Two fails in one

[mcatrophy]

... Why then must this bike be so touchy about all this?

I'm also curious as to what causes the fault detection in the CC. ....

Don't overthink this. If the CC sees a brake or clutch activation it will disengage. The only "fault" is that a switch may operate when it shouldn't.One of the FJR's few unwanted "features" is that the rear brake lever pivot can get sticky if not lubricated, leading to the switch remaining on after braking. Then the CC won't engage (and the rear brake disc gets hot).

There are a few instances of switch problems (quite rare). Again, all the CC is doing is obeying its rule, "if brake switch on then disengage CC". Simple as that.

It's the fault detection logic I don't get. What causes it to decide there's a fault to persist? The rule you cite is fine. But even when the brake lights are NOT on (or at least no longer on) the fault remains and you can't use the CC.

My guess is this:
The brake switches have two contacts each. One of each is used to activate the brake lights (or the brake light relay), the other of each are wired in series and in series with one of the two clutch lever switch contacts. This circuit is used by the ECU to deactivate the CC.

The brake light circuit is monitored by the ABS computer. The computers obviously communicate with each other.

Should the ABS computer say the brake light is on, but the ECU doesn't see the deactivate circuit (or vice versa) this is clearly an anomaly. Presumably the system is programmed so that if such an anomaly has been detected, the only safe course is to disable the CC until any problem has been sorted.

So, maybe you have a switch contact problem.

That's my guess.

 

[20valves]

In all the others I never had a single problem with any of these. Not one. And I almost never lubricated a footbrake or either hand lever.

Are you lubing it with something that sets up thick? Why did you feel the need to lube it in the first place? Maybe the lube is creating the problem?

 

[RichDoyle]

In all the others I never had a single problem with any of these. Not one. And I almost never lubricated a footbrake or either hand lever.

Are you lubing it with something that sets up thick? Why did you feel the need to lube it in the first place? Maybe the lube is creating the problem?

I hadn't lubed the rear brake pedal at all, though it was supposed to be done at 4,000 miles. It had the factory lube, which looked OK to me. Nevertheless, as suggested by some on this site, I cleaned it up (noting nothing obvious) and lubed it with the recommended grease (lithium) and adjusted the travel of the rear brake switch. Even though I could detect no problem with the travel of the rear brake. So we'll see what ensues.

Looking at the wiring diagrams (I used to design stuff like this for a living, BTW) there is a misleading representation of the brake light relay in the diagram on 8-85 and it differs from all other representations of this relay in the other diagrams I have found. It's also somewhat incomplete, which added to the confusion. As shown in this diagram, you would deduce that the brake lights would be on unless you pulled the brake lever or pushed the pedal, which would turn them off. This is because the one contact represented is a NO (normally open) one. All the other diagrams show it as a single-pole, double-throw relay, with the brake lights wired to the NC (normally closed) position and the NO position not being used. This would work.

The 8-12 diagram shows what one would expect for disabling the CC, with the 4 disabling NC switches all in series. Circuit opens when one or more switches open and CC disengages the setting, but remains available. (As an aside, with a discrete detection of the clutch handle being released, I know now why the idle could be caused to increase upon releasing the clutch a bit).

Of course, all the diagrams have the weakness that the computers are black boxes and internal logic can only be deduced in a general way. Since the clutch and both brake operators are double-pole switches I could see that a case where the two poles of any of these 3 switches were found to be in an inconsistent state, that would reasonably trigger a fault. For example, the brake lights are on (implying a brake light activation pole is open), but the disengage poles of the switches are closed. This could be caused by some dirt or whatever on one of the switch contacts or in the internal mechanism of the switch between the external activator rod and the switch poles themselves.

As I said, I'll see how this all behaves now with the adjustments I have made. If it pops up again, I'll be looking at the internals of the switches themselves.

Something to keep in mind: This bike is 1 year old, with less than 6,000 miles. I've hardly ever driven it in the rain, this is far from a dusty area, it hasn't been driven on dirt roads, and while it's not garaged, it is in a 5/6 enclosed carport. While it doesn't have a lot of miles, it is driven regularly on 40-ish mile rides. So it really doesn't sit, either.

 

[RichDoyle]

Tentatively, since making the adjustments/lubes described previously things are behaving again. For good measure I also worked over the clutch lever movement and lube. Given this, I infer that at least part of the fault detection algorithm involves the length of time the brakes appear to be on continuously or appear to be on in combination with the bike not behaving like they're actually on.

