What cancels the turn signals?

[txmxrider]

Okay, I'll admit up front that I haven't done a scientific study on this yet, and I'm taking the lazy man's way out by posting the question here rather than trying things in the garage first. But I'm just curious, on a '99 R1100S, why is it that sometimes the turn signals self-cancel and sometimes they don't? At least that's what I think is happening. About 1 out of 20 times I use the turn signal, after I complete the turn I notice the blinker isn't on anymore even though I hadn't pressed the cancel button yet.

Well I did try a few things in the garage. It doesn't seem to be time related because I've deliberately ignored them to see how long they go and finally ended up using the thumb switch to douse it. Then I thought maybe there was a trigger in the handlebars that senses when you counter steer to straighten back up from a turn. But sitting in the garage I can turn the bars from lock to lock and the signal stays on the whole time.

It happens so rarely that I'm wondering if I mistakenly failed to turn the signal on or perhaps inadvertently hit the cancel button somehow.

[cageyar]

They shot off automatically after 10 seconds or 650 ft. if you don't turn them off.
Its in the owners manual.

[markjenn]

Been discussed before. It's a time/distance algorithm. For the signals to cancel BOTH a time constraint AND a distance constraint must be exceeded. The time constraint governs on the freeway when you're signaling lane changes and the distance constraint governs when you're sitting at a light. I've never seen anyone do an investigation to get what the precise constraint values are; would be an interesting project.

It's important to note that this feature is not designed to be the regular way of canceling as they typically run too long. It is designed as a safety net to keep them from running when you completely space out and forget to cancel them altogether.

- Mark

[signit98]

...that's all BS...

They are BMW turn signals... they are smart!

You switch them on and they watch where you are going... when they deem it safe, they switch themselves off... that's how easy that is!

Sometimes, when they are tired, they switch themselves off quicker than other times... sometimes, they don't like to work at all... they are German after all, Stubborn!

[txmxrider]

Well, it's NOT in the manual btw. Unless you got the deluxe manual and I didn't. Page 10 and 11 does say "automatic canceling while riding." So that and Markjenn's response would explain why they never cancel while sitting in the garage, if it takes both time and distance. But it clearly isn't as simple as 10 sec or 650 ft, whichever comes first. 650 feet is close to an 1/8th mile and 10 seconds is a long time when you figure the normal street corner shouldn't take more than 3 or 4 seconds. And the occasional normal street corner is where it seemed to self-cancel. So that still begs the question, why does it sometimes cancel and sometimes not?

Next time I'm out on a desolate stretch I'll just turn on the indicator and ride a straight line and see how long it goes. Then if/when I can find a corner where it actually does self-cancel, I'll go back and try the same turn about a dozen times and see if it's consistent or if it makes up the rules as it goes.

[Boybiker3]

Good question and here's my take on the stock timing for the R1100S: it is a bit smart.

There is a base time-out period which is fixed for standing still. But the length of flashing is modified in inverse relation to the voltage that drives the speedometer. Simple as that.

Of course, that is too simple. One of the first mods you should consider is a Kisan SignalMinder. All kinds of benefits, esp. to tired thumbs and for bikers who don't want their signals flashing long after the move is over.

[roger albert]

But the length of flashing is modified in proportion to the voltage that drives the speedometer. Simple as that.

Where did you get that idea? Wouldn't seem to be the case, as even ones with a dead signal to speedo still work fine as far as cancellation goes.

[Boybiker3]

Either David Robb has a cousin who wanted to unload some blue wire on BMW or the blue wire between the speedo and the turn signal controller has some useful purpose. I'm not positive which it is.

[chewie]

fwiw - on the 12s at least, I think they can tend to stay on a bit too long, especially if you think about when you are at a 'T' intersection and you stop and then take off to the right for example. If the T intersection is close to a bend in a 100km/h zone, it's not out of the question for someone coming from the left to steam around the bend and see a bike with it's indicator on (heading off in the same direction) assume it's turning and for the cage to duck up the inside when in fact the bike has just turned out onto the main road with the indicator still on. If someone then comes the other way around the same time - there's a bit of a space problem.

I have a bit of road just like this near my house. I consciously turn mine off because of that potential predicament.

So for me the answer to your question is ~ my right thumb cancels the turn signals

[roger albert]

> I'm not positive

This is normal bb.

Yes, there is a pulsetrain, but they cancel fine w/o it.
Maybe over a longer period. I've not timed both ways.
Just noted cancellation isn't cancelled w/o speedo signal.

[EenyBear]

On the 12S at brisk highway speeds, the time and distance factors (whichever is dominant) seem to cancel the signal at an appropriate time - for a relaxed lane change, for example. Around town, the time and distance are always more than required so I manually cancel.

Once my turn signal was stuck so I had to circle the block until I ran out of gas.
Ian

[Yellow12]

Originally Posted by chewie

So for me the answer to your question is ~ my right thumb cancels the turn signals

That's true Chewie.
And like Mark said, it's just a backstop in case you have a brain fade and forget to do it yourself.

PS I think the 12s manual says 10 seconds or 200m.

Ant

[sgoodwin]

This is news to me, mine have never self cancelled ever. When friends ride it's always fun watching them try to figure out how to cancel them. Wonder if I can retrofit the k1300's switch pods? Normal would be nice for a change as everything else I own is the same.

