HID Headlight Transformer Interference

5 messages Options
Embed this post
Permalink
Jeff928

HID Headlight Transformer Interference

Reply Threaded More More options
Print post
Permalink

 Hello EvListers,
  First off Happy 4th of July !
  My problem today, actually for over a year, I just didn't know it, is this dc-dc converter(transformer) for my HID headlights.

http://www.flickr.com/photos/13947721@N03/3688081608/
http://www.flickr.com/photos/13947721@N03/3688081342/

 After nearly two years of driving my 928 Porsche conversion I finally put a radio in it. Only to find I have a lot of radio electrical noise, but not from the usual suspects,controller,brushed motor.
  Here is what I have tried so far to minimize the interference.
Tried capacitors on the 12v input (220uf & 2200uf), lights wouldn't come on !
Diode to ground on the input side
ran the 12v source through a relay directly from my 12v battery
put the unit in a aluminum sealed box (trying to eliminate conductive inter.)
  None of these made a difference.
  I also wanted to know if the noise was inductive or conductive, so I powered the transformer from a separate 12v battery (all interference went away, so its Inductive, right ? )
  Now I admit I'm an electronic hack, so please be gentle and not to technical with your advise.

Thank you all in advance !
             Jeff



_______________________________________________
General EVDL support: http://evdl.org/help/
Usage guidelines: http://evdl.org/help/index.html#conv
Archives: http://evdl.org/archive/
Subscription options: http://lists.sjsu.edu/mailman/listinfo/ev

evan foss

Re: HID Headlight Transformer Interference

Reply Threaded More More options
Print post
Permalink
On Sat, Jul 4, 2009 at 3:39 PM, Jeff Mccabe<[hidden email]> wrote:

>
>  Hello EvListers,
>  First off Happy 4th of July !
>  My problem today, actually for over a year, I just didn't know it, is this dc-dc converter(transformer) for my HID headlights.
>
> http://www.flickr.com/photos/13947721@N03/3688081608/
> http://www.flickr.com/photos/13947721@N03/3688081342/
>
>  After nearly two years of driving my 928 Porsche conversion I finally put a radio in it. Only to find I have a lot of radio electrical noise, but not from the usual suspects,controller,brushed motor.
>  Here is what I have tried so far to minimize the interference.
> Tried capacitors on the 12v input (220uf & 2200uf), lights wouldn't come on !

Adding capacitors in parallel to the input side of the dc/dc should
have at worst done nothing, not stopped the lights from turning on.

Also I would have started with a smaller value. Larger caps are for
dealing with lower frequency fluctuations (more often sudden spikes in
power use). Small caps will pass RF.

> Diode to ground on the input side

This was not likely to do anything.

> ran the 12v source through a relay directly from my 12v battery
> put the unit in a aluminum sealed box (trying to eliminate conductive inter.)
>  None of these made a difference.

So the noise isn't traveling threw the air but back up stream from the
HID DC/DC to the +12V source and then to everywhere else.

>  I also wanted to know if the noise was inductive or conductive, so I powered the transformer from a separate 12v battery (all interference went away, so its Inductive, right ? )

You got good data from your experiments but not the right conclusion.
Your last experiment isolated the two things and stopped the problem,
so it is conductive. I think you might try an inductor in series with
the lamps to block the noise from propagating. How far (physically) is
the DC/DC from the Lamps? You want them and anything blocking the
noise to be close in proximity to minimize any RF prorogation (Not
that you seem to have evidence of any).

>  Now I admit I'm an electronic hack, so please be gentle and not to technical with your advise.

The capacitors were a good guess.

> Thank you all in advance !
>             Jeff

Evan


--
http://www.coe.neu.edu/~efoss/
http://evanfoss.googlepages.com/

_______________________________________________
General EVDL support: http://evdl.org/help/
Usage guidelines: http://evdl.org/help/index.html#conv
Archives: http://evdl.org/archive/
Subscription options: http://lists.sjsu.edu/mailman/listinfo/ev

Roger Stockton

Re: HID Headlight Transformer Interference

Reply Threaded More More options
Print post
Permalink
In reply to this post by Jeff928
Jeff Mccabe wrote:

>   My problem today, actually for over a year, I just didn't
> know it, is this dc-dc converter(transformer) for my HID headlights.

>   I also wanted to know if the noise was inductive or
> conductive, so I powered the transformer from a separate 12v
> battery (all interference went away, so its Inductive, right ? )

The more usual terms are conducted and radiated.  Conducted means the interference is getting from the source to the destination via a conductive path (the 12V power or ground connections).  Radiated means the interference is getting from the source through the air as RF energy (same as the signal from the radio station gets to your radio receiver).

If powering the DC/DC from a separate source eliminated the interference, then you have determined that the interference is conducted.

> Tried capacitors on the 12v input (220uf & 2200uf), lights
> wouldn't come on !

