HID/PWM issue DIY fix for BOW (Bulb Out Warning), flickering, and poor starting

[theshadow27]

SwedeSpeed P1 (C30, S40, V50, C70) HID Retrofit Index

1. SKBOWE Plug-and-Play harness for P1 HID conversion
2. Background info, general questions, and share your HID setup
3. KBOWE How-To (Original 2009 Version) build your own PWM filter (you are here)
4. Original 2009 investigation and technical discussion on S40 HID issues
5. ForceFed’s E46 D2S Bi-Xenon Projector Retrofit guide
6. Rewire OEM Bi-Xenon headlights to work on the halogen harness
7. DRL Disable Alone not enough for safe Aftermarket HID retrofits
8. LizardOfBodom's EU/RHD Bosch ("E46") Bi-Xenon replica retrofitting writeup + angle eyes + running lights

This thread is the official "How To" writeup of the KBOWE, a DIY PWM filter for P1 cars - please only post here for support in building, installing and testing an KBOWE.

As of June 2017, if you don't want to build this yourself there is finally an assembled version of the KBOWE called the SKBOWE. See skbowe.com for a plug-and-play version of this PWM filter for HID retrofits in P1 cars.

If you'd like to share your experiences on an aftermarket HID installation working (or failing!) in a P1 car, that is S40/V50/C30/C70N, with or without this sort of box, please post in the statistics thread. Thanks!

---

NOTE: If you've got an S60, it might not be possible to disable the PWM modulation on the low beams. The KBOWE described below should (additional information would be appreciated) allow a successful retrofit (although the install instructions will be different), however there may be an easier way with a relay harness and the parking lights.

"Told Ya So" disclaimer:

WARNING: Do not attempt to operate an aftermarket HID kit in a P1 S40 / V50 / C30 / C70 before reading and understanding the issue described below. While a kit may turn on, the PWM DRL regulation will drastically shorten the life of the HIDs, and can potentially damage the (Windshield) Wiper Motor Module (WMM) and/or the Central Electronic Module (CEM), both of which are several hundred dollars to replace.

Newer Volvos, specifically the P1 chassies (S40N, V50, C30, C70N), with halogen (incandescent) headlamps use the low-beam projector for the Daytime Running Lights (DRLs). Because the bulb is on whenever the car is running, Volvo engineers decided to add a form of voltage regulation called Pulse Width Modulation (PWM) which extends the life of the halogen bulb. This form of regulation prevents aftermarket HID ballasts from operating correctly. he KBOWE described below is one method to allow an aftermarket HID system to be safely under such conditions. It works by filtering the PWM voltage, turning the pulses into clean DC voltage.

The device described below only partially addresses the problems with HID retrofits in P1 cars. PLEASE read this overview completely before you proceed!

KBOWE Write Up

This write up covers the construction of what DDM calls a "BOW Eliminator", or what I call a PWM filter. Its purpose is to let aftermarket HIDs run on factory wiring on Volvos (or any car) with PWM voltage regulation on the low beams. The S40N/V50/C30 have this "feature", as do some other newer Volvos. so this thread is just how to build it.

Required tools:
- Soldering iron + solder
- Wire nips
- Crimper
- X-Acto knife
- Step bit (for grommet)
- Dremel abrasive disk (or cutoff wheel)

Required parts:

2x 22,000uF (22mF) 25V capacitors minimum. Recommended: 33,000uF (33mF) or 47,000uF (47mF) 35V, rated for at least 2000 hrs at -40c - 105c. e.g. ECE-T1VA333FA

2x 10A 50V TO-220A Schottky rectifiers minimum. 40A 200V recommended for safety margin. Ultra low forward voltage drop is key, MBR1045PBF-ND

1x Prototyping board, perforated.

1x Plastic project enclosure large enough to hold the capacitors. You can use any enclosure you want, or none at all (just pot with silicone and wrap with electrical tape).

1x 1/2" Rubber grommet

4x Male bullet connectors (or Quick Disconnects)

4x Female bullet connectors (or Quick disconnects)

~2m 14 or 16 gauge stranded automotive wire, various colors and black

1x Tube of RTV silicone


The capacitor specification is a minimum, the bigger the better up to around 50mF. Capacitors in parallel add - so you can get less expensive smaller capacitors and wire them together:

The enclosure doesn't matter as long as it can hold everything.

Parts Cost - If you order enough to make just one and have nothing lying around the shop to use.
2x Diodes @ $0.31
2x Caps @ $2.36
1x PCB perf board @ $1.99
1x Enclosure @ $2.69
1x Grommet pack @ $1.99
1x Pack bullet connectors @ $1.95
1x Tube RTV silicone @ $5.97
6' 14AWG stranded 4 conductor @ $0.90/ft
Total = 25.33. If there's no radioshack near you order everything from mouser/ebay. Shipping will probably be more than parts, so I'd buy extras.

