Headlight modulator causing voltage spikes

Status
Not open for further replies.

Thread Starter

jfairman

Joined Jul 2, 2010
22
I'm rethinking the placement of the PI filter after the factory plug. Why not get two more "factory" plugs and put the PI filter between the motorcycle wiring harness plug and the light unit plug? This approach seems easier to un-do, and perhaps easier to install. It puts the filter only slightly further away from the modulator.
 

retched

Joined Dec 5, 2009
5,207
I'm rethinking the placement of the PI filter after the factory plug. Why not get two more "factory" plugs and put the PI filter between the motorcycle wiring harness plug and the light unit plug? This approach seems easier to un-do, and perhaps easier to install. It puts the filter only slightly further away from the modulator.
Now you are thinking.

By using the same plugs as the factory connection on the filter, you can EASILY revert the system back to factory without cutting any wires.
 

SgtWookie

Joined Jul 17, 2007
22,230
jfairman,
I received your communique'; however I'm not "firing on all cylinders" so to speak.
One of my sons gave me a cold over the Thanksgiving break. I'd rather have my head on a bit straighter before I reply. Hold off on ordering the caps for a bit yet.

I really goofed up when I fed the parameters to the filter program; 240 cycles per minute is 4 cycles per second! However, all is not lost; you can simply stack up more of those toroids to get far greater inductance. The basic design will be the same as I posted earlier; only the values will change somewhat.

Might go to a 5-pole filter, which will mean another couple of caps and another inductor.

Meanwhile, you say the headlamps are rated for 45W. Is that the high beam? I need to know the total wattage that will be switched on and off by your flasher. If it's indeed 45W, then it has a resistance of about 3.26 Ohms once it's up to operating temperature. When the filament is cold, it will measure as just about a direct short. If it's an LED lamp, your results may vary - considerably.
 

Thread Starter

jfairman

Joined Jul 2, 2010
22
Sgt Wookie,

Each of the two factory headlight bulbs is rated at 45 watts. I have not found specs for the wattage of high-beam vs. low beam settings. I expect this means that the total wattage being modulated is 90 watts. Unfortunately I don't know if the previous owner changed from factory original, but suspect not. Apparently the common replacement bulb is a 55W H4 like these: http://www.electricalconnection.com/other-lighting/bulb_euro-hl.htm If the PI filter is built to handle 90W will it work ok with 110W too?

Your description of the expected ohm measurements makes sense with my experience. Thanks for explaining that. I thought I was going nuts measuring a zero resistance through the bulbs.

I have placed another inquiry to the modulator mfr to confirm the switching cycle specs.

Happy to hold off with further design or purchase of parts until you're recovered and I hear back from mfr.

Good news is that I purchased 10 toroids rather than the minimum order of five.
 

Thread Starter

jfairman

Joined Jul 2, 2010
22
the modulator mfr confirmed the modulator has a 4 cycle per second rate. i have confirmed a source for more of the "factory" plugs so that the filter can simply be plugged in between the existing plugs connecting the headlight module to the bike's wiring.

I am unclear if there is a preference or requirement to have the ground for the PI filter be different than than the existing ground for the headlight module.

still no hurry on this. i took 3 months to pick the project up again. hope you're feeling better.
 

Thread Starter

jfairman

Joined Jul 2, 2010
22
Sgt.

I hope the new year is treating you well!

I’d love to start putting a circuit together with any further advise you’re able/willing to provide.

Your last note suggested that you thought you might have used some incorrect values for modulator cycle time … and maybe something other than the PI design would be appropriate.

Are you willing to take a look at this project again?

jfairman
 

SgtWookie

Joined Jul 17, 2007
22,230
Hello again jfairman!

I'm feeling a good bit better, thanks. That darn cold just wouldn't go away for the longest time. I'm still not back 100% yet, but I'm pretty much over it.

Anyway, going to change tactics a bit with the winding of the toroids. Instead of just one winding, you'll use two windings; this will make a choke that basically eliminates the switching noises, because the transition times will be slowed to a crawl. Possibly the easiest thing to use would be "zip cord" aka lamp cord if you don't have heavy-gauge magnet wire. Either that, or you could order some magnet wire online. It would need to be AWG-21 or larger.

But tell me, do the headlamps have a wired ground, or do they depend upon grounding within the fixture to the chassis?
 

Thread Starter

jfairman

Joined Jul 2, 2010
22
Sgt.

The ground for the headlights is wired via the 4 pin plug that serves the dual headlight module.

Has your recommendation changed from the plan to use a PI filter?

What will be the impact on appearance of the modulation if the transitions are slowed?

jfairman
 

SgtWookie

Joined Jul 17, 2007
22,230
You shouldn't notice a difference in the lighting, just an absence of noise in your radio. :)

So, do you want to try this with some lamp aka "ZIP" cord, or do you want to get the right wire for the job?

The zip cord would be cheap for a test, but I wouldn't want to leave it in there, as the insulation is thick (limiting your turns) and would eventually get hard, crack and fall off, leaving bare wires.
 

SgtWookie

Joined Jul 17, 2007
22,230
Looks like that could be a decent source for wire. I buy stuff like that from a local surplus shop, so don't have a recommendation for a wire supplier.