I can't think of any reason why capacitors between 12V and ground would possibly interfere with the DC/DC starting.  You did try the caps between 12V and ground, not between 12V and the DC/DC, right?

You are on the right track with the capacitors.  I would try much smaller values, such as 0.1uF or 1uF,and connect it/them between 12V and ground as close to the input to the DC/DC as you can.

The idea here is that capacitors look nearly like a short circuit to high frequencies, so connecting one or more capacitors between 12V and ground "shorts" out the high-frequeny noise that comes out of the DC/DC on the 12V or ground lines.

Smaller value capacitors are more effective for higher frequencies.  The 220uF and 2200uF caps you tried are relatively large values and will be electrolytic types, which may not be very effective if the noise from the DC/DC is fairly high frequency.

I would start with a 0.1uF film or ceramic capacitor and see how it does.  If the noise isn't reduced enough, add a 1uF film in parallel with it.  If it still isn't reduced enough, add the 220uF or 2200uF in parallel with the other caps.

If you can't reduce the noise sufficiently with caps alone, try adding an inductor in series with the 12V line, between the caps and the car.  If the noise still isn't reduced sufficiently, try moving the inductor to between the caps and the DC/DC.

This article may explain a bit better or in greater detail:

<http://www.popularmechanics.com/how_to_central/automotive/1272491.html?page=3>

You may be able to find an off-the-shelf noise filter that you can install between the DC/DC and the 12V system.  From your pictures, we can see that the DC/DC draws 3.2A in normal operation and a peak of 20A, so you need a filter capable of handling these sorts of currents.

This one, for instance is rated 15A/250W:

<http://www.mcmelectronics.com/product/60-1055>

As long as it can tolerate the 20A peak (I'm assuming it is a turn-on surge), this is an inexpensive option.

Here's some beefier (and more costly) options:

<http://www.tessco.com/products/displayProducts.do?groupId=619&subgroupId=65>

> Diode to ground on the input side

I wouldn't expect this to help.

> ran the 12v source through a relay directly from my 12v

This has the potential of helping, but you would need to use a double-pole relay so that you switch both 12V and ground to the DC/DC.  Twist the 12V and ground leads together between the battery and relay and between the relay and DC/DC, and use the double-pole relay to connect the DC/DC directly to the 12V battery.

Cheers,

Roger.

_______________________________________________
General EVDL support: http://evdl.org/help/
Usage guidelines: http://evdl.org/help/index.html#conv
Archives: http://evdl.org/archive/
Subscription options: http://lists.sjsu.edu/mailman/listinfo/ev

Jeff928

Re: HID Headlight Transformer Interference

Reply Threaded More More options
Print post
Permalink
In reply to this post by Jeff928

Happy 4th  Roger,


Sent from my iPod

On Jul 4, 2009, at 9:14 PM, Roger Stockton <[hidden email]> wrote:

Jeff Mccabe wrote:

 My problem today, actually for over a year, I just didn't
know it, is this dc-dc converter(transformer) for my HID headlights.

 I also wanted to know if the noise was inductive or
conductive, so I powered the transformer from a separate 12v
battery (all interference went away, so its Inductive, right ? )

The more usual terms are conducted and radiated.  Conducted means the interference is getting from the source to the destination via a conductive path (the 12V power or ground connections).  Radiated means the interference is getting from the source through the air as RF energy (same as the signal from the radio station gets to your radio receiver).

If powering the DC/DC from a separate source eliminated the interference, then you have determined that the interference is conducted.

Good, got that one right


Tried capacitors on the 12v input (220uf & 2200uf), lights
wouldn't come on !

I can't think of any reason why capacitors between 12V and ground would possibly interfere with the DC/DC starting.  You did try the caps between 12V and ground, not between 12V and the DC/DC, right?

Oops , no , I tried them in series with the positive side .


You are on the right track with the capacitors.  I would try much smaller values, such as 0.1uF or 1uF,and connect it/them between 12V and ground as close to the input to the DC/DC as you can.

Thanks, I'll try this one tomarrow.


The idea here is that capacitors look nearly like a short circuit to high frequencies, so connecting one or more capacitors between 12V and ground "shorts" out the high-frequeny noise that comes out of the DC/DC on the 12V or ground lines.

Smaller value capacitors are more effective for higher frequencies.  The 220uF and 2200uF caps you tried are relatively large values and will be electrolytic types, which may not be very effective if the noise from the DC/DC is fairly high frequency.

I would start with a 0.1uF film or ceramic capacitor and see how it does.  If the noise isn't reduced enough, add a 1uF film in parallel with it.  If it still isn't reduced enough, add the 220uF or 2200uF in parallel with the other caps.

