Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[yhafting]

I have been looking for a way to implement traction control on my Zero 2015 SR, and i believe i have found two possible ways of implementing it.

I had a look at the DVT customer manual from sevcon- which goes through some of the settings  you may set on a sevcon controller.
One of these is the ability to set the maximum acceleration rate. From the image (pasted below) it seems like the default is 10.000 rpm/s.

If this was set to for example 1550 rpm/s, it would correspond to the approximate rate that would take the bike from 0-100km/h in 3 seconds- which is slightly faster than the bike will do anyway. (1549 rpm/s is based on 28 / 130 tooth gear ratio, and 17" wheel (1,66m circumference with 9,8cm tire height))

Has anyone tried to do anything similar / and or tested that setting? I know there are a few having manipulated their sevcon controllers for allowing regen brake levers and such, so all having done anything like that should be able to make a test.

At the moment the availability and cost of the equipment for talking to the sevcon seems slightly expensive for attempting to adjust one parameter alone.

Another option that strikes me is that the Zero app likely does program the sevcon controller when setting custom modes. Thus- depending on how the app sends data- it could in theory be used to change arbitrary settings on the sevcon controller. (It would have been nice to do things by the app, rather than by buying  a PC-to sevcon interface for 400$ and opening the bike for accessing the sevcon port).

 

[Burton]

I am not sure how this would add "traction control" which to me suggests you detect when the wheel is slipping and when it is gripping.

That said when I get a chance I will look at my DVT to see which the stock values are for these ... I can tell you the max RPM is set to 7000 though ... that I know on the IPM motors. But I think this is a cut off value not an actual max which is set to another value.

Also "local limits" is more of a summary of what is set ... there are a lot more settings when it comes to your motor which are not shown in this general window

[yhafting]

I am not sure how this would add "traction control" which to me suggests you detect when the wheel is slipping and when it is gripping.

That said when I get a chance I will look at my DVT to see which the stock values are for these ... I can tell you the max RPM is set to 7000 though ... that I know on the IPM motors. But I think this is a cut off value not an actual max which is set to another value.

Also "local limits" is more of a summary of what is set ... there are a lot more settings when it comes to your motor which are not shown in this general window

The maximum RPM of 7000 makes sense, because that corresponds to 150,6 km/h, as fast as the bike goes.

A poor mans traction control is something i would define as a setting that allows for some spin, but not necessarily going totally wild. If you set a maximum acceleration of RPM/s to about 1550 (or slightly higher) it should never interfere with normal driving (it would be bad for burnouts and wheelies). It would be interesting to see whether such a limit would make recovery more easy when the wheel loses traction, compared to spinning up to 150km/h in a flash which is the case today.

But i agree that traction control is probably not the best word for it. I should perhaps said acceleration control.

I'm looking forward to read about what you learn with your DVT. I guess there are other things that would have been nice to tweak, such as allowing for coasting in ECO-mode. (That would have made it much more edible for rainy days). 

[Burton]

I guess there are other things that would have been nice to tweak, such as allowing for coasting in ECO-mode. (That would have made it much more edible for rainy days).

That is a really easy "fix" when it comes to settings in the DVT software.
Haven't tried it though since I run "custom" which has coasting ... when I get a chance I will test it in Eco as well as I never use roll off regen.

[MrDude_1]

I am not sure how this would add "traction control" which to me suggests you detect when the wheel is slipping and when it is gripping.

The short answer to that... if the rate of acceleration of the rear wheel increases at a rate faster than it is possible for the wheel to propel the bike, the wheel is slipping... cut power, measure again.
its super simple...and is how a lot of traction control systems work.

[BrianTRice@gmail.com]

The Sevcon toolchain became proprietary/authenticated, but previously (and pre-IPM motors still work with this) an unlicensed version of DVT was available that could implement this.

Now, regarding the Sevcon programming, the toolchain is made for developing electric forklifts and light industrial trucks, and the maximum acceleration rate is definitely a key control for "poor man's traction control", and unless you want to do burnouts, a sensible value is a good idea, like 1550 RPM/s suggested above. IIRC I have set this on mine to a value like 2000 (but I need to recheck what DCF I've applied).

For what it's worth, setting roll-off regen to 0 does have a safety concern that if you're an extremely low traction surface like wet grass or ice, the wheel can enter a state where the controller keeps turning it without throttle input. I've had it set to a negligible value for a while just to hopefully prevent this.

