Bafang m510

[SchadowMaster]

I am testing the engine but so far I am very impressed.

 

[temon10]

Do U guys think that bafang can decrease the calibration time like M500/M600 by updating the firmware?

 

[Waynemarlow]

Do U guys think that bafang can decrease the calibration time like M500/M600 by updating the firmware?

Yes and no. It depends on how and why Bafang have chosen to calibrate the torque sensor every time on start up. They may think that they will get a far better and more linear sensor by first " calibrating " the sensor with all its associated weight and drag of the pedals and cranks attached ( not drag as in rotational drag ) that will give probably a more progressive sensor reading particulalry at very light readings.

I suspect that on the M600 I have that Bafang only use part of the sensor curve and once it gets to a particular reading and cadence then the motor simply moves to cadence only eliminating the human ergonomic problem of creating torque at higher cadences ( we are like an electric motor where we can create max torque at 0 rpm, its much harder to create that same torque whilst rotating the pedal at higher cadences ). If that section of the curve whilst under very light loadings is not very linear then you have to either calibrate the sensor to get optimum results or add firmware / electronics to compensate, but every sensor may not be the same, hence the calibration on startup.

From the TSDZ2 engine the firmware guys even went further and setup a calibration routine that was individual to the users weight to get an even better optimisation of what is tiny tiny tiny voltage variations from the torque sensor, with both no load and a full loading of the users entire weight on one pedal. Only by doing this can you work out the " range " of the sensor and thus capitilise on a full and maximum conversion of analogue voltage to a digital signal.

On the other hand Bafang may simply have screwed up and there is a firmware bug which simply loops continuosly until say there is a software timeout forcing the firmware to abandon the software loop.

My view is that you will always get a better more optimised light loading sensor if you calibrate it on startup. Put up with the startup delay guys and you may have a better more responsive motor.

 

[Waynemarlow]

I notice one of the developers of the TSDZ2 engine and who is now looking at his M500 has commented the following re the startup calibration issue.
That is very strange, because as we know, the only connection between the torque sensor and the motor controller is CAN. And on CAN, we see that the torque sensor sends every 10ms his value to the CAN bus to se motor controller can read the value from the bus. And we always see a value of 750 at 0kgs, on my M500 as also other developers M500, so, seems they are calibrated on factory.

Bafang M500/M600 thread

Since this seems to be an interesting motor, why not have a separate topic for it? I'll start with three quotes I have found about the motor on this forum, feel free to add more information that you can find. Over the last couple of days I had heard that there had been reports of the M600...

endless-sphere.com

This guy knows way way way more than I on the subject, maybe it can be sorted easily with a firmware update ?

 

[captainhightop]

Is there an optimum delay to have between power on and putting feet on pedals? I'm finding although it no longer cuts out completely, it does still occasionally drop out to 50% of the assistance. Almost like the dynamic range of the of the torque sensor is not very granular. I'm not totally disappointed, but it's very far from perfect. I've tried varying delays from 30sec to 120secs and other than the wattage no longer dipping to zero it's still unpredictable. When the wattage dips to 50% it feels like someone has thrown an anchor off the back of the bike, not a gradual attenuation at all.
I suppose a CAN bus logger could maybe log the torque sensor data to give an objective view of this, but this has to be a software problem that Bafang should fix like the battery voltage / power attenuation thing.


Both my M510 motors had identical behaviour (as much as you can tell without measuring it in some way).

 

[thaeber]

Is there an optimum delay to have between power on and putting feet on pedals? I'm finding although it no longer cuts out completely, it does still occasionally drop out to 50% of the assistance. Almost like the dynamic range of the of the torque sensor is not very granular. I'm not totally disappointed, but it's very far from perfect. I've tried varying delays from 30sec to 120secs and other than the wattage no longer dipping to zero it's still unpredictable. When the wattage dips to 50% it feels like someone has thrown an anchor off the back of the bike, not a gradual attenuation at all.
I suppose a CAN bus logger could maybe log the torque sensor data to give an objective view of this, but this has to be a software problem that Bafang should fix like the battery voltage / power attenuation thing.


Both my M510 motors had identical behaviour (as much as you can tell without measuring it in some way).

Sorry to hear mate! I don’t observe anything like this. My motor is running flawless.

 

[captainhightop]

Sorry to hear mate! I don’t observe anything like this. My motor is running flawless.

