LED turn signal question.

[Wallowgreen]

Steve,

Here's my 0.03 (I've already chipped in my 0.02!)...

From looking at the wiring diagram and doing a bit of research on electronic flasher relay units here's what seems most logical, assuming the three contacts on the electronic flasher relay are: +12v, GND and switched +12v out.

I think the controller makes the GND connection for the electronic relay when it receives a demand from the handlebar switch to enable either signal pair. The relay then has 12v across it and begins to operate sending out the pulsed 12v on the Lbl wire back to the handlebar switch. It's at the handlebar switch the correct lamp pair are fed with the pulsed 12v on the Lbl wire.

If all this guess work is correct, the self cancel feature should not be impacted at all by the installation of LED signals. As you're well aware, the problem with LED signals is one of loading on the relay output side of things, not the control of whether the relay is enabled or not - that's the job (and only job) of the controller unit.

I'll be figuring this out for my own reasons before too long so I'll try and do a little digging on my 1100 this weekend, that has the same set up. If so, I'll post what I find but having said all that, I still haven't actually tried what you're proposing.

For reference, I believe the wires to be used as follows...
on the relay:
O/G - +12v
B/Bl - GND
Lbl - switched +12v out (pulsed to make the signals flash)

on the controller:
O/G - +12v
Bl/R - speed sensor
B/R - speed sensor
Lg/B - signal request input from switch
B/Bl - switched GND to relay
Br/Y - Cancel input from switch
B/W - GND

Hope this sheds a little more light - no pun intended of course!

[sponsored1]

The badlands load equalizer WORKS! At least for me it did. With the stock flasher unit and all led turn signals. It blinks at a steady rate and and my self cancelling feature works fine. It's only $25 and fixed all my problems on my 78gs1000. Why even mess around with anything else? It's 3 wires and installs in less then 2 mins.

Someone allready posted this link, but here it is again:

http://www.denniskirk.com/jsp/product_catalog/Product.jsp?skuId=H50046&store=&catId=&productId=pH50046&leafCatId=&mmyId

[Skateguy50]

I will add what I can but am sure you know more on this than me, but I have some "experience"

Ok this was on my wifes 85 honda rebel 250, so didnt have any auto off feature, and come to think of it my 83 gs 650gl doesnt seem to shut off unless I shut it off.

Anyways her bike came with some stock parts and some after market stuff and I pieced together what I could to get it goind as cheap as possible, the PO supplied some LED signals for the back and had enough parts to make the stock fronts work again, so thats what I did.

My first attempt included using load equalizers bought off the shelf at autozone, they worked great and did the job... for about a week. The vibrations where to much for them to stay together, I had them mounted under the rear fender loosely tied in an out of the way spot that would also allow them to cool and not hit anything. Regardless they didnt hold up.

I then got the electric turn relay and it worked great wired with led in the back and stock in the front. Again no auto off stuff on the rebel, but the moral of my little story is the load equalizers might work great in a car but didnt hold up on the bike.

[Steve]

From looking at the wiring diagram and doing a bit of research on electronic flasher relay units here's what seems most logical, assuming the three contacts on the electronic flasher relay are: +12v, GND and switched +12v out.

I think the controller makes the GND connection for the electronic relay when it receives a demand from the handlebar switch to enable either signal pair. The relay then has 12v across it and begins to operate sending out the pulsed 12v on the Lbl wire back to the handlebar switch. It's at the handlebar switch the correct lamp pair are fed with the pulsed 12v on the Lbl wire.

If all this guess work is correct, the self cancel feature should not be impacted at all by the installation of LED signals. As you're well aware, the problem with LED signals is one of loading on the relay output side of things, not the control of whether the relay is enabled or not - that's the job (and only job) of the controller unit.

Wallowgreen,

I think there are too many assumptions there, and I'm not even going to make any of the usual jokes about that.

I should finally be able to get around to doing my own testing on this matter in about three weeks. So far, the only thing I have been able to verify is the result of one quick test, and that was explained in post #33, back on 12 JUN 2008. I turned on the signals, then used a test light to probe the B/Bl wire to see if it was HIGH or LOW. As soon as I touched the wire, the signals stopped. Because of this one test (I have not even gone back to test it with a high-impedance DVM), I am guessing that the controller puts a HIGH signal on the line to turn on the flasher unit. I am guessing that the flasher unit is a typical thermal unit because it responds to varying loads, but also has something in there that allows the signal from the B/Bl wire to enable it to flash. When the signal is removed by grounding it, the flashing stops.

