Programming the Bafang Middrive BBS01+BBS02

[bgt]

antihistamines and paracetamol are wunderful things wot!?? Got grumpy and had to go for a ride.

Grumpy or not, thanx for the info

 

[Ken Taylor]

I've recently been doing hours of testing on a Kurt Kinetic rolling road and plotting the motor power, speed and cadence data on Strava while trying to optimise the controller settings. With this set up I can rotate the pedals by hand so as not to add any human power and vary the motor load by changing gears. I can't see any explanations here on what particular settings do that fully match what I see the motor doing. Settings explained wrongly in this thread include "Keep Current(%)" and Current Decay (1-8)". I have established they interact and can't give a good explanation of their behaviour though I can see the explanations here are not correct.

A discovery from which a lot of other things follow is that my motor controller never uses the speed sensor but does all its speed sensing from the cadence. Is this the same for everyone? On the basic settings page I use the "Speed Meter Model: External, Wheel Meter" as does everyone else but it doesn't use the speed sensor to influence controller behaviour. It behaves as I would expect it to if I selected "Internal, Motor Meter". The speed shows on the C961 meter so the sensor is working OK. Looking at some data:

You can see here that the cadence remains the same regardless of gear setting. The controller adjusts power to match the changing load as the gears are changed so that it maintains the same cadence. Limitspd% settings of 40% and 50% are tested here with power settings of 100% and maximum speed set to 40 km/h. Just 18 Watts are needed, even though the power setting is 100%, to keep the cadence at 34 rpm with a 40% speed setting and the controller has no trouble controlling such a small load. This equates to 0.85 rpm per speed setting% which also works for 50% but it would imply 100% is 85 rpm. Max rpm is higher than 85 rpm and depends on battery charge/voltage. Also, changing the wheel diameter setting doesn't change cadence.

This contrasts with:

PAS cuts power as a function of road speed but dials down assist (limits current) as a function of crank speed. With the standard programming, the faster you pedal, the less assist you get. A feature I personally hate and have removed via programming. Lets hope my drive doesn't blow up now :lol:

Road speed is irrelevant on the tested motor, so is it different to Kepler's motor or does Kepler's ignore speed as well?

Hi Guys. Ok, this is what I have recently been telling people:

Great to see the thinking behind the settings on Em3EV supplied motors but I think it is wrong in parts which leads me the the view the settings are sub optimal for most people.

First:

Did i mention the controller seems to have little or no phase current limiting, well it hasn't and the fets are cheap low level things that are easily blown. At least now, they have fitted 9 low quality fets to their 25A controller instead of 6 low quality fets, like the previous version 25A controller.

and this corresponds to other posts, particularly where they mention controller failure or hot motors. Without phase current limiting there is no way to guarantee a motor will not overheat. You rely on the operator keeping the motor at a reasonably high rpm or minimising voltage applied to the motor at low motor speeds. The best a seller can do to avoid overheating is to keep the Limitspd% low so that little current is applied at low cadences but for an operator who avoids low cadence for more than a few seconds this is bad because the motor will provide no power at normal cycling cadences. An operator that keeps pedaling cadence in normal ranges, above about 60 rpm, with the BBS01 will never overheat the motor but an operator that lets motor speed drop below this risks overheating the motor with a wide range of controller settings. This would go away with phase current limiting but the BS0X doesn't have that feature. The seller suffers potential warranty claims if the Limitspd% is set high but the operator suffers poor performance if the Limitspd% is set low. The current to the motor is also affected by "Keep Current(%)" and Current Decay (1-8)". Even with the settings applied by Paul I suspect you can overheat the motor by going up a steep hill at low enough cadence. It is what I suspect happened here:

Just recently my second controller went belly up while riding along the Wasatch Mountain range... I used only the first two preset PAS levels(out of the nine PAS level setting) PAS1 and PAS 2 (current 52% and 56% respectively) with no throttle in the ride... The motor enclosure was noticeably hot.

particularly as the controller failed twice.

... Forget about assist level, IMO, that is not how it should be viewed. The PAS level should be used to set the pedal cadence.

I disagree. Assist level equates to power. Steady state power is controlled by the combination of Limit current(%), Keep Current(%) and Current Decay (1-8). Power is also a function of cadence with the relationship depending on the value of Keep Current(%) and is modified further when the cadence is varying. Limitspd% will only limit the maximum cadence at which the motor will supply the specified power. It will reduce the specified power to keep the cadence at the Limitspd% level. This is what you see in the graphs above.