Sounds like a plan, thanks

If you can't reduce the noise sufficiently with caps alone, try adding an inductor in series with the 12V line, between the caps and the car.  If the noise still isn't reduced sufficiently, try moving the inductor to between the caps and the DC/DC.

This article may explain a bit better or in greater detail:

<http://www.popularmechanics.com/how_to_central/automotive/1272491.html?page=3>

You may be able to find an off-the-shelf noise filter that you can install between the DC/DC and the 12V system.  From your pictures, we can see that the DC/DC draws 3.2A in normal operation and a peak of 20A, so you need a filter capable of handling these sorts of currents.

This one, for instance is rated 15A/250W:

<http://www.mcmelectronics.com/product/60-1055>

As long as it can tolerate the 20A peak (I'm assuming it is a turn-on surge), this is an inexpensive option.

Yes, the current quikly drops as the lights brighten.


Here's some beefier (and more costly) options:

<http://www.tessco.com/products/displayProducts.do?groupId=619&subgroupId=65>

Diode to ground on the input side

I wouldn't expect this to help.

ran the 12v source through a relay directly from my 12v

This has the potential of helping, but you would need to use a double-pole relay so that you switch both 12V and ground to the DC/DC.  Twist the 12V and ground leads together between the battery and relay and between the relay and DC/DC, and use the double-pole relay to connect the DC/DC directly to the 12V battery.

I'll try this also, I have a double pole laying around someware.



Cheers,

Roger.

Thanks again Roger !

Jeff
_______________________________________________
General EVDL support: http://evdl.org/help/
Usage guidelines: http://evdl.org/help/index.html#conv
Archives: http://evdl.org/archive/
Subscription options: http://lists.sjsu.edu/mailman/listinfo/ev


_______________________________________________
General EVDL support: http://evdl.org/help/
Usage guidelines: http://evdl.org/help/index.html#conv
Archives: http://evdl.org/archive/
Subscription options: http://lists.sjsu.edu/mailman/listinfo/ev

Jeff928

Re: HID Headlight Transformer Interference

Reply Threaded More More options
Print post
Permalink
In reply to this post by Jeff928

Evan,
 Thanks , your advice agreed with Rogers.
  I have several things to try tomarrow.
 Happy 4th,
Jeff

Sent from my iPod

On Jul 4, 2009, at 9:12 PM, evan foss <[hidden email]> wrote:

On Sat, Jul 4, 2009 at 3:39 PM, Jeff Mccabe<[hidden email]> wrote:

 Hello EvListers,
 First off Happy 4th of July !
 My problem today, actually for over a year, I just didn't know it, is this dc-dc converter(transformer) for my HID headlights.

http://www.flickr.com/photos/13947721@N03/3688081608/
http://www.flickr.com/photos/13947721@N03/3688081342/

 After nearly two years of driving my 928 Porsche conversion I finally put a radio in it. Only to find I have a lot of radio electrical noise, but not from the usual suspects,controller,brushed motor.
 Here is what I have tried so far to minimize the interference.
Tried capacitors on the 12v input (220uf & 2200uf), lights wouldn't come on !

Adding capacitors in parallel to the input side of the dc/dc should
have at worst done nothing, not stopped the lights from turning on.

Also I would have started with a smaller value. Larger caps are for
dealing with lower frequency fluctuations (more often sudden spikes in
power use). Small caps will pass RF.

Diode to ground on the input side

This was not likely to do anything.

ran the 12v source through a relay directly from my 12v battery
put the unit in a aluminum sealed box (trying to eliminate conductive inter.)
 None of these made a difference.

So the noise isn't traveling threw the air but back up stream from the
HID DC/DC to the +12V source and then to everywhere else.

 I also wanted to know if the noise was inductive or conductive, so I powered the transformer from a separate 12v battery (all interference went away, so its Inductive, right ? )

You got good data from your experiments but not the right conclusion.
Your last experiment isolated the two things and stopped the problem,
so it is conductive. I think you might try an inductor in series with
the lamps to block the noise from propagating. How far (physically) is
the DC/DC from the Lamps? You want them and anything blocking the
noise to be close in proximity to minimize any RF prorogation (Not
that you seem to have evidence of any).

 Now I admit I'm an electronic hack, so please be gentle and not to technical with your advise.

The capacitors were a good guess.

Thank you all in advance !
            Jeff

Evan


--
http://www.coe.neu.edu/~efoss/
http://evanfoss.googlepages.com/

_______________________________________________
General EVDL support: http://evdl.org/help/
Usage guidelines: http://evdl.org/help/index.html#conv
Archives: http://evdl.org/archive/
Subscription options: http://lists.sjsu.edu/mailman/listinfo/ev


_______________________________________________
General EVDL support: http://evdl.org/help/
Usage guidelines: http://evdl.org/help/index.html#conv
Archives: http://evdl.org/archive/
Subscription options: http://lists.sjsu.edu/mailman/listinfo/ev