[theshadow27]

Okay, someone asked for a way to cut down on the AM/FM noise that HID ballasts produce. There are three factors that could produce interference in an HID system:

Quote, originally posted by theshadow27 »

1) The ballast itself. This is unlikely since it is enclosed in a metal case which is grounded to the chassis (see Faraday Cage) 2) The wires between the CEM and the ballast. The ballast operates via a 200-500khz boost converter. If this is poorly filtered it will spit out EMI back down the supply lines. Since they were designed for halogen bulbs they are unshielded, and basically act as a 15 or 20 foot antenna to broadcast that interference. This can be fixed pretty easily, with a capacitor-inductor filter at the ballast, which will prevent high frequency transients down the line. 3) The HV wires between the ballast and the bulb. These are shorter and thus have less radiative length. But it's possible that these are the culprit, as they do transmit HV at HF. Unfortunately there's not much you can do about these, because the ballast requires a direct connection to the bulbs to ensure proper ignition and arc stabilization. What you can do is shield the HV lines, and the easiest way to do this is with stainless steel hose sleeving, and then ground that shield to the chassis. This will greatly cut down on the amount of radiated EMI.

So lets get started.

1) Ground the ballast.

Zip ties are not good enough. The ballast should be physically bolted to the car's chassis. Most kits come with brackets, and sheet metal (preferably aluminum, or stainless steel) can be used to make custom brackets. If you're lazy, use self tapping screws. I drilled holes and used stainless steel hardware (hex cap bolts) that threaded into my slim ballasts.

Even though the case is grounded through the negative wire, there's a bunch of wire between the ground point and the ballast which has the potential to act as an antenna. Grounding the case of the ballast cuts that wire out of the loop. It will also help keep it cool, which is never a bad thing.

2) Filter the input.

Even if the negative lead is grounded, there is about 15 feet / 5 meters of wire in two harnesses between the positive (yellow) wire in the headlight and the battery. This can easily act as an antenna for back EMF coming off of the flyback or boost converter. To cut this noise out, we use a low-pass pi-filter:

The capacitor-inductor-capacitor "pi" form the lowpass filter. It's tuned to cut anything above 1khz. The larger capacitor and series diode act to filter input ripple from the PWM, and the reverse-bias diode eliminates any negative transients that develop within the filter.

Components:

Quote »

BOM - HID INTERFERENCE REDUCTION FILTER (2x) 2x M9975-ND INDUCTOR FIXED 1.0UH 10% RADIAL $1.13 http://search.digikey.com/scri...75-ND 4x 565-1782-ND CAP 2200UF 16V ELECT LXY RAD $1.06 http://search.digikey.com/scri...82-ND 2x 493-1546-ND CAP 10000UF 16V ELECT HE RADIAL $2.04 http://search.digikey.com/scri...46-ND 4x 497-2738-5-ND DIODE SCHOTTKY 25V 10A TO-220AC $1.50 http://search.digikey.com/scri...-5-ND TOTAL COST, COMPONENTS, 2 UNITS: $16.58

The circuit should be constructed on a solder perf board or with fly leads. Then it should be put in an enclosure and completely covered (potted) with RTV silicon. For this to effectively filter noise, it should be as close to the ballast as possible, preferably next to it. For best results, connect the negative wires to the harness, the ballast, AND to a close chassis ground (maybe the bolt securing the ballast, eh?). I'd also bolt it to the car, but that's not necessary for this guy.

3) Shield the HV lines.

A great deal of noise may come from the HV lines between the ballast and the bulb, especially if your ballast has an external ignitor (small box inline with HV wires). There's no easy way to directly filter this, nor would you want to, since it would reduce the efficiency of the ballast. The best way to combat this EMI is to get some stainless steel hose sleeves (~20 from Advance Auto/Pep Boys) and run the HV components inside of the sleeve.

Remember, to act as a shield the sleeve must be grounded. The easiest way to do this is using a hose clamp at one end to clamp a ground wire to the sleeve. Don't over-tighten, or you may damage the wires underneath. It doesn't really matter if it's a loose fit, as long as it runs continuously from the ballast to the headlight assembly.