IMO, the BBS02 works best when the speed and current progressively increase from assist 1 through to 9 (as is done in the standard settings) and you use the PAS level to set your pedaling cadence and forget about it as assist or power level. Do not view it as an assist, it's not. If setting 5 is the right 1 for you, use setting 5, if you want to do more work, pedal faster and you will take the load away from the motor.

I don't agree. I use the gears on the bike to keep nearly the same cadence on all but the steepest hills which makes any assist levels set with too low a Limitspd% useless and I increase the assist level when I can't keep up with the guy in front and reduce assist when I can.

The SW changes can only make the best use of the hardware, when combined with using the kit in an appropriate way. You can't change the hardware limitations of the controller by changing a couple of settings, but you can use the software to tweak things to disguise it's shorfalls and hopefully avoid it operating in ways that are likely to blow a fet due to overcurrent.

Yes that is true but the operator keeping the cadence above 60 rpm on the BBS01 will also avoid overheating due to over current without requiring controller settings that impede performance.

The BBS02 controller are very easily damaged with a single bad setting, so for all these guys changing settings, they should be aware they are easily damaged if incorrectly configured and it’s not fair to expect a warranty replacement on controllers, after changes have been made, no matter if those settings seem quite reasonable or not.

Yes, I can see that.

Has anyone found that the speed sensor is used by the controller to vary power output on their motor?

 

[astmacca]

Has anyone found that the speed sensor is used by the controller to vary power output on their motor?

The limiting speed (in km/hr) set from the display chops motor power when reached. It doesn't "vary" the power, just cuts it.

Given the units, that information HAS to be coming from the speed sensor at the wheel rather than the motor controller since actual road speed is independent of cadence because of the gearing.

However, in a hub motor, for a given wheel size (which is a variable you can enter in this program), a sensor that measured the motor speed WOULD give you road speed wouldn't it??

On page 22 pjgold opined that the Keep Current setting may be a hangover from hub motor operations (or at least thats what i think he means).

I believe this setting is a carry over from hub motor settings as it makes perfect sense when you have no gears.

Which has got me wondering what else in the GUI available to vary settings is only hub motor related as well? Manufacturer may have just recycled a "standard" program they use to set up and flash their firmware on a number of products??

Road speed is irrelevant on the tested motor, so is it different to Kepler's motor or does Kepler's ignore speed as well?

Well, my first session of testing round and round the park today had me convinced the motor was controlled by road-speed. Was in a mid range gear where the speed limit % on the PAS setting happened to me very close to what i thought it should in km/hr as a % of the display set speed limit. Thought to be consistent i should do all the testing in the same gear so ration wasn't a variable.



Not until i purposely tested in an "extreme" config (set 50km/hr on the display and was in lowest possible gear) that it became glaringly obvious i was wrong and that the road speeds / PAS level were nothing like what they should have been in km/hr as a % of the display set speed limit and that the motor was not even trying to get there, that the penny dropped.

Ken, you say you had the power set to 100% and speed to 40km/hr but on your plot it only gets to a max speed of 21.6km/hr?? How come? Is that all the BBS01 will do on your trainer set up or is there something else going on there??

Does anyone have screenshots of the setting used by other sellers that "custom" program these units?? Would be interesting as comparisons.

 

[leelorr]

Just something that might be interesting to try, but you could move the PAS pickup and magnets or whatever (I don't actually have a BBS0X yet) to the wheel so that the inputs WERE actually coming from road speed. You could install more magnets on the wheel to compensate for the longer distance traveled by each revolution of the wheel than by the crank. This way the gear that you were in would not have an effect on the distance traveled like it affects the cadence.

(possible noob post) :^)

 

[kjetilhk]

An operator that keeps pedaling cadence in normal ranges, above about 60 rpm, with the BBS01 will never overheat the motor

Even if assist is set to 100% of "current limit"(15), which allows a 36V250W engine to provide 540w? In a long steep hill where I can not get above the road speed limit? And "Keep Current 100%"...