I wonder if it's because I run shorter cranks (150mm)? My cadence is super consistent, so I'm pretty sure it's not that.

I've not got a lot of options till the replacement M600 controller turns up, but although I love the fact that the M510 is so much lighter and seems more efficient, I'm really bugged by this issue, it's more of an on road issue (boring fire roads/flat trails) etc. when commuting or going to the off road trails, or maybe I don't notice it so much during off road adventures, as I'm usually holding on for dear life!

 

[thaeber]

I run 160 mm cranks. Hard to judge if the 7-8% reduced torque of the 150 mm cranks really makes a difference. Do You run the same firmware (latest on GitHub) as I do?

 

[captainhightop]

I run 160 mm cranks. Hard to judge if the 7-8% reduced torque of the 150 mm cranks really makes a difference. Do You run the same firmware (latest on GitHub) as I do?

I'm running the original firmware, not the latest from GitHub, so might be worth the risk of updating again, although I'm a bit wary after trashing my previous M510 controller, and then somehow killing the M600 controller when it all went wrong last time.

On reflection, I'll probably wait to try the one you're using until I have the new M600 controller in hand and installed (mid July I'd estimate).

I have got some 170mm cranks though, trying them will be low risk (apart from the inevitable pedal strikes!).

I'm pleased you did the math on the difference in torque though, that doesn't seem enough that it would make a difference, although I'm a software engineer, not a mechanical engineer so what do I know!

 

[SchadowMaster]

I will share with you my opinion about the new Bafang engine M510
In fact, I wasn't expecting this huge improvement in performance, power and energy savings, especially since my experience with the M500 engine was not satisfying in some points, but the new engine is so powerful that I can climb hills for a long time without any problems. The acceleration is also pretty awesome with this engine.
I like that the torque sensor is very sensitive to the movement of the pedals.
I did the first 40 km with only 30% of the battery and most of the time I used the 5 level with several hill climb tests, but i noticed that the engine shuts off when I exceed 26 km/h even though I set the maximum speed to 25 km/h did you notice that too?

 

[thaeber]

I will share with you my opinion about the new Bafang engine M510
In fact, I wasn't expecting this huge improvement in performance, power and energy savings, especially since my experience with the M500 engine was not satisfying in some points, but the new engine is so powerful that I can climb hills for a long time without any problems. The acceleration is also pretty awesome with this engine.
I like that the torque sensor is very sensitive to the movement of the pedals.
I did the first 40 km with only 30% of the battery and most of the time I used the 5 level with several hill climb tests, but i noticed that the engine shuts off when I exceed 26 km/h even though I set the maximum speed to 25 km/h did you notice that too?

Yes, same here and pretty normal, with other manufacturers also. With my M510 the engine support is drastically reduced when hitting the 25 km/h limit, only very little support remaining up to 26 km/h. I remember that from legislation perspective this is ok until +10% (27.5 km/h). My buddys Fantic XEF 1.9 with Brose Mag S is a little faster then my E10, pushing the bike up to 27 km/h. Always forcing me to use the boost button

 

[krokan]

Yes, same here and pretty normal, with other manufacturers also. With my M510 the engine support is drastically reduced when hitting the 25 km/h limit, only very little support remaining up to 26 km/h. I remember that from legislation perspective this is ok until +10% (27.5 km/h). My buddys Fantic XEF 1.9 with Brose Mag S is a little faster then my E10, pushing the bike up to 27 km/h. Always forcing me to use the boost button

How do you enable the boost feature? I have DP C241 but there is no "Boost" button, only power button.

 

[thaeber]

How do you enable the boost feature? I have DP C241 but there is no "Boost" button, only power button.


View attachment 90351

Looks like the boost function is not available on Your setup. I‘m using the old control unit and DC241 display of my M500, only replaced the motor itself with M510.

 

[ficorama]

today's ride. consumed 25% of battery. The m510 runs smoothly.

 

[Pantofla]

How is the noise of M510 when compared to M500?

 

[SchadowMaster]

How is the noise of M510 when compared to M500?

The noise is a little higher than M500 but the difference is not much.

 

[thaeber]

Today I rode the bike until the battery died. That was a real challenge, because between 10% and 0% the motor only provideed 200W. That’s enough to climb additional 300m and ride 10km. Another hour for me. The battery cut off at 41V. So overall, I‘m very happy with the attenuation. Attached 2 pics of the SoC and related Volt at 1% SoC. I had to stop for the pictures. So Voltage was appr. 0.5 - 0.8 Volt lower under pedaling.