[markjenn]

We've been talking about this on this forum off/on for awhile.... I'm leaving on a trip in a week and I'll take a stopwatch and see if I can get some hard data. I think it should be pretty simply to test: just start a timer and distance measurement simultaneously right when you signal. Do one trial at very high speed and this should give you the time constraint. Do another at very low speeds and this should give the distance constraint.

Obviously, this is a very imprecise way to try and anticipate all of the possible signaling situations. That's why there should be no expectation that this system should cancel at the appropriate time - in fact, the time and distance constraints are set on the long side so that riders won't complain of premature canceling. It's a safety net to catch the gross errors - we've all been behind other riders who forget to cancel and traverse a county or two before they notice the blinking light on the dash. This is the scenario the system is designed to prevent - nothing more.

I think may riders come at this from a car driving mentality expecting a level of consistency in automatic canceling that is simply impossible on a bike. With cars, the steering wheel is a control that determines the rate of turn - for any given speed and assuming no skidding, X angular displacement of the wheel will always yield a consistent Y turn rate. This means that a turn in a car has the driver cranking in a steering wheel angle and holding it until the end of the turn when the crank the other way. This isn't so with bikes where steering deflections are used it initiate and recover from turns, but during the turn there may be no deflection of the steering. The steering changes the turning rate, but doesn't set the turning rate. (For you pilots out there, it's like the ailerons in an airplane.) And counter-steering makes the situation even more complicated.

- Mark

[txmxrider]

FWIW, I gave it a quick test yesterday as I was running some errands and made a few discoveries. Funny how I hadn't noticed something this basic before. For one thing, it's not as easy as you might think to keep one eye on the seconds-counter of your watch, one eye on the turn signal indicator, and still keep from hitting the Toyota in front of you. I eventually gave up on the wrist watch and started using the one-one thousand, two-one thousand method. All tests were done in city traffic never exceeding 35 mph so I still need to try the same thing at higher speeds just to compare. But from my initial test it appeared that distance had nothing to do with it, just forward motion of some sort, and time...19 seconds actually. That's why I need to try it at a faster pace tho, to see if distance does come into play. But it seemed that it always canceled after 19 seconds *unless* I stopped for a light or a stop sign. Then it appeared to stop counting for the duration that I was stopped, then started counting again as I started moving.

19 seconds is a surprisingly long time. I found that it's less annoying if I engage the blinker further from my actual turn, then it cancels at what seems like a more reasonable distance after the turn. Either way it seems too long to me and the thumb switch is still the best bet. Or maybe one of those Kisan Signalminders.

If it ever stops raining around here I'll try to ride some more and take more measurements.

Cheers all.

[AntonLargiader]

It's more like a minimum time/distance requirement, not whichever comes first. That way they don't shut off when you're stopped at a light, but they will not stay on for 30 seconds when you change lanes at 80mph. It's more like ten seconds at speed, I think.

The flasher unit reads the conditioned (by the cluster) wheel speed speed signal TAA, and simply makes the decision to cancel based on that and its internal timer. The cable-driven-speedo bikes don't have this feature. Just K-bikes, R11S, Rockster and R1200C I think, until the CAN-bus bikes.

[Boybiker3]

Originally Posted by AntonLargiader

It's more like a minimum time/distance requirement, not whichever comes first. That way they don't shut off when you're stopped at a light, but they will not stay on for 30 seconds when you change lanes at 80mph. It's more like ten seconds at speed, I think.

The flasher unit reads the conditioned (by the cluster) wheel speed speed signal TAA, and simply makes the decision to cancel based on that and its internal timer. The cable-driven-speedo bikes don't have this feature. Just K-bikes, R11S, Rockster and R1200C I think, until the CAN-bus bikes.

Right. Just as I explained in post #6 above. It is the blue wire (post #8).

Easy enough to check: use the centerstand and run at 0, 15, 30, and 60 mph.

Or cut the blue wire.

[markjenn]

Originally Posted by txmxrider

I found that it's less annoying if I engage the blinker further from my actual turn, then it cancels at what seems like a more reasonable distance after the turn.

Do what works for you, but again, the system is not designed to auto-cancel routinely - it's way too unpredictable and you'll be running long periods of time with the blinkers on. You should always cancel manually and then let the system catch the occasional time you forget.

- Mark

[txmxrider]

Originally Posted by markjenn

Do what works for you, but again, the system is not designed to auto-cancel routinely - it's way too unpredictable and you'll be running long periods of time with the blinkers on. You should always cancel manually and then let the system catch the occasional time you forget.

- Mark

Right on. I usually stick to the legal definition of 100 feet before the turn and then use the thumb switch to cancel immediately after the turn.

In my previous post I didn't make clear that I was referring to when I was testing. I found it a little unsettling to have the blinker still going for so long after the turn so I started initiating it a little further up the road than normal so it was more like 60/40. But that was just for testing, not my normal riding mode.

Thanks for the thought tho.

[markjenn]

Originally Posted by txmxrider

In my previous post I didn't make clear that I was referring to when I was testing. I found it a little unsettling to have the blinker still going for so long after the turn so I started initiating it a little further up the road than normal so it was more like 60/40. But that was just for testing, not my normal riding mode.

Gotcha. Didn't mean to come down on you. It's just that one of my pet peeves is folks disliking the BMW self-canceling feature because it doesn't cancel exactly when they want. That's not its design nor is it feasible to build an self-canceler for a bike that is the routine way of canceling - at least not without getting into a system with gyroscopes that can accurately track lean angle. And I think BMW is to be applauded for providing this feature as a "safety net".

- Mark