The ABS warning light never came on in all of this, FWIW.

As an epilogue, I'm inclined to put a small monitor light where I can see it that would light with the brake lights. Just a small LED would do. That's because the only safety-related aspect of the condition I encountered would be the brake lights being on ALL the time, which impedes their effectiveness as a warning to those behind.

 

[HotRodZilla]

Be careful adding stuff Rich. Any perceived added resistance in the brakes' electrical system may keep the CC from engaging. I got the TailBlazer rear lights I have when they prevented a CC from engaging on a 2008.

Word is, there is a difference between aftermarket installs and factory CC, but I'd check with the electrical guys, like Ionbeam, before I did anything...Unless you're also one of those guys. Then have at it and let us know how it goes. Lol...

FWIW, an indicator that the brake lights are on is a very good idea. Doing it without screwing with the CC is the challenge.

 

[RichDoyle]

So - it's back again. I guess I'll start going through the checklist of other stuff to, uh, check.

 

[audiowize]

Be careful adding stuff Rich. Any perceived added resistance in the brakes' electrical system may keep the CC from engaging. I got the TailBlazer rear lights I have when they prevented a CC from engaging on a 2008.
Word is, there is a difference between aftermarket installs and factory CC, but I'd check with the electrical guys, like Ionbeam, before I did anything...Unless you're also one of those guys. Then have at it and let us know how it goes. Lol...

FWIW, an indicator that the brake lights are on is a very good idea. Doing it without screwing with the CC is the challenge.

You could use a darlington transistor to feed the indicator light. A few microamps of current fed to the base of a darlington will turn it on, and I doubt the FJR's computer is going to notice a miniscule amount of current like that.

 

[Harald]

All of the bikes I've had in the last 20 years have had brake lights controlled by switches in both brake controls and electrical interlocks with the clutch handle and the side stand. In all the others I never had a single problem with any of these.
Why then must this bike be so touchy about all this?

I wonder if this may be a case of you not knowing that you had a problem because those bikes didn't have ABS or CC that protest when brake lights are on. Unless a buddy tells you that your brake lights are stuck on, you would never know on one of those "antique" bikes.

 

[Fred W]

Rich,

I do not see where you mentioned this yet: Are you aware that there is another cruise control interruption switch located in the throttle twist grip? If you turn the throttle grip "closed" it opens that switch and cancels the cruise. Could this be the cause of your mysterious cancellations?

As for the brake light circuit's relay logic, it looks like this to me: Both the front and rear brake switches are normally closed contacts supplying +12V to the Brake Switch relay. The rear brake switch supplies power to the relay coil so it is normally energized, and the front brake switch supplies power to the relay's N.O. contacts, which are held closed on the normally energized relay.

That 12V passing through the Brake Switch Relay contacts goes to the Brake Light Relay's coil. So that relay is also normally energized all the time. The power input to this relay's NC contacts is right from the Signaling System Fuse (same power as the horn), but because the relay is energized only gets passed on to the output side when a brake switch is operated. The output side goes to the two brake light filaments and also to the ABS ECU.

If you hit the rear brake switch the Brake switch relay de-energizes, which cuts power to the Brake Light Relay, which allows power out through the NC contacts to the brake lights.

If you hit only the front brake switch it cuts the power to the (still energized) Brake Switch relays contacts, so the Brake Light Relay still gets de-energized, allowing signal power to pass through the NC contacts to the filaments.

The reason for all the funky relay logic is to make the brake lights "fail safe", meaning that a failure of the relays to energize, or of either of the switches to make good contact, will result in the brake lights being on continuously, which you are more likely to notice than the lights not coming on when you do hit the brakes.

The ABS ECU independently monitors the two 12V signals from the two switches to the Brake Switch Relay, so ostensibly it knows if you hit the front brake, the rear brake or both.

The Main ECU gets a signal from the output of the switch relay that is the combined status of the two brake switches. I assume that could be compared to the dedicated daisy chained NC switches that run to the Cruise Control and could trigger a brake switch error code of some sort.

 

[RichDoyle]

Fred, I used to design control circuits like this. You are correct and I got it. The deal is the black-box nature of the ECU inhibits full analysis of what's triggering it. Definitely complicates the process.

But there's new data. I fortunately had it happen a couple of times a few days back and was able to get some more analysis done.

