba]

Why the power assist curve goes up? The rpm is more multiplied in the formula?

More torque and more cadence means that the Power Assist assists more. The torque actually decreases with higher cadence but the cadence increases faster so the assistance you get is still more -> curve goes up.

(With Torque Assist you only look at torque so at a higher cadence you get less assistance unless you push harder and apply more torque on the pedals. But no matter what, the torque is never zero when the pedals are rotating so it has an offset.)

 

[e3s]

Ok. Understood. If the relation between human power and motor power is always the same, there are some condition that is difficult to maintain only a level.

 

[thineight]

I think you may need to look at the debounce again. As you asked I have tried A8 out on the route that I use walk assist. Very steep steps in flights of about 6 or 7 steps cut into a cliff. I set assist level to 9 at the bottom as I always do and lifting the front wheel up to start each flight and pushing the saddle to help get some traction I also always use panniers and run a heavy bike. Then I noticed that the assist was getting weaker and weaker and this was because the assist level had dropped down to 5 so I upped the assist level back to 9 and it was not long before the assist level had fallen down again. When I got to the top I discovered that this was due to fast finger bounce on the walk assist button (the down button) manually lowering the assist level.

Good points! Will fix and deactivate the button events during Walk Assist and Cruise!

Yes, I mentioned this "issue" in one of my previous posts:
- during the w.a. the DOWN button should not change the current assistance level
- of course the brake sensors shall reset everything, if activated

The use of the DOWN button shall of course be restored once the walk assist is interrupted (whether if brake sensors are applied or just the DOWN button is released for the proper amount of time).

At this point I will wait directly the A9 to test the latest features (I'm currently on A7).
Thanks :thumb:

 

[maximusdm]

Did some more tests. I just installed the motor on a new MTB 29".
I am testing A7 with the default assistance levels.
I tested power mode because I wanted to see improvements form 0.19 first, at the moment only on flat city road.

1. Acceleration. Previously I used 8A/s and was very pleased. Now I am using 80 and seems to me a smaller acceleration.
@Buba can you please detail what is the corresponding between old mode and new one?
2. Startup boost does not work for me. Tested with 40-60-80 sensitivity. On flat this is nice to have, but on steep hill it is a must if you stoped and want to start with a 25kg bike. I think the problem here is that on old boost mode you could set a separate assistance level, but still I cannot feel no difference between 0 and 80.
3. Power delivery. On motor mode 0 (48 std) it is impossible to get 600W+ on flat with assistence 9 whatever you do. With mode 2 (48 HC) it is better, with my custom HC mode it is a BEAST! On 5 assist i get constant 800w+ at 45km/h, on huge MTB tires. With road tires i assume i can hit 50km/h with small effort. Obvious it is burning the battery but what is important is if you need speed, it can provide.

One good thing in 18.2 and 19 was that it was possible to make the motor behave violent . Starting with boost of 800W is . Now even with 9 assist it is starting with 150w
Overall impression is a more powerful motor, but it delivers the power on a linear way with smaller angle and acceleration is slower. Can this be configured or it is possible only with software changes?

Thank you!

 

[buba]

I think you may need to look at the debounce again. As you asked I have tried A8 out on the route that I use walk assist. Very steep steps in flights of about 6 or 7 steps cut into a cliff. I set assist level to 9 at the bottom as I always do and lifting the front wheel up to start each flight and pushing the saddle to help get some traction I also always use panniers and run a heavy bike. Then I noticed that the assist was getting weaker and weaker and this was because the assist level had dropped down to 5 so I upped the assist level back to 9 and it was not long before the assist level had fallen down again. When I got to the top I discovered that this was due to fast finger bounce on the walk assist button (the down button) manually lowering the assist level.

Good points! Will fix and deactivate the button events during Walk Assist and Cruise!

Yes, I mentioned this "issue" in one of my previous posts:
- during the w.a. the DOWN button should not change the current assistance level
- of course the brake sensors shall reset everything, if activated

The use of the DOWN button shall of course be restored once the walk assist is interrupted (whether if brake sensors are applied or just the DOWN button is released for the proper amount of time).

At this point I will wait directly the A9 to test the latest features (I'm currently on A7).
Thanks :thumb:

Will fix! I actually read your great suggestions. But when developing and testing I noticed that with enough debounce time I had time to switch assist level without disengaging Walk Assist. Was almost like a feature that I thought would be appreciated when going uphill or in challenging terrain. The assistance does not have to disengage to increase or decrease the assist level. And it is set so that if there is a single button bounce it will not change anything. But if there are multiple bounces it will change.

However, I understand that this is maybe not that necessary so will of course disable the buttons! Maybe only disable the DOWN button and keep the other buttons active? Regardless, I understand the problem reported and will fix!