If not: how hard can I push a 250w engine in long steep hills, when I always pedal above 60 rpm? And "Keep Current 100%"

 

[bgt]

Ken, what you are really saying here is, having a low cadence with a high gear can be crucial to the live of the motor. Especially in my country where you can have strong headwinds keeping your speed down.
This sounds like a bad motor controller design(=no phase current protection)...................................people who don't know about this design error and how to deal with it can easily destroy their motor............... :x

 

[astmacca]

no phase current protection

Can someone explain to electrical terminology challenged me, what "phase current protection" is and its significance in this context??

It makes sense to me that if you are working a motor hard (uphill, headwinds) and it is forced to operate below its optimum revs (which all motors have i think?) then its going to struggle suck power and heat up. The positive about a mid drive over a hub as far as i can see is that you have gears to get it closer to that optimum and should use them where as with a hub you just wear that limitation.

Just something that might be interesting to try, but you could move the PAS pickup and magnets or whatever

Nope. That's built into the motor on these.

 

[Ken Taylor]

Has anyone found that the speed sensor is used by the controller to vary power output on their motor?

The limiting speed (in km/hr) set from the display chops motor power when reached. It doesn't "vary" the power, just cuts it.

From the test data above you can see that in this case the controlled parameter is the cadence not the speed. In this test case the limiting speed of 40 km/h was set in the controller rather than from the display and Limitspd% was tested at 40% and 50%. The controlled parameter should be speed rather than cadence, is it for anyone else?

I don't think the power is just "cut" which is often referred to as bang bang control http://en.wikipedia.org/wiki/Bang%E2%80%93bang_control as the power data does not show the on/off pulsing which you would expect in bang bang control. However, there is 24 cadence sensor pulses per pedal revolution so at a cadence of 33 rpm in the test data graphed above there is an opportunity with bang bang control to turn the power on and off at a rate of up to 13 Hz. The sampling interval in the test data is 1 Hz and power is measured by sampling at 1 KHz and averaging every 1000 samples. The pulsing you would expect in bang bang control could be smoothed in the test data by the sampling rate being much lower than the switching speed but it gets academic at this point because bang bang control at high frequency and low hysterisis can behave the same as proportional control.

From the graph in the post above it looks like there is a control loop adjusting the power to achieve the set point cadence. At the start of the graph where the motor is accelerating from stationary and just after the orange line where it is accelerating from stationary again you can see the power is higher than that necessary to maintain the set point cadence. You wouldn't see this dynamic behaviour without a good test bed but the Kurt Kinetic rolling road has a lot of inertia which makes load during acceleration similar to what you would get on the road.

Given the units, that information HAS to be coming from the speed sensor at the wheel rather than the motor controller since actual road speed is independent of cadence because of the gearing.

Yes but the test data shows it is not which is why I've asked here if it does for anyone else.

However, in a hub motor, for a given wheel size (which is a variable you can enter in this program), a sensor that measured the motor speed WOULD give you road speed wouldn't it??

Yes.

On page 22 pjgold opined that the Keep Current setting may be a hangover from hub motor operations (or at least thats what i think he means).

I believe this setting is a carry over from hub motor settings as it makes perfect sense when you have no gears.

Pjgold's explanation for the "Keep current(%)" parameter is not consistent with my test data. However, pjgold also says "Current set to 100% the assistance level does not change at all from what you select." which is close to, but not exactly what I see. I speculate that it is a PID controller http://en.wikipedia.org/wiki/PID_controller and that Keep current(%) is the proportional gain parameter. With a PID control algorithm, the higher you set the proportional gain the closer the controlled parameter, in this case current, tracks the set point value. If the proportional gain is set too high the controlled parameter becomes unstable and oscillates which I haven't seen in my testing. Therefore, if you are aiming to keep power constant as specified by the level setting 1-9 regardless of cadence, which is what I want, then setting Keep current(%) to 100 works best. Pjgold and Kepler both recommend a "Keep current(%)" setting of 100% and Paul from Em3ev recommends against 100%. I suspect that once "Keep current(%)" is at 100 then "Current Decay (1-8)" has no effect but I'm not sure of this or exactly what "Current Decay (1-8)" does. I suspect "Current Decay (1-8)" is the integral parameter in the PID algorithm and the proportional gain is high enough at 100% that adding an integral component does not produce an obvious effect but this is pretty speculative. "Current Decay (1-8)" definitely modifies the effect of "Keep current(%)" at values other than 100. Following the suggestion that it is a PID controller we are adjusting to its conclusion, I suspect "Slow-Start Mode(1-8) is the derivative parameter in the PID algorithm but this is untested.