Why it doesn't come set that way from the factory with an override via the MBB (in say Sport mode) is anyone's guess. But the MBB code of course only implements what Zero thinks to implement, and there aren't any parameters (I'm aware of) for this.

Another related "poor man's traction control" setting is a ramp rate for regen, so it doesn't come on in a step function. I want to engage this because I use 0%-100% rolloff vs braking regen, and this occasionally surprises my girlfriend and her helmet gently bumps into mine. But also it would allow exceeding the limit set by Zero while also avoiding the concern of wheel lockup when traction is extremely limited (on gravel, in the rain, say).

There is also a dedicated traction control mechanism, BUT it is not usable for highway capable vehicles, warns the manual. I gather that the controller's CPU is just not fast enough to modulate the sine/cosine waves correctly for these cases.

[Burton]

For what it's worth, setting roll-off regen to 0 does have a safety concern that if you're an extremely low traction surface like wet grass or ice, the wheel can enter a state where the controller keeps turning it without throttle input. I've had it set to a negligible value for a while just to hopefully prevent this.

I think the stock MY17 is set to 0.01 if I recall (would have to check)

Why it doesn't come set that way from the factory with an override via the MBB (in say Sport mode) is anyone's guess. But the MBB code of course only implements what Zero thinks to implement, and there aren't any parameters (I'm aware of) for this.

It could be because they are using only one node on the controller and are limited to what they can control "on the fly?" Not sure but if I recall the RPDO's are pretty much all taken up stock.

[BrianTRice@gmail.com]

Why it doesn't come set that way from the factory with an override via the MBB (in say Sport mode) is anyone's guess. But the MBB code of course only implements what Zero thinks to implement, and there aren't any parameters (I'm aware of) for this.

It could be because they are using only one node on the controller and are limited to what they can control "on the fly?" Not sure but if I recall the RPDO's are pretty much all taken up stock.

That makes more sense; the operational model is much simpler than just "a CPU strictly controls everything" which is appropriate given the risks of running a switching power electronics controller.

And I do remember/realize that the feedback loop would just modulate / override the throttle input, which invalidates what I said about the CPU, anyway. So, it sounds like they just don't have a free node to process the loop given all the functions they're already exercising.

And Zero and Sevcon have probably already discussed this and not gotten anywhere, based on the observation that the only revision since 2013 for these controllers is the "Gen4.5" modification to use more efficient MOSFETs so that the newer controllers produce less waste heat. Sevcon makes what seems to be a reasonable product but never changes the feature-set and locks down access to get a little tooling licensing revenue...

[Keith]

I'm glad to see this discussion and I hope it goes somewhere. I noticed the Sevcon acceleration rate control parameter too, and had the same idea. While I've since learned different throttle techniques for off road riding that help, I'd still really like to limit the rate of rpm rise to a much lower value. When the rear wheel is off the ground and throttle is applied, the torque control mode leads to insane wheel speeds very quickly. Especially at low or even no speed, like when the front wheel is stuck on a root or rock, the drive wheel breaks loose and spins up to a speed that makes getting traction impossible. And small jumps that get the rear wheel up while accelerating lead to roosting without any benefit where lower rpm would hook up better. I've even considered adding a device that would detect excessive acceleration and trigger a regen pulse to dampen it. I realize my off road needs do not match those of most riders but the same problem does occur on slick pavement. Limiting acceleration is comparable to adding flywheel weight, a common modification for some off road bikes. For fun, try twisting the throttle quickly with the bike on a stand so the rear wheel is free. You can get 90mph on an FX in less than one second. On second thought don't try it, just take my word for it!

[yhafting]

I guess there are other things that would have been nice to tweak, such as allowing for coasting in ECO-mode. (That would have made it much more edible for rainy days).

That is a really easy "fix" when it comes to settings in the DVT software.
Haven't tried it though since I run "custom" which has coasting ... when I get a chance I will test it in Eco as well as I never use roll off regen.

That was what i hoped for. Now all i have to do is find someone in my vicinity that has this tool. Unfortunately my dealer/workshop does consider this "rocket science"- my heart really sunk when i read that , as i had high hopes for them.

If only one could use an arduino or such to access the CAN bus- but i guess it makes sense from a safety point of view to prevent that from happening in general.

Do you know if it is possible to set different max rpm/s for each mode?
For example i could set a much lower limit for rpm/s for eco and custom mode compared to sport. For eco i guess the (40% or so) torque could match a maximum acceleration of  1550*0,4 =  620 rpm/s or so.