 

[temon10]

Today I rode the bike until the battery died. That was a real challenge, because between 10% and 0% the motor only provideed 200W. That’s enough to climb additional 300m and ride 10km. Another hour for me. The battery cut off at 41V. So overall, I‘m very happy with the attenuation. Attached 2 pics of the SoC and related Volt at 1% SoC. I had to stop for the pictures. So Voltage was appr. 0.5 - 0.8 Volt lower under pedaling.

View attachment 90801

View attachment 90802

your M510 already reach 3000 km?

 

[thaeber]

No. I swapped engine at 2.400 km. So 600 km on my M510 now.

 

[ficorama]

Yes, same here and pretty normal, with other manufacturers also. With my M510 the engine support is drastically reduced when hitting the 25 km/h limit, only very little support remaining up to 26 km/h. I remember that from legislation perspective this is ok until +10% (27.5 km/h). My buddys Fantic XEF 1.9 with Brose Mag S is a little faster then my E10, pushing the bike up to 27 km/h. Always forcing me to use the boost button

what is Fantic, Brose like compared to Dengfu, m510? climbs, forest roads, singletrack ......?

 

[SchadowMaster]

Today the bike, screen and motor suddenly turned off more than ten times. One time the bike didn't run at all for ten minutes Although all cables are connected. In the end everything worked again normally.
I couldn't identify the problem. Have you had similar problem?

 

[thaeber]

what is Fantic, Brose like compared to Dengfu, m510? climbs, forest roads, singletrack ......?

At least same power, I allways outperform my buddies at steep climbs. Only the Brose / Fantic is on same level. Overall consumption very good. But hard to judge, due to personal fitness differences.

 

[ficorama]

At least same power, I allways outperform my buddies at steep climbs. Only the Brose / Fantic is on same level. Overall consumption very good. But hard to judge, due to personal fitness differences.

ok, means they are roughly in performance. and what was the difference with the m500?

 

[savas]

Today the bike, screen and motor suddenly turned off more than ten times. One time the bike didn't run at all for ten minutes Although all cables are connected. In the end everything worked again normally.
I couldn't identify the problem. Have you had similar problem?

Might be cables,especially where you have the cable coming from the motor connecting to the display or where the cable connects to the motor.These connectors are really bad.Not sure why bafang uses them

 

[thaeber]

ok, means they are roughly in performance. and what was the difference with the m500?

Much more power and torque. 0.5 kg lower weight. Better battery consumption. Little more noisy (whining).
With my M500 and 13A firmware the M500 was good, but it was hard to follow the Fantic/Brose on long and steep climbs. With the M510 it‘s easy, and I rarely need my biggest climbing gear.

 

[Rob Rides EMTB]

Much more power and torque. 0.5 kg lower weight. Better battery consumption. Little more noisy (whining).
With my M500 and 13A firmware the M500 was good, but it was hard to follow the Fantic/Brose on long and steep climbs. With the M510 it‘s easy, and I rarely need my biggest climbing gear.

Feels a tons stronger now compared to the M500. First ride today in a car park!

 

[ficorama]

new bike finished? please after testing the bike, a brief comparison of the engine with, bosch, brose, yamaha, shimano. thank you

 

[Rob Rides EMTB]

new bike finished? please after testing the bike, a brief comparison of the engine with, bosch, brose, yamaha, shimano. thank you

CHEEB V2 - £2999 Electric Mountain Bike Build

Here's the finished build, under £3k for an ebike in this day and age is pretty decent! Had a quick spin in the car park and the M510 motor is way more punchy and responsive than the older one - and lighter! Bike weight is exactly 24KG (with DH tyres!). Not too bad! Full specs:

www.emtbforums.com

 

[krokan]

Much more power and torque. 0.5 kg lower weight. Better battery consumption. Little more noisy (whining).
With my M500 and 13A firmware the M500 was good, but it was hard to follow the Fantic/Brose on long and steep climbs. With the M510 it‘s easy, and I rarely need my biggest climbing gear.

So what is the M510 Watt peak?

Do i understand it right that the max W peak is the main factor? -more W->more power?

 

[Kyokushin]

Do i understand it right that the max W peak is the main factor? -more W->more power?

Yes. The 250W is nominal power (the power where engine have temperature equilibrium). More peak power (that value is not written on bike or engine) = more Wats and stronger engine.