Manufacturer may have just recycled a "standard" program they use to set up and flash their firmware on a number of products??

That would be the way I would do it.

Well, my first session of testing round and round the park today had me convinced the motor was controlled by road-speed. Was in a mid range gear where the speed limit % on the PAS setting happened to me very close to what i thought it should in km/hr as a % of the display set speed limit. Thought to be consistent i should do all the testing in the same gear so ration wasn't a variable.



Not until i purposely tested in an "extreme" config (set 50km/hr on the display and was in lowest possible gear) that it became glaringly obvious i was wrong and that the road speeds / PAS level were nothing like what they should have been in km/hr as a % of the display set speed limit and that the motor was not even trying to get there, that the penny dropped.

At high speed in low gear you might be exceeding the maximum rpm of the motor. Better to test at low speeds, perhaps 20km/h. Try your highest gear and lowest gear on flat ground and see if you always get the same speed. You need to pedal but try not to add any human power.

Ken, you say you had the power set to 100% and speed to 40km/hr but on your plot it only gets to a max speed of 21.6km/hr?? How come? Is that all the BBS01 will do on your trainer set up or is there something else going on there??

The controlled parameter is cadence. The gearing then defines the speed. See that the speed changes as the gear is changed but the cadence remains the same.

An operator that keeps pedaling cadence in normal ranges, above about 60 rpm, with the BBS01 will never overheat the motor

Even if assist is set to 100% of "current limit"(15), which allows a 36V250W engine to provide 540w? In a long steep hill where I can not get above the road speed limit? And "Keep Current 100%"...

If not: how hard can I push a 250w engine in long steep hills, when I always pedal above 60 rpm? And "Keep Current 100%"

I have tested the motor at full power up long steep hills, specifically Black Mountain http://en.wikipedia.org/wiki/Black_Mountain_(Australian_Capital_Territory) achieving a minimum 65 rpm at the steepest bits with me pedaling as well. I was testing the gear ratios and trying to determine the largest cog I needed. The motor does not get hot and I'm pretty sure can run forever at full power as long as the cadence is kept up. The battery goes flat fast when you suck that much power out of it though. A 350W motor is limited to 18 amps but the BBS01 controller info box in the programming interface says "Max. Current: 20A". I haven't tried increasing it to 20A because I don't like to risk damage but it may be OK at 20 amps too. I usually run with the maximum set at 15 amps to match the spec of the 250 Watt motor and 15 amps already provides more power than I ever need.

Ken, what you are really saying here is, having a low cadence with a high gear can be crucial to the live of the motor. Especially in my country where you can have strong headwinds keeping your speed down.
This sounds like a bad motor controller design(=no phase current protection)...................................people who don't know about this design error and how to deal with it can easily destroy their motor............... :x

Yes that is right. You need a reasonably high cadence or phase current limiting to avoid overheating but most electric bikes do not have phase current limiting. What saves most people from disaster is that it feels unnatural to pedal at low cadences so they change to a lower gear. There is a more in depth discussion of this at http://endless-sphere.com/forums/viewtopic.php?f=2&t=55283&start=25#p913803.

no phase current protection

Can someone explain to electrical terminology challenged me, what "phase current protection" is and its significance in this context??

It makes sense to me that if you are working a motor hard (uphill, headwinds) and it is forced to operate below its optimum revs (which all motors have i think?) then its going to struggle suck power and heat up. The positive about a mid drive over a hub as far as i can see is that you have gears to get it closer to that optimum and should use them where as with a hub you just wear that limitation.

You should look elsewhere for a good explanation of phase current limiting. A simple explanation is that limiting current on the battery side of the controller does not in general limit current on the motor side of the controller. Current on the motor side is determined by (applied voltage less back EMF) divided by electrical resistance. Electrical resistance is constant and back EMF increases with motor speed. If the motor speed is too low back EMF is low and current can get too high. A motor has a maximum torque it can produce and current beyond that required to generate maximum torque generates only heat and does nothing useful.

I'm still looking for responses as to whether anyone has found their BBS0X behaves as I think it should and controls speed. Or do all the others behave as mine does and control cadence?