As I see it, the controller unit gets a START signal from the turn signal switch via the Lg/B wire. It turns ON and sends a signal to the flasher via the B/Bl wire, then starts its sequence of counting time and measuring speed. When its turn-off criteria is satisfied, it removes the signal from the B/Bl wire and the flashing stops. If this is, indeed, the way it works, then I will determine the limits of that signal and use it to power a relay that will, in turn, provide steady power to a generic electronic flasher that won't care what kind of lights it's powering. Hope to know in less than a month. Stay tuned. (I am really surprised that this thread is still active after what? almost a full year. )

.

[Riding Again]

Was looking for something else and had stumbled upon this which features a timed auto cancel.



found at http://www.chromeglow.com/catalog.asp?prodid=500390

Haven't tried it, but you can check it out.

[Matchless]

Some feedback from my side:

1) Replacement LED bulbs are not efficient for starters, even with 36 leds's. I have a set and they are still less bright than the stock bulb.

2) Led's require a regulator to protect them and to give out maximum light - regardless of the voltage across the battery at idle and high rpm.

3) Truck led clusters are brighter than stock bulbs and they use individual reflectors for each led and only 19 superflux led's make a good light.

4) Loading the led flashers to draw more current and make the OEM flasher work properly and thus retaining the self cancelling feature is not a good idea. One of the reasons for going to led's is to reduce the current consumption and heat, the load resistors defeats this purpose. The other is the difficulty of proper mounting of such resistors and the eventual vibration failures if not fitted properly.

5) Retaining the self cancelling feature is very important and may even save your life at some time!!! It may sound cool to tell people you do not need it, but....

6) The self cancelling feature can only be retained properly when using
led flashers if a purpose built flasher unit is designed and used, that matches the self cancelling unit. There is no ready built one on the market.

7) Led's have much faster response when coming on and going off than the incandescent bulbs and thus will react faster and more visible to voltage fluctuations as well.

8) Led's have a designed voltage and current limit specific to a type and make of the led. If these limits are not considered then the output will not be optimal or the lifespan of the led will decreased. They are not all the same! In our case the led's need more than just a series resistor.

9) The GS flashers do not have enough space to fit a proper off the shelf led cluster with built in voltage regulation. As said earlier standard replacement led bulbs are just too weak light wise. Output and mounting position are not optimal.

10) Basically all this leads to further complications, truck led clusters come with built in voltage regulation and spike protection, but are too bulky to fit.

11) I have put together an electronic FET P-channel flasher unit that can flash any led lights and also puts out a regulated voltage to the flashers. It retains the self cancelling function as it uses the inhibit signal coming from the self cancelling unit. The flasher unit plugs into the existing flasher socket and just needs an extra ground wire which can be picked up on the black/white just under the seat coming down to the self cancelling unit.

12) I have cannibalised some 100mm truck flashers and have the round 100mm PCB's with 19 Superflux led's mounted out and managed to trim them down slightly on a bench grinder so they just fit inside the lens width. With the reflectors on they are very bright. I just have to give some thought on how to insulate and mount them in place of the old bulb holder and reflector.
The layout and the method of powering the leds' is quite important and these have 3 led's in series with a resistor, with 6 sets of 3 in parallel and a single led with resistor in the center. The led's run at about 46 mA and have 2.2 Volt across them with the regulator regulating the voltage so that at 15V charging you do not exceed the voltage and blow the led's and 12 volt you do not have them glowing faintly. The regulator keeps the voltage to the led clusters at about 9.5V give or take a couple of mV.

OK that's it, my flasher prototype version 2 is working and is built on a small PCB, in a small plastic enclosure and fits in the same spot as the OEM flasher unit.

At this stage we know that there are different versions of self cancelling units out there, but does anyone know if any of the OEM flashers units are interchangeable with other manufacturers who also use a 3 prong flasher unit, but uses a different self cancelling design?

Hope this helps a bit.