You can use any combination of the methods described above. They will all help. I personally don't have interference issues so I can't say for sure, but I'd recommend doing them in this order: 1, 3, 2 and check each time to see if the problem is fixed. Cheers

 

[ryanMR2]

Re: Reduce HID ballast radio interference (theshadow27)

Thanks for the write up, new project for tomorrow http://********************/smile/emthup.gif

 

[theshadow27]

Re: Reduce HID ballast radio interference (ryanMR2)

Np. Let us know if you find something that works.

 

[thisdothack]

Re: Reduce HID ballast radio interference (theshadow27)

Definitely ground the ballast, that should be in the installation instructions I would think. LOL.
For cheap effective shielding; Get some decently thick copper tape from a craft store and run it down the HV wires lengthwise rolling it around them as you go (leaving no gaps, so you'll probably need like 1/2" or more to overlap depending on the gauge). Next, wrap this in heavy-duty weatherproof tape leaving a small exposed amount of copper on each end. Shrink wrap the end of the tape down for pro durability. Then use a small metal hose clamp or metal cable tie-down (pressed tight but not crushing) and tie that into the frame somewhere, metal to metal, on the exposed copper at the ends of your wire. I'll have to post pics of this technique if I find any time. Shadow's stainless tubing idea will work, but I think it would be a bit awkward to fit nicely in the tight spaces of the engine bay and might rattle with the vibs of driving.

 

[theshadow27]

Re: Reduce HID ballast radio interference (thisdothack)

The copper tape would be easier, but I don't like the look of wires spiral wrapped with tape

Plus it'd be a PITA if you had to disconnect the wires in the future.

The SS stuff isn't that bad to work with, it's really flexible and actually looks pretty good. This is the stuff:

which expands and goes around wires like:

then contracts by pulling it lengthways. There's no real rattling potential since it snugs up against the wire as long as it's held at both ends.

Amazon has a set of every size imaginable for $20:
http://www.amazon.com/Mr-Gaske...WCG8Q

They also sell the copper foil you mention for $20:
http://www.amazon.com/JVCC-CFL...tcorr

 

[thisdothack]

Re: Reduce HID ballast radio interference (theshadow27)

Oh... I thought you were talking about the in-wall type spiral wrap ss tubing. You meant braided cable wrap. That stuff would be nice, might get a little gross after a while collecting stuff in the braiding. But then again, tape tends to rot out over time too.

 

[fkon]

I have a similar problem, has anyone tried this low-pass filter? EDIT: I'm referring to the filter described by theshadow in the post at the top.

The inductor for instance is rated for 10A but measurements show that my ballasts use approximately 70A peak current each for 1ms. Is this acceptable, i e is 10A for steady state but higher peak currents are OK for short durations?

Will the inductor not saturate from the high DC-current?

 

[fkon]

Some details of my issues:
I have installed after-market(chinese) Xenon HID High-beams in my car. These have a ballast which transforms 12VDC from the car battery to 23kV AC required to ignite the lamps. The problem is that every time i turn on the HID:s the car stutters and warning lights inside the car(ABS, EPS etc)turn on.

I work with R&D for a car company so I've conducted measurements both on the car itself and on separate HID-kits and a spare car battery. Unfortunately my area of expertise is combustion engines so i have only a rudimentary knowledge of electronics. My conclusion so far is that there is high frequency interference returned from the ballasts which propagate through the DC-cables(the error disappears when i use an external battery so airborne interference is not the cause) to the battery which in turn cause a reset of some node on the CAN.

On the battery, oscillations with and amplitude of of ~150V and a frequency of 15-50MHz occur when the HID:s are turned on. I don't know if they are common mode or differential mode, it is difficult to get repeatable measurements with the oscilloscope since it picks up the airborne noise also.

The ballasts use approximately 70A each in peak current during the first ms after the lamps are turned on but it appears(also a bit uncertain) that the disturbances occur before the large current peak.

I've tried filtering the current cables from the battery to the ballasts close to the ballasts with capacitors, 1, 10, 100 nF(which should have low impedance in this frequency range) and ferrite chokes scavenged from computer cables. These help to some extent but don't eliminate the problem completely. The error still occurs from time to time.

My next step would be to order ferrite cores with correct permeability from an electronics supplier. However, if the EMI is not common mode i will have to pass the + and - cables from different directions and the ferrite cores will become saturated(because of the 70A current) and will not be effective against the EMI, right?