 

[pjgold]

There are serval places in the software that you can set a speed limit and they behave differently.
Setting a limit at the display unit cuts power at road speed set.
Setting speed % on the basic tab is motor rpm speed
Setting speed on the PAS tab is a road speed
And setting speed on throttle tab is road speed

I assume which ever cuts in first overrides the others

 

[dirkdiggler]

I still haven't fixed my speedometer, so I don't think the Bafang relies at all on the speed sensor. I believe it is a function of cadence.

Also, here are my settings before any changes. Kit was purchased from GE bikes (aliexpress 4/2014). 750 watt 48v.

These settings were changed - sorry no before screenshot. Limit speed % was changed on 0, 7, 8, and 9.

Thanks again to Darth elevator for helping me change the settings. Sorry for the size - a pain to resize them!

How do you remove the throttle delay again? We couldn't remember and it still has a bit of lag, better but still present. One other thing I might recommend is lowering the LVC. I seem to find my batteries are usually at 3.6v or higher when it shuts down on 12s. Good for the batteries, but I could use just a bit more juice. I had a positive lead pull out of my Lipo pack on one 4s battery, so I'm down to only 5ah until I can get another battery. I'm having to cut back on the throttle and pedal more to get me through my ride.

 

[Ken Taylor]

I still haven't fixed my speedometer, so I don't think the Bafang relies at all on the speed sensor. I believe it is a function of cadence.

Yes.

There are serval places in the software that you can set a speed limit and they behave differently.

Ah, that is useful advice. It never occurred to me speed might mean different things in different places.

Setting a limit at the display unit cuts power at road speed set.
Setting speed % on the basic tab is motor rpm speed
Setting speed on the PAS tab is a road speed
And setting speed on throttle tab is road speed

I assume which ever cuts in first overrides the others

I only tested Setting speed % on the basic tab and assumed the others behaved the same. That makes sense thanks.

 

[pjgold]

From what Paul has it appears that the basic tab speed % setting causes the controller to lower current gently to maintain motor rpm. The other settings appear to just be a power cut based on road speed.

Also remember BBS01 is 80 rpm max and the BBS02 is 120 rpm max.

 

[astmacca]

pjgold wrote:
Setting a limit at the display unit cuts power at road speed set.
Setting speed % on the basic tab is motor rpm speed
Setting speed on the PAS tab is a road speed
And setting speed on throttle tab is road speed

I assume which ever cuts in first overrides the others

Makes sense.

These two certainly behave that way in my experience.

dirkdiggler, can you link to those settings again. Cant see them.

 

[Kepler]

Some good testing carried out here and I think we are all a little wiser on how the controller operates. Its good to know that % speed is cadence speed and not wheel speed however in my opinion nothing has really changed in relation to how to configure this controller. whether it is wheel speed or cadence, I see no advantage in limiting the speed and consider any limitations here counter productive. The standard control strategy makes no sense to me. It tends to force you into into a harder gear and to spin the cranks slower if you want to maintain your assist. What is the point in that?

What the control strategy should be doing is encouraging you to spin the the cranks faster to maintain assist. In my opinion if you want good performance, economy, and longevity out of the drive, keep the the current limit down and the cadence up. This has served me well over 5000km of hard weekday commuting and weekend single track fun.

Speculation is now that the drive doesn't have effective phase current limiting. My experience is that with poor phase current control or no phase current control you end up with savage power ramp control. Take for example an RC controller with no phase current limiting or a standard square wave controller with the battery to phase current ratio set too tight. I personally find that the BBS controller has a really smooth and gentle ramp capability suggesting it either has really good battery current limiting or the phase current limiting isn't so bad after all.

Perhaps I have just been lucky not to have had a controller failure so far. To be honest, even if I did suffer a controller failure at this point I would quite happily replace the controller and program it with exactly the same settings again.

 

[bgt]

I see no advantage in limiting the speed and consider any limitations here counter productive.

Thanx, good to know. I had changed them back again to default on my 250W BBS-01. I am also convinced it has a bit of phase protection wisdom build into the micro controller. I have run a lot on full throttle control lately and the motor becomes a bit warmer when I do a "throttle only" ride, so no pedalling=no cadence. Difficult subject if you have no insight in controller setup? I wonder if the ready made bikes with a Bosch/Yamaha/MPF midmotor(which are sold by an enormous amount here) do have phase protection because they do torque assist i.o. PAS which means a heavy load at first start and I see people choosing a high gear so they get more assist. This is in sheer contrast with the Bafang controller stories.
PS does any here has the specs of the powerfets used in the controller? Maybe they can absorb a lot of waste energy......have some kind of limiters?

