Motorcycle Flasher Issue

Posting your schematic would be best. This way we can better assess where noise decoupling and surge protection will be most effective.

Chris

 

It may be interference as you surmise but it may also be excess voltage. Motor cycle voltage control is primitive to say the least.

 

There's no Vcc decoupling there and there's probably none on the input to your converter either. It's also not a bad idea to precede the input to the converter with a low value resistor followed by a 9V Zener with anode connected to Gnd.

Are you really using a 2N3055 power transistor? This is a very high power model with very low beta. With an automotive lamp in the collector circuit that 1K base resistor would not provide enough base current to saturate it. It would not be my choice for diving high current LEDs either.

Chris

 

Before we go there, what's the specs on the converter? 6-12V converter doesn't tell us much. What's the min-max input range, max output current, input current demands? Do you have a link to it?

EDIT: Did you measure the bike's 6V line while the bike was running? Also, while a DMM is slow, you might be able to switch the DMM to AC and see if it picks up any readable noise or ripple on the line.

Chris

 

See attachment for addition of noise filtering. I also recommend that you lower R2 as shown. Even this value is high for a 3055 but it ultimately depends on what the LED current is in the collector circuit. The general rule of thumb when saturating a transistor is that the Base current should be Ib = Ic/10 or times less than the collector current. Admittedly, this rule isn't that relevant when the collector is switching light load currents though. Do you know what your LEDs draw.?

Chris

 

Yes, a 330 Ohm should put you close enough to the ball park. I suspect you will see an increase in the LEDs brightness too. By the way, to give you an idea of how different a 3055 is from a common small signal switching transistor, the 3055 has a max base current of 7A! No, that's not a typo.

Chris

Edit: I didn't want R2 to go any lower than 330 Ohms because I stuck that 10 Ohm decoupling resistor in the Vcc leg. I didn't want the Vcc to dip too much when the output is driving the base.

 

Yes, it's big time overkill for your needs. Your 350mA load would have been best handled with a small power MOSFET. My second choice would have been a TIP120 Darlington.

Chris

 

Sorry that those mods didn't help. Perhaps a photo of the board and the interconnects to the bike might help. I think ignition noise is getting in there. It can be very severe and difficult to eliminate. Wire length, grounding points, decoupling cap lead length, all can play a part here. Spark noise extends well into the RF bands. If it gets onto the Trigger and Threshold pins it can make a 555 very erratic. You may have to resort to a metal case with RF feed - through caps..

Chris

 

Ive seen people try to run Motor bike blinkers from a magneto before. Rarely (never in my experience) goes well. Unless its OME.

Could I suggest a new approach? Get a small secondary 6v battery, maybe 4 rechargeable AAA's. Or maybe even a super cap. Charge it from the magneto, then when you turn the blinker on, have the magneto disconnect from the battery and run a flasher circuit, isolated from the rest of the electronics. If you don't need blinkers a lot, I would consider a small solar panel and not have it connected to the bikes circuit at all.
You might even get away with 4 AAA's(6v) to your boost circuit then run the lights you have...


If you find another solution, please PM me, as I have come across this problem 20-30 times in the last 20 years... Never seen a solution work with the magneto... Whats the rectifier look like? If it's not FWBR (doesn't look like one in your circuit diagram) You might want to try putting one between original battery and boost circuit. Look for one that can handle a few hundred volts...

I wouldn't be surprised if the magneto killed you boost circuit either. They can put out some wacky voltages... Ive seen headlights blow at high rpm... Especially with enduros from the 70's. Not so much with modern bikes.

 

I've been thinking about it too. Early in the thread you said that the boost converter is connected to the battery through a bridge that you added. Long shot, but have you tried eliminating the bridge? A storage battery is a very low impedance device that filters garbage well close to its terminals. I'm wondering about the necessity of the bridge and it's effect on this noise. Is it an aid or hindrance? Is there a large filter cap on the output of the bridge?

Chris

 

You've certainly done a good job of trying to isolate the problem. You're consistently one step ahead of me.

Have you tried connecting both the + & - input of the boost converter directly to the battery?

Grabb'n at straws.

Chris