(I use 66% torque custom mode for commuting in nice weather, but that has proven to be hazardous on wet asphalt).

[yhafting]

Another related "poor man's traction control" setting is a ramp rate for regen, so it doesn't come on in a step function. I want to engage this because I use 0%-100% rolloff vs braking regen, and this occasionally surprises my girlfriend and her helmet gently bumps into mine. But also it would allow exceeding the limit set by Zero while also avoiding the concern of wheel lockup when traction is extremely limited (on gravel, in the rain, say).

Thanks for a lot of useful information! I am pondering on whether i should invest in this tool, although it is expensive. If i knew someone close having it, that would be preferable though...

And i do know that gentle "bump" too  Id love to have a smooth transition for brake regen.

[rgutt]

I realize this is a really old post, but simply put, that is not nor does it even barely emulate what would be a functional traction control. If you set the max RPM/s high enough never to interfere with max throttle on dry pavement, a tail wind, and possibly even some negative elevation change, the wheel will easily spin out of control on a loose or wet surface. The maximum delta RPM limits can vary wildly depending on surface conditions and tire.

Someone mentioned the roll-off regen set to 0 versus the OEM setting of 0.1 being a safety concern. First, the system doesn't even enter roll-off until the controller has determined the motor has stopped, so this is only even a consideration after you've come to a complete stop. I can't imagine the situation shy of poor programming, poor tuning, or a defective controller, in which the wheel will try to turn with the output torque set to 0, but maybe this is the case and Sevcon (BorgWarner now) communicated this to Zero, hence the setting. That said, I've had mine set to 0 for almost 2 years now. There's no good reason for it really. It has 0 impact on efficiency since as I said before, it doesn't take affect until you've come to a complete stop, but it does make me feel better

[DerKrawallkeks]

I think limiting the acceleration in rpm/s still has an advantage in the fact that it limits the wheel speed that will be achieved when losing traction, until the rider cuts the throttle.
I agree that in the wet it won't act as traction control since the acceleration is still more than twice as much as might be possible in those bad conditions.

[Specter]

But you'd need some sort of input as to how much power being asked for, being put in, rpm .vs. theoretical should be rpm etc.  We kind of do this in the power plants to keep an eye on mvar  creep, if something gets too out of whack, you get a Power Level Imbalance Runback.  It's something like, You are giving it enough energy / torgue to generate 500 megawatts, yet you are only putting out 440 mw...something is wrong, back it down NOW.  In your case it'd be, you should only be doing xx rpm, or yy ramp rate yet are hitting zzz. excess..back it down NOW.  The math can get a bit complicated, and very dynamic quickly too !!  it'd have to also  honestly be millisecond accurate to be of any real use too.  How good is a traction control that kicks in only after the wheel been spinning out of control for 2 or 3 seconds.  When you are already doing 60 mph, by that time, you are already lowsiding into the ditch.  Worse case is this will probably happen 90 percent of the time during an acceleration, which again, you  need to act FAST to be of any use.  Remember, some of these bikes can go from 0 to 60 in a few seconds, and driving like THAT is when you are going to need this the most,  so yah, it has to be fast.

In an anecdotal way, this reminds me of when I ran over a spark plug in the highway and took out a tire.  I was pulled over and on the phone with the tow truck driver, when Fkn worthless ONstar interrupts my phone call to tell me I have a flat tire.  Oh wow REALLY!!!  that 'data' is not very useful at that point.

Aaron

[yhafting]

... The math can get a bit complicated, and very dynamic quickly too !!  it'd have to also  honestly be millisecond accurate to be of any real use too. ...
Aaron

No, the math here is dead simple. You set a parameter, and then that parameter is used whenever any other control input suggests other use. It does not have to be reactive if you limit acceleration or maximum speed.

Now a proper traction control would be better, and making something that is a lot faster than human and cuts power until traction can be regained is not that hard. Doing it so it gives a good feel of control is a different matter.

Personally i went and bought a bike with proper traction control two years ago, so I found a solution for me, but it would of course be interesting if someone had the chance to try out these settings, but I will not encourage anyone to go out and try breaking on asphalt to see if it works. If someone has the equipment setting the max acceleration low and then trying on gravel or some place you would expect to spil (snow, mud etc) to feel if it makes the difference
 or what it takes to make it work reasonably.