Headlight modulator causing voltage spikes

Status
Not open for further replies.

SgtWookie

Joined Jul 17, 2007
22,230
The two strands of wire together need to pass through the center of the toroid 18 times, for the pair. In your photograph, you have 4 turns on the toroid for the pair.

If you have two strands of wire that were originally about 3 feet long, wind them on until you have either completed 18 turns, the toroid is too full to wind any more, or the wire gets too short to wrap any more turns on.

Try to keep track of how many turns that you have wound on.
 

Thread Starter

jfairman

Joined Jul 2, 2010
22
good.

i believe you previously explained that I need to make sure the strands don't cross or overlap each other. true?

on my first attempt i had the entire inner circumferene of the toroid full with a single layer of wrapped pairs after just 10 turns, and stopped because I observed I was going to start creating an overlap of the wires passing through the toroid center.

since I know mathematically the inner circumference of the toroid, calculated at 62mm should be able to easily accomodate the combined diameters of thirty six 1.02mm AWG 18 wires, i suspect my wrapping technique is flawed.

somehow I expect I need to figure out how to lay the wraps more tightly against the inner wall of the toroid.

i imagine this is something that takes practice, patience and maybe a trick. all suggestions are welcome!
 

SgtWookie

Joined Jul 17, 2007
22,230
No, I expected that you would wind up with more than one layer of turns. What I meant was to avoid "tying a knot" with the wires. If you accidentally get wires tangled around each other so they wind up twisted, it tends to reduce the quality of the inductor.

It is OK to have more than one layer of wire wound on the toroid. Simply try your best to keep it neat and evenly wound.
 

Thread Starter

jfairman

Joined Jul 2, 2010
22
Almost ready to test it. Waiting for some warmer weather to take it outside and connect it to the 'Wing'.

You'll likely get a kick out of SEEING what all your good guidance has produced! Thanks again for your help so far.

I'll hold off putting foam in the container until it works as expected in testing.

I'll absolutely post a summary of my testing ... probably a couple more weeks.


Cheers!
 

Attachments

Thread Starter

jfairman

Joined Jul 2, 2010
22
I've been doing resistence checks throughout the build to ensure that I'm connecting components correctly at each step of the way.

Relative to the design in post 34: I find that I have less than infinite ( ~5500 ohms ) resistance between the + and - connections on this filter ( eg. between +OUT and -OUT ). Is this to be expected? I assume this is the combined resistance of the two capacitors?

Measured zero resistance between +IN and +OUT.
Measured zero resistance between -IN and -OUT.

Any other bench tests you'd suggest to confirm it's all connected as intended?

Stay safe!!
 
Last edited:

SgtWookie

Joined Jul 17, 2007
22,230
Hmmm - 5.5k Ohms? That might be OK for the capacitor leakage rate. It should be less than 15mA leakage current per cap at 75v, which is 5k Ohms per cap; since they're wired in parallel, then 2.5k is the minimum acceptable resistance. You'll have about a 2.5mA leakage current when the bike is running, which isn't a lot compared to the load of the headlamps.

Are you absolutely certain that you have the polarity on the caps connected properly? If they are connected with the polarity reversed, they will fail very quickly; perhaps rupturing the package(s).

If you are absolutely certain that you have them connected properly, then try connecting it up to 12v DC power (making certain to have polarity correct) via a 1k Ohm resistor for a few days, and monitor the voltage across the resistor to see if it goes down. The resistor is to limit current flow in case a capacitor tries to short out, or the internal resistance is too low.

The leakage rate is measured by the voltage across the resistor. I=E/R, so if you have a 1k Ohm resistor in series supplying current to the capacitors, then each 1v/1k Ohms is 1mA current flow. If you conducted this test right now, you'd see about 2.2v across the resistor with a 12v supply.
 

Thread Starter

jfairman

Joined Jul 2, 2010
22
All systems perform as before ... without the noise on the audio.

This is a success that many told me could not be achieved! Sgt Wookie, I thank you for sticking with me and answering all my foolish questions.

Nothin like some good ole fashion rollin up the sleeves and getten 'er done!

Today you can notch up one more terrific success for this forum.

In the spirit of sharing knowledge I will plan to post a link to this thread from the goldwingfacts.com Technical Forums site, which has just been demoted to my SECOND favorite forum.

Now that all systems are working, I'll proceed to fill my enclosure with Good Stuff to buffer the components from shock.
 

SgtWookie

Joined Jul 17, 2007
22,230
I am glad to hear that you have achieved the desired result - it certainly took a while to get there!

Your questions were not foolish at all, and I DID change directions slightly during the course of the thread. Instead of a simple "PI" filter (cap to ground, single inductor in series, cap to ground) you wound up with a bit more sophisticated setup with a common mode choke.

If you do post a link to this thread on another site, you should mention that people will need to register AND validate their account by responding to the E-mail before they will be able to view the attachments.
 

Thread Starter

jfairman

Joined Jul 2, 2010
22
a photo of the finished contraption ( attached )

really nice to have this be so modular that it can easily be removed.

i did find a little bit of noise ( with volume turned way up ) remains in the circuit, but i think that's what you expected.
 

Attachments

Last edited by a moderator:
Status
Not open for further replies.
Top