1st case: As I was coming to a stop to wait for the ferry, the ABS light blinked a few times and stayed on. I then pushed the CC button and it blinked a bunch of times. I shut the bike off with the kill switch and saw code 90. Went behind and the brake lights were on. Pulled up on the foot brake - no change. Cycle the front brake lever - brake lights went off.

2nd case: As I was stopping in front of my gate to open it (I had the CC "on" already") the CC light starting flashing. I again stopped the bike with the kill switch. Brake lights were stuck on. As gently as I could I pushed inward on the front part of the brake lever where it engages the pushrod for the front brake switch. If this piece moved, it was so minuscule I couldn't detect it. But the brake lights went off.

I have aftermarket levers, but they've been on there for many months, well before this started. And I can detect no gap between where the lever comes to rest against the mount when the lever is released. Anyway the saga continues, with the front brake switch/lever in the center of it all.

 

[Fred W]

Well, at least you would appear to have it narrowed down to the front brake switch or the relay that it operates.

 

[RichDoyle]

Ok, so here's the final outcome. It was some sort of grunge in the front brake switch assembly. The fix was to remove the switch assembly and partially pull it open, then spray the inside with contact cleaner and exercise the plunger a bunch of times. I made some interesting discoveries along the way.

Some definitions to clarify my description:

"Arming" the CC is pressing the round button with the little speedo symbol on it. That lights up the same symbol on the dash. As long as this is a steady light, the system is armed.

"Engaging" the CC is choosing a speed to hold. It gets "disengaged" by hitting the brakes, clutch, or pushing the throttle fully closed. But the CC remains armed.

"Enabling" the CC means that none of the aforementioned switches is currently in the "disengaged" position.

So:

There are two separate switches in that front switch assembly and they engage/disengage at different positions of the operating pin. If you remove the brake lever and very carefully push in and out on the pin, you can hear/feel each of the two distinct operations, maybe 1 mm in pin travel apart. Using diags 82/83 (and watching the brake lights) I determined that the first one to operate when pulling the handbrake lever (i.e., releasing the pin) is the cruise control disengage switch. Pulling the lever (or releasing the pin further) turns on the brake lights. If the pin is between the two positions, the CC would be disengaged but the brake lights are not on. I didn't test the rear brake light switch in this fashion but I figure it probably works the same way. Transiently this would be a normal condition.

The wiring diagram in the CC section of the service manual leaves out certain information, focusing only on the direct operation. But looking at the full wiring diagram it's clear that the ECU theoretically can determine which of the four enabling switches are being operated and even that the brake lights are on independently of switch positions.

The fault I've been seeing is probably that the CC is "enabled" but the front brake light switch is triggered, which is an inconsistent state, so it prevents the system from being armed, hence the blinking. This fault condition latches and won't reset until the bike is turned off and back on. This is when you have, like I did, an intermittent condition.

It may also be that if the brake lights are not on and the CC is disabled via a brake light switch for some excessive period of time, this would also be a fault, preventing arming the CC. This is kind of a questionable condition to react to. But it would be true that a munged up brake lever or pedal could produce this as a maintained state. At the very least it would prevent engagement, though possibly not preventing arming the CC.

I'm tempted to dig into the switch assembly a bit further to better understand it, but it has this spring inside and I know if that thing gets loose it'll travel through an alternate dimension to who-knows-where. But I can say that after spraying the contact cleaner in there and while operating the pin, it felt like some grit was coming loose inside. I have no clue where it could have come from or what it really was. This would seem more likely for the rear brake switch, given its location, but this is, as I said, the front switch.

 

[steve_g]

Isn't that what i said in post 11? Not as many words but same outcome. The brake switches have 2 contatcs, 1 for cruise and 1 for brake lights.

 

[RichDoyle]

Isn't that what i said in post 11? Not as many words but same outcome. The brake switches have 2 contatcs, 1 for cruise and 1 for brake lights.

Sort of. The thing I learned is they can be independently tested, as they intentionally trigger at different points in the travel of the lever. That, and the the fact that stuff gets involved here that does NOT appear in the SM in the CC section. Because the ECU gets involved, the actual logic causing the bike to detect a "fault" isn't clear from either the manual or any prior explanation. This was complicated by the fact that my situation was intermittent. So the 82/83 diags didn't give me clear direction. In fact, they still don't. I only believe it cured because it hasn't happened in weeks. The 82/83 diags did NOT tell me exactly what was wrong. Both tests were normal every time I ran them.