Have a look at the attached; it's pretty simple. Basically, you'll stack two of those toroids up, and wind 1 layer of masking tape around the outer edge to keep them held together while winding on the turns of wire. You'll be winding two strands of AWG-18 magnet wire at once. You'll need to be careful to not stress the toroids during winding, as they're brittle and somewhat fragile.
 

Attachments

Thread Starter

jfairman

Joined Jul 2, 2010
22
Sgt.

I have incorporated your drawing into a diagram of how this will plug between the motorcycle and the headlight assembly using standard factory plugs. ( .jpg below )

I have a bunch of implementation questions:

1) Are the two “in” wire strand ends of the toroid winding electronically connected to each other & two “out” strand ends connected to each other?

2) I plan to use epoxy to join the toroids as you'd suggested a while back. OK?​

3) Do I need to worry about clock-wise or counter clock-wise winding?​

4) What AWG size would you recommend to connect the components to each other and to the plugs?​

5) What recommendations for mounting, securing or covering these components to protect against each other, the motorcycle fairing housing, vibration, etc? How fragile are they?

6) Will any of the components get warm or hot?

Looking forward to building this!​

Thx.​
 

Attachments

SgtWookie

Joined Jul 17, 2007
22,230
I have incorporated your drawing into a diagram of how this will plug between the motorcycle and the headlight assembly using standard factory plugs. ( .jpg below )
Yes, you have it right.

I have a bunch of implementation questions:

1) Are the two “in” wire strand ends of the toroid winding electronically connected to each other & two “out” strand ends connected to each other?
There are only two ~3' lengths of wire that are wrapped on the paired toroids. You'll wind up with four "ends" in two pairs.

If you want to make it easier to identify the positive and negative "ends", then use a couple of pieces of black shrink tubing for the ground (return) side before you wind the wire on the toroid.

If you get the ends swapped around, it'll make your electrical noise problem much worse instead of better. If you accidentally apply power to your blinker module backwards, you'll probably smoke something.

2) I plan to use epoxy to join the toroids as you'd suggested a while back. OK?
That's fine.

3) Do I need to worry about clock-wise or counter clock-wise winding?
Not really - just make sure you have ground to ground, positive to positive on both ends.

4) What AWG size would you recommend to connect the components to each other and to the plugs?
AWG-18 stranded or larger should be fine.

5) What recommendations for mounting, securing or covering these components to protect against each other, the motorcycle fairing housing, vibration, etc?
One easy way to keep the flexing to a minimum would be to use some low-expansion construction foam that's designed for sealing up around door and window frames. Big Orange and Big Blue hardware stores sell it under the trademark name of "Great Stuff". Make sure to get the low expanding foam. You can just put the assembly in a small can (perhaps like a plastic orange juice concentrate can or something similar), This should keep the solid-core wiring flexing to practically nothing, and the toroid will be completely supported.

How fragile are they?
About like a china coffee cup handle.

6) Will any of the components get warm or hot?
Not if you have the capacitor polarities wired correctly. Power dissipation will be very low.
 

Thread Starter

jfairman

Joined Jul 2, 2010
22
i believe my question 1) above resulted from a mis-interpretation of the drawing and explanation. I thought that L1 was a stack of two toroids with a two strand wrap, and L2 a separate two toroid stack with a two strand wrap.

after some puzzling and re-reading, my understanding is that there is just a single two torid stack, with two strands. one strand is L1 and one strand is L2.

i'll post a photo when i get it put together. with ordering plugs and capacitors, i expect it'll be a couple of weeks.

i like the idea of using some great stuff in a plastic container. C1, C2, L1/L2 all together in the container?

cheers!
 

SgtWookie

Joined Jul 17, 2007
22,230
i believe my question 1) above resulted from a mis-interpretation of the drawing and explanation. I thought that L1 was a stack of two toroids with a two strand wrap, and L2 a separate two toroid stack with a two strand wrap.
Sorry, I guess I didn't explain that well enough. L1 and L2 are coupled inductors; basically a transformer, but in this circuit they are common-mode inductors used to reject changes in current.

after some puzzling and re-reading, my understanding is that there is just a single two torid stack, with two strands. one strand is L1 and one strand is L2.
That is correct!

i'll post a photo when i get it put together. with ordering plugs and capacitors, i expect it'll be a couple of weeks.
OK, if you are still uncertain, I suppose that I could wind one up here and take some photos.

I like the idea of using some great stuff in a plastic container. C1, C2, L1/L2 all together in the container?
Certainly. Just make sure that all of your connections are correct prior to injecting the foam, or you'll have one heck of a mess trying to chisel it out without destroying just about everything. Make certain that the stranded wire you get for making the connections to the wiring harness is pretty finely stranded; if it only has a few strands, then it is very likely to break right at the project container due to vibration and metal fatigue.
 

Thread Starter

jfairman

Joined Jul 2, 2010
22
Allright!

Just starting. Ran into an important clarifying question:

Post 33 on 2/6 specifies two strands and 18 turns around the toroid. Is this 9 turns for each of the two strands? or 18 turns for each of the two strands, for a combination of 36 turns through the toroid?

I attempted 18 turns for each of the two strands and ran out of room on the inner circumference of the toroid. admittedly my turns could have been laid down closer agains the inner circumference.

thanks!
 

Attachments

Status
Not open for further replies.
Top