 

[astmacca]

keep the the current limit down and the cadence up.

Which HAS to keep the load off the motor.

Knowing that speed% on the Basic page is actually a reference to cadence and not road-speed is important i think. Means to me that by using the right gear you can actually get to, or close to, what Paul from EM3ev referred to as "motor no-load speed"

It seems to me like the controller when working in PAS mode, just does not seem to be able to function well at a speed significantly below the motor no-load speed. This motor no-load speed is adjusted by changing the PAS speed setting associated with that particular "Assist" level.

This has served me well over 5000km of hard weekday commuting and weekend single track fun.

Lol! I really have to take whats said about these beasts by someone who has run up that many km seriously. Particularly since i have only about 200km so far. Fun km though.

While also acknowledging that Paul has a lot of experience with these from maybe a different perspective, direct access to the manufacturer, and has been good about sharing info here.

To be honest, even if I did suffer a controller failure at this point I would quite happily replace the controller and program it with exactly the same settings again.

I suspect i would do the same. Although i have different settings from kepler (at the moment, i may go back to more like his) as a user i really wasn't happy with the stock settings.

Did see on another thread there was discussion about replacing the FETs with better quality ones.

http://endless-sphere.com/forums/viewtopic.php?f=2&t=62983&start=50

Anyone know if thats been done as yet?? That sounds like it may be a good long term solution to me. Not that i would hack up a working controller, but if one ever died on me i'd probably have a go at that, as well as ordering a replacement.

PS does any here has the specs of the powerfets used in the controller?

I think the specs or a link to such is on the thread i linked to above.

 

[pjgold]

There may be a bit of efficiency to be gained by setting the cadence speed limit to match your preferred cadence so the controller eases of the power at the right point rather than continuing to try and drive up to 120 rpm. But you sort of do this manually by setting the right amount of power assist level to support your cadence.

 

[astmacca]

There may be a bit of efficiency to be gained by setting the cadence speed limit to match your preferred cadence so the controller eases of the power at the right point rather than continuing to try and drive up to 120 rpm.

Worth trying i reckon, I have no interest in pedaling at 120rpm. Considering on this drive there a fully 9 levels of PAS to play with.

And if you use the gears properly, to get to that cadence reasonably quickly, there should be no problems with loading up the motor and generating heat.

Found this which may be useful in working out the speed% to set to get a particular cadence for a particular wheel size and gearing.

http://www.machars.net/bikecalc.htm

Ok, have had a play with a spreadsheet to plan next iteration of settings to try.

https://www.dropbox.com/s/6g3v743oonvq28i/Ebike%20PASCadence%20levels%201.xlsx?dl=0

Using the gears will mean i can hit the cut-off cadence for a PAS level either below my target road speed, almost bang on it, or if i was using top gear have assistance going all the time. Easy enough to change down to avoid the last and loading up the motor if needed.

I'm thinking this:

keep the the current limit down and the cadence up

may turn out to be the golden rule for these drives.

 

[Kepler]

I also think there is merit in setting the % speed that matches the max cadence you are happy with. Personally I take my cadence over 100 on a regular basis so the 100% speed on all assist settings suits me.


If you decide to use settings similar to what I use (and if you have purchased a drive off Paul, do so at the risk of losing your warranty), I think another good strategy is to only go to higher assist settings then say "3'' once you are moving at a reasonable speed and are in the correct gear for efficient cadence. I personally never go above PAS assist "4" which is around 500W. I do however use my throttle button on a regular basis for short sharp hills which pushes my power out to around the 800W mark.

 

[Rusty123]

If you are interested in maximizing battery range, while still getting benefit from electric assist, I think setting the speed cutout at (or slightly below) your natural cadence makes sense. That gives you assist when accelerating, or climbing hills, but cuts power when you're cruising along on your own power.

 

[Aushiker]

If you are interested in maximizing battery range, while still getting benefit from electric assist, I think setting the speed cutout at (or slightly below) your natural cadence makes sense. That gives you assist when accelerating, or climbing hills, but cuts power when you're cruising along on your own power.

How does one determine the relation between the cutout (%?) and cadence? Say if one's natural cadence is 80 or 90 what would the cutout value be?