Some questions:
1 Do you know if EMI from these types of transformers is usually common mode or differential?
2 How should I wind the cables around the cores?
3 Should I even use ferrites or something else entirely? Inductors?
4 If the EMI is differential mode and i cannot find ferrite cores which do not saturate, how should I go about in adressing the problem?

As I said, this is not my field of expertise so maybe I'm asking the wrong questions and maybe I'm not providing sufficient background information, in that case comment and I'll try to be more clear!

I'm in a bit over my head here so I'm extremely grateful for any assistance!

 

[Flechette]

Hello,

I'm about to build the filter circuit you've posted. I have one question though - I know for sure that I have no PWM on the input whatsoever, my HID ballasts are purely relay driven, so I assume I can just leave out the rectifier section and only do the pi filter? Won't that affect the filters frequency cutoff tune though? It would look somewhat like this:

Cheers.

 

[theshadow27]

Hello,

I'm about to build the filter circuit you've posted. I have one question though - I know for sure that I have no PWM on the input whatsoever, my HID ballasts are purely relay driven, so I assume I can just leave out the rectifier section and only do the pi filter? Won't that affect the filters frequency cutoff tune though? It would look somewhat like this:

Correct, looks good... But if you are just looking for a filter, for $10 you can get a pair of DDM BOW3 which implement the complementary inductive T filter with the same cutoff and frequency response as the above circuit.

 

[Flechette]

Correct, looks good... But if you are just looking for a filter, for $10 you can get a pair of DDM BOW3 which implement the complementary inductive T filter with the same cutoff and frequency response as the above circuit.

I would, but since I'm from the Europe, it wouldn't be economically feasible to order it since the shipping would cost me way too much, and I don't mind building it myself, got the gear to do it I will see if I can integrate this filter into the slim ballast case, if not, I'll use connectors off some extension wires that were provided with the ballasts.

 

[theshadow27]

Also, not sure how I missed the above (2012!) but for posterity and the benefit of future googlers:

the filter described by theshadow in the OP... The inductor for instance is rated for 10A but measurements show that my ballasts use approximately 70A peak current each for 1ms. Is this acceptable, i e is 10A for steady state but higher peak currents are OK for short durations? Will the inductor not saturate from the high DC-current?

Yes, higher current is acceptable for short durations. However, the peak current itself will be reduced by the inductor, which will restrict di/dt. Saturation is based on the steady state current at the time of measurement (not the AC component), which is 0 at initial turn-on, so by definition it's not saturated under this condition.

I've tried filtering the current cables from the battery to the ballasts close to the ballasts with capacitors, 1, 10, 100 nF(which should have low impedance in this frequency range) and ferrite chokes scavenged from computer cables. These help to some extent but don't eliminate the problem completely. The error still occurs from time to time.

Sounds like you've done some good research, but you can't approach power filtering from a signal filtering standpoint. There is no level of acceptable AC in the car electrical system - therefore, the bypass capacitor should have a low impedance at ALL frequency ranges (very large capacitor).

Also, the large currents involved make parasitic elements non-negligable; meaning you have to account for the resistance in each connector and section of supply wire. This means putting the filter as close as possible to the source.

Some questions:
1 Do you know if EMI from these types of transformers is usually common mode or differential?
2 How should I wind the cables around the cores?
3 Should I even use ferrites or something else entirely? Inductors?
4 If the EMI is differential mode and i cannot find ferrite cores which do not saturate, how should I go about in adressing the problem?

1) Differential. You can insert the filter on the positive or negative wire, the convention being to use positive.

2) Ferrites like the ones from computer cables have an extremely high mu and are designed to prevent low level common mode HF with a single pass of all conductors through core. They will saturate immediately (reducing the inductance to that of an air core coil) at the currents involved here. To get the inductive effect your are looking for, you'd need a less permeable core and a number of wraps (10-20t) of one of the wires or a stock inductor.

3) It is easier, cheaper, and smaller to shunt the objectionable current using a bypass capacitor than it is to block it using a series inductor. A large (mF aka 10,000uF) capacitor placed across the ballast leads will short-circuit the initial startup surge (regardless of it's frequency) thus preventing it from reaching the car. The bypass diode protects the capacitor from reverse voltage, and the series input diode prevents any negative transients (regardless of voltage) from making it back to the car.

4) Use a capacitor, don't bother with an inductor.

