Two Prong Flasher Auto Cancel Mod

[crazy steve]

I thought (?) I read that the relay was heating, not sure how it would do that without full voltage. And to heat, it would have to be on for quite a while...

Low voltage is what causes the heat to some degree, and will certainly cause overheating in this case. The relay is a inductive device (same as an electric motor), and it's trying to do 'work'. On a resistive load (like a lamp or heater), reducing voltage simply lowers the heat/light output with no harm other than not getting full output. But on an inductive load (in this case, where the coil is trying to move the plunger to operate the contacts), it will still try to do the same 'work' (however many watts that takes) and if voltage is low, the circuit will draw more current to try to make it up. More current = more heat. This rarely comes up in electronics, but is a factor in any electrical device that physically moves something, whether it's a solenoid plunger or an electric motor. In this case, you're trying to move that plunger against a fixed load (the spring) and it'll take X amount of watts to do that.

[DAVINCI]

Thanks for the lesson.

OFF = cool. Gotta turn that coil off.

[cywelchjr]

I thought (?) I read that the relay was heating, not sure how it would do that without full voltage. And to heat, it would have to be on for quite a while. I'll wait for Cy's reply before I yell D-uh!

Steve, it's heating cause it's on 90% of the time, what I've drawn will turn it off. I think even if he gets full voltage by curing some ground problem or whatever, the heating problem will still be there. The trick here is to turn that coil off.

Turns out it's the auto cancel not being able to reliably source the 30MA that the relay wants, and all the relays that can handle the ~10amps needed to insure working with normal bulbs need at least that much. The answer appears to be a TWO relay approach, at least if you are not using LEDs. I found a high sensitivity relay that only needs 12ma to pull in that I can use as the driver off the auto cancel to then supply the cancel trigger relay off the brown wire. I'll do a schematic of the modified circuit. I found I supposedly can get the relay I need from the same store I got the other, so I'm going shopping again today, then installing this new relay into the line.

[cywelchjr]

Thanks for the lesson.

OFF = cool. Gotta turn that coil off.

Yeah, but that turns the flashers back on. Catch 22. As I just posted, the problem isn't after all the relay, but the auto cancel that's getting overloaded and dropping it's voltage, it's a solid state device. So the answer is to lower the load on it with a relay it CAN handle, to trigger the relay that can handle the current that normal bulbs pull.

[crazy steve]

Turns out it's the auto cancel not being able to reliably source the 30MA that the relay wants, and all the relays that can handle the ~10amps needed to insure working with normal bulbs need at least that much...

Yep, that would cause the problem. But you might check to see if the input voltage to the canceller is up to snuff, as that could cause this also. The load current on the canceller with the stock flasher should be about 160 MA (there's 75 ohms between the C and L terminals), more than enough to drive that relay. I think what you have should work, but there's other problems causing the low current/voltage. You may have a marginal canceller too... hard to say.

Like I said, what you have should work....

[DAVINCI]

Yeah, but that turns the flashers back on. Catch 22. As I just posted, the problem isn't after all the relay, but the auto cancel that's getting overloaded and dropping it's voltage, it's a solid state device. So the answer is to lower the load on it with a relay it CAN handle, to trigger the relay that can handle the current that normal bulbs pull.

Duh, because the handlebar switch is still in the L or R position. For it to work as i envisioned it, the switch would have to be centered, which is the same as turning it off manually, round and round I go.

[DAVINCI]

OK, after this I shut up. Use a transistor to run the relay.

[crazy steve]

OK, after this I shut up. Use a transistor to run the relay.

Yeah, I thought that too, but wasn't sure if one could carry that much current...

[cywelchjr]

Yep, that would cause the problem. But you might check to see if the input voltage to the canceller is up to snuff, as that could cause this also. The load current on the canceller with the stock flasher should be about 160 MA (there's 75 ohms between the C and L terminals), more than enough to drive that relay. I think what you have should work, but there's other problems causing the low current/voltage. You may have a marginal canceller too... hard to say.

Like I said, what you have should work....

Ahh, but you forget the bulbs that are between that load and ground, that L pin isn't hooked directly to ground, but is rather the hot line for the directional bulbs, which includes 2 27 watt bulbs and one what 7 watt I think for the indicator?

Ok, I just ran the calculations and the bulbs only add something like 2.631 ohms. Now your testing it when it's idle, and the only way to truly test it is to run the test with it in circuit and measure the current it draws while it's cancelling. Then you can calculate the apparent resistance of the unit in use, which may well be higher than the static resistance. I had started out to calculate it on paper, then thought, "self, you have a smartphone, and I'll bet there is an ohm's law calculator for it". So I looked and there was, so I just plugged in the numbers .