Thanks
Andrew

 

[astmacca]

How does one determine the relation between the cutout (%?) and cadence? Say if one's natural cadence is 80 or 90 what would the cutout value be?

Depends what gear you are in. I had a go at this last night and linked a calculator (that does cadence vs roadspeed for each gear) and my planned settings a couple of posts up. I did them on a spreadsheet but it wouldn't let me attach a .xlsx file.

Did 22km on those this morning and its ok. Think though i will bump the speed% up a little. Having it cut out 10rpm below my actual target road-speed was not comfortable. I'm really still sorting what cadence i am comfortable with as have not really thought of this in these terms before.

 

[Ken Taylor]

Maximising battery range is important to me. The goal is to ride as everybody else does but not get left behind. I had the pleasure of riding with three pro riders today. I couldn't have kept up without assist, particularly on the long climb.

If you are interested in maximizing battery range, while still getting benefit from electric assist, I think setting the speed cutout at (or slightly below) your natural cadence makes sense. That gives you assist when accelerating, or climbing hills, but cuts power when you're cruising along on your own power.

You describe mine as supplied by Em3Ev before I started fiddling and I didn't like it. When you are on your own you don't notice it much but if you are riding with others and starting to fall off the back as the grade increases you need to change gears up where you would normally change down to go faster on an increasing grade. That reduces your pedaling efficiency. I tried the first time today with constant (i.e cadence independent) power levels and it is better but far from perfect. Previously I wasn't getting assist on descents as the cadence invariably increased beyond the assist range but now it gives help on descents which I don't need and is wasteful. I can dial back the assist level but I've put them close together, the steeper it is the more levels I've gone up and it is a pain to go down 8 levels on a descent then back up again a bit later. Maybe I'll get used to it or change the spacing.

... in my opinion nothing has really changed in relation to how to configure this controller... The standard control strategy makes no sense to me. It tends to force you into into a harder gear and to spin the cranks slower if you want to maintain your assist. What is the point in that?

Yes, I'd agree with that.

Speculation is now that the drive doesn't have effective phase current limiting. My experience is that with poor phase current control or no phase current control you end up with savage power ramp control.

Speculation is because

Did i mention the controller seems to have little or no phase current limiting, well it hasn't and the fets are cheap low level things that are easily blown. .

and I didn't check it but can you can tell from the power ramping whether it has phase current limiting? I thought you could do power ramping without phase current limiting.

How does one determine the relation between the cutout (%?) and cadence? Say if one's natural cadence is 80 or 90 what would the cutout value be?

Thanks
Andrew

As stated a few posts back http://endless-sphere.com/forums/viewtopic.php?f=28&t=58780&p=958593#p958358 from the accompanying graph I get 0.85 rpm per speed setting% for a BBS01 but even in that graph there is a slight load dependence. I would guess the BBS02 is 1.3 based on the maximum cadence being 130 rpm. The load dependence would be much higher at Keep Current(%) values less than 100 and Current Decay (1-8) changes things too at values of Keep Current(%) less than 100. I didn't test the dependence on Keep Current(%) during cadence control but I think it follows from the observation that current becomes a function of cadence at Keep Current(%) values less than 100.

Depends what gear you are in.

No it doesn't. Well, very little at the values I tested. Again see the graph at http://endless-sphere.com/forums/viewtopic.php?f=28&t=58780&p=958593#p958358 Notice that cadence is almost constant regardless of gear which goes through a range of 12 to 34 teeth and working against a load intended to be equivalent to the load at that speed on a flat road.

 

[pjgold]

The limited percent rpm is a direct percentage of the max rpm of the BBS unit which is...
120 rpm for a BBS02
80 rpm for a BBS01

So on a BBS02 75%=90 rpm and 67%=80 rpm
However a BBS01 100%=80 rpm and 75%=60 rpm

 

[astmacca]

No it doesn't. Well, very little at the values I tested.

I think it actually does in the context i was looking at, which is a speed% (cadence) setting that equates to an actual "target" road-speed. That's ratio dependent, and you can see that from your graph since the road speed drops as the gears change, while cadence stays pretty constant.

I may well be barking up the wrong tree with this approach in terms of comfort and usability but its fun trying out variations and learning. There seem to be a range of settings out there that at least dont break the controllers straight away.

And i'll just get it set up nicely and then my 42T chairing will arrive.