 

[theshadow27]

I would, but since I'm from the Europe, it wouldn't be economically feasible to order it since the shipping would cost me way too much, and I don't mind building it myself, got the gear to do it I will see if I can integrate this filter into the slim ballast case, if not, I'll use connectors off some extension wires that were provided with the ballasts.

I don't think it will fit inside a ballast case... but if you get it post pictures

 

[Flechette]

A large (mF aka 10,000uF) capacitor placed across the ballast leads will short-circuit the initial startup surge (regardless of it's frequency) thus preventing it from reaching the car. The bypass diode protects the capacitor from reverse voltage, and the series input diode prevents any negative transients (regardless of voltage) from making it back to the car.

So you're saying it will actually be more beneficial to include that 10,000uF low ESR cap afterall even though my car does not power that ballast with PWM? My problem is pretty much that when I switch the HID fog lights on, I get an ABS, Brake failure and DSC lights on, then they disappear. And while the ballasts are working, I get slight FM radio interference.

 

[theshadow27]

So you're saying it will actually be more beneficial to include that 10,000uF low ESR cap afterall even though my car does not power that ballast with PWM? My problem is pretty much that when I switch the HID fog lights on, I get an ABS, Brake failure and DSC lights on, then they disappear. And while the ballasts are working, I get slight FM radio interference.

If there is no interference when running the HID off a separate battery (not in the car) then I would try the diode+cap first and see if it fixes it. Maybe order 6x or 8x 10000uF and wire them in parallel.

The advantage of a pi/tee filter is that the cutoff frequency is sharper - a slope of 20db / decade, where a straight R-C (resistor in the wires) filter is 10db / decade. But with a very large capacitor, the cutoff frequency is very low (<100Hz) so by the time it gets to 1KHz (2 decades) the advantage is negligible.

 

[Flechette]

I forgot how small they were, no way I'm putting anything in it! Although I think I will be able to ground the metal backplate of it. Looks like it's a metal PCB and the back is grounded. I had it zip tied inside the bumper only, not good enough. But I wasn't getting noticeable interference until I tinted my rear window (which has traces that are used as an antenna) and I had to add a second FM antenna in the front of the headliner. When I put a typical clamp on ferrite core on the power wire close to the ballast, the interference while it is fired up and running got smaller, but then it started to mess up my on board instrumentation on startup. Wish I had a scope, I'll get one in less than a month, will be fun!

Here's some pics of the client in question. The connector is already rusty, and looks like it rusted from the inside too - and that's only a year of use. Must be China's luxury model.

 

[theshadow27]

Maybe it is time for new ballasts? That one look very cheaply built!

 

[Flechette]

Maybe it is time for new ballasts? That one look very cheaply built!

Yeah, I've looked around the market, but it's flooded with generic Chinese units over here. I tried to buy Xentec, but the import costs for me are just immense! I guess I'm stuck with this crap.

Running a grounding wire from the PCB back plate to nearest body screw helped with the radio interference it seems, but I still get some burst of disruption when they start up which "reset" my DSC / ABS module for a second or two. Interestingly when they were left for 5 minutes or so, then turned off and on, I havent got that reset issue. Must be using lower voltage right from the start on the hot ionized lamps. I wasn't quite able to test whether or not they produce the "reset issue" on startup from an external power source, my spare gel acid decided to pack up. I'm quite sure this issue won't come up when I test power it that way.

Either way, waiting for parts to arrive, then I can print some PCBs and get on with it

 

[theshadow27]

So a real HID enclosure will look something like this:

i.e. Potted, shielded, PCB screwed to metal chassis, etc and I am concerned that you are wasting a lot of time/money for not much gain... Your ballast PCBs look like toys.

There are many reputable vendors in Europe, my friend Les in the UK has a excellent product... http://hid50.com I actually import my ballasts from him into the US.

My feeing is that a decent ballast will resolve 100% of your issues.

 

[Flechette]

I would gladly buy from him, but I just can't spend that sort of money on it. It's just a Fog light that gets rarely used, my main headlights have OEM factory HID setup which works fantastically. I've looked at some online stores, but all that I can see are no name brandless copy of a copy replica of a replica Chinese products with God knows what's inside of them. If I can't fix the current ones I have, I'll attempt to find something that has listed picture of the inside, fairly easy to spot a decent ballast by looking at the electronics.

Cheers!