So, I think your right, it SHOULD work as is. So, the question is, why is it not. I may need to go through and see why I'm not getting the voltage I should, since I should be able to source a good double what I'm getting, unless the OEM flasher has a significantly higher dynamic resistance than it's static resistance. Hmm, time for some more research. It sucks that I don't have an OEM flasher anymore, on either of my bikes (my 400 as I recall uses the same flasher and the 11).

[cywelchjr]

Duh, because the handlebar switch is still in the L or R position. For it to work as i envisioned it, the switch would have to be centered, which is the same as turning it off manually, round and round I go.

Even centered the brown wire is still hot until you press the switch in, in which case the signals are off anyways . If it's held to one side or the other, then the auto cancel unit will never kick in, cause it's held in a disabled state and doesn't even start counting the pulses from the reed switch until you let the switch return to center.

BTW, I just went back and referenced the schematic of the OEM flasher that ELDR posted back in 2008, and as I thought, it looks like the resistance of the OEM flasher would be MUCH higher when it's in cancelled mode, probably many times higher actually, so a static reading of it's resistance is just testing it's lowest level of resistance, which is BEFORE it cancels, and doesn't even take into account that 12v is being fed in from the other pin at the same time and there is a connection in there. So I'm guessing the draw when cancelled is WAY less than 30ma, or at least a bit less, like maybe half, maybe even less than that. So I'm going to go ahead with your transistor idea, as I think at this point it's the way to go as the first choice, with second choice being the second high sensitivity relay, but the transistor is cheaper and it at least can be sourced almost anywhere, although if your ordering the relay, you may as well order all the parts at the same time.

[DAVINCI]

Even centered the brown wire is still hot until you press the switch in, in which case the signals are off anyways . If it's held to one side or the other, then the auto cancel unit will never kick in, cause it's held in a disabled state and doesn't even start counting the pulses from the reed switch until you let the switch return to center.

BTW, I just went back and referenced the schematic of the OEM flasher that ELDR posted back in 2008, and as I thought, it looks like the resistance of the OEM flasher would be MUCH higher when it's in cancelled mode, probably many times higher actually, so a static reading of it's resistance is just testing it's lowest level of resistance, which is BEFORE it cancels, and doesn't even take into account that 12v is being fed in from the other pin at the same time and there is a connection in there. So I'm guessing the draw when cancelled is WAY less than 30ma, or at least a bit less, like maybe half, maybe even less than that. So I'm going to go ahead with your transistor idea, as I think at this point it's the way to go as the first choice, with second choice being the second high sensitivity relay, but the transistor is cheaper and it at least can be sourced almost anywhere, although if your ordering the relay, you may as well order all the parts at the same time.

Right, the lever self-centers, but the switch stays on until you push in and center the switch itself.

[cywelchjr]

Right, the lever self-centers, but the switch stays on until you push in and center the switch itself.

Right, which turns off the signals right away. The auto cancel is for if you don't push it in. Further reading of how the OEM unit works (ELDR did a really good job there) shows me the problem is in my auto cancel unit, as there should be MORE than enough capacity from the auto cancel unit to run this relay, and this relay should handle the load easily. So, I've got to find why I'm losing voltage, which means I'm going to be cleaning contacts . But it also means the schematic as published is correct, no other components are needed, as the relay draws a fraction of what the OEM unit does (30 mA vs 180mA). I'm probably losing the voltage in my stupid 4 way switch, the darn thing is worn out. I may have to find a replacement, as looking at the wiring diagram that is where the brown wire gets it's voltage, so the whole system may be low, then again it might be another connection, I gotta check each of the voltages, good thing I've got a good DVOM with a fresh battery in it.

[CaptonZap]

Last year, my canceling unit would not work. Using the "pry the crimped cover off" method, ala speedometer repair routine, I was able to look inside the canceling unit.
There are some contacts inside, and they were really corroded. After a bit of emery paper treatment, it works a treat, to quote James.
You might check the voltage coming out of yours, as well as the drop during current flow. JAT CZ

[cywelchjr]

Last year, my canceling unit would not work. Using the "pry the crimped cover off" method, ala speedometer repair routine, I was able to look inside the canceling unit.
There are some contacts inside, and they were really corroded. After a bit of emery paper treatment, it works a treat, to quote James.
You might check the voltage coming out of yours, as well as the drop during current flow. JAT CZ

Yep, I gotta check input and output, but I'm suspecting that it might be in there. If so, I may have some cleaning to do, and thing may then work just sweet.

Of course I've got to get the thing open and then get it cleaned up and sealed back up afterwards if this is the case .

Knowing it's not solid state (which I thought it was) makes it more likely I can fix it if it's the problem).

[3Phase]

Cy, Randy, Steve, PMFJI but I just got back online after yet another pending hard drive failure and had to test the new drive offline before transferring any data to it.

A Darlington pair, prepackaged or discrete, should be able to use the low current to drive the flasher directly and get rid of the extra relay or am I totally misreading what's wanted here?