Modding the case
If you decide to use 4x2x1 radioshack enclosure, you'll find it's too short to fit the 30mm caps. I used two 7/8" sockets and a heat gun to indent the lid so the cap:

Move the heat gun quickly over the case until it softens up. Then apply an even pressure to smoothly bring out the bumps. You may have to try this a couple times before it works out right - just keep test fitting the caps until they just slightly have wiggle room. Silicone will fill in the gaps later, so it should not be a press fit.

Next, you'll need a 1/2" hole for the grommet:

It's easiest to put it in before everything else.

Circuit overview
The schematic for the circuit is:

In real life, these parts look like this:

Building the diode board
Install the diodes, bending the leads down so the TO-220 case lies flat with the board.

Insert wires into the PCB with the diodes. You should use two different colored wires here, I used red for "in" and orange for "out". Make sure the right color is going to the right terminal on the diode (see circuit overview). If you mount them like below one will be backwards (see blue dot in pic for the anode) so be careful.

Here's how we're gonna wire it:

Flip over the board and solder the wires to the diode leads. Make sure to get nice hot joints, cold solder joints break fast in a car.

Not shown here is two additional wires that come off and go to the caps. They are connected to the red pads on the reverse side. You'll see them later on.

Wire the capacitors
For the negative terminals on the capacitors I used a single black wire, stripped the end, and sliced the insulation further down. This lets you push the insulation towards the stripped end and makes two nice places to solder. If it's easier you can do it with two wires.

Here's the two extra red wires. They go from the back of the board (this was just to save space) on each red pad to a capacitor.

I nipped, stripped, and soldered them in place to make sure the wires weren't too long. Also everything was threaded through the grommet.

It's easiest to see in the next photo, there's a small white piece of plastic between the PBC and the caps. I put that in there thinking it would stop the caps from rolling into the PCB, but looking at it now it's unnecessary since the whole thing is gooped anyhow.

Everything assembled at tucked in:

You might have to use a Dremel tool to take off the smaller plastic standoffs in the bottom of the case to get everything to lie flat.

Testing
At this point (with everything how it will live when the top goes on) it's a good time to do some tests.
Physical testing: Wiggle all the wires by hand and make sure nothing breaks off. Yank hard on each wire in the bundle and it should remain intact.

Electrical testing: Unless you have a bench-mounted HID work light (like me ) the first test will have to be in the car. There are a couple of options for this, I tested it after I cut the wires at the CEM (see the appendix), using an alligator clip to ground at the cigarette lighter. It's also possible to test in the engine bay, assuming you have a long enough ground lead to make it to the frame. Best to try one side (of the circuit) at a time.

It will be pretty obvious if you've hooked something up wrong, either the HID won't light or a capacitor will explode. If the HID doesn't light up, the most likely cause is either the diode hooked up backwards, or you have switched the input and output when installing it in the car (see troubleshooting).

Potting
If all goes well, disconnect everything and fill the case past the rim with clear RTV silicone (the colors have other stuff in them). Make sure some gets under the capacitors and there are minimal air bubbles and gaps. This is half full:

Once it's topped off (everything completely covered) screw on the top and wipe the excess silicone from around the joint. As long as you're not too hard on the wires you can install it right away. To be extra safe you can wait a day or two, but it will cure just as slowly in the car as out.

Install
Steps for install are:
0) Unlock all doors, remove the key, open the driver's door, and wait for the interior lights to turn off.
1) Remove two clips and cloth kick panel cover from under glove box.
2) Unscrew the two thumbscrews securing the CEM (fuse box) to the chassis.
3) Swing down the CEM (fues box) - it is on hinges.
4) Unplug the right most connector (green) on the fuse access side by unlatching the retaining clip and swinging it about 100 degrees.
5) Determine the two correct wires to tap. Consult Appendix A for additional information.
6) Cut both wires about 3" from the connector, and strip off some insulation.
7) Crimp female bullet connectors onto the wires on the connector side of the cut.
8) Crimp male bullet connectors onto the wires on the harness side of the cut.
9) Plug the BOWE input wires (that go to diodes) into the female bullets on the connector side.
10) Plug the BOWE output wires (that go to the capacitors) into the male bullets on the harness side.
11) Zip-tie the BOWE wires to the factory harness.
12) Re-insert the green connector into the CEM and latch it in place.
13) Swing the CEM back into position
14) Secure the CEM using it's two thumb-screws

15) Remove the small plastic nut securing the carpet to the wall (near the door jam) on the passenger side.
16) Remove the door still by pulling straight up.
17) Fold the interior carpet away from the door.
18) Follow two black wires that run along the body down to a ground stud
19) Bolt the black wire to one of the ground studs

20) Shove the BOWE somewhere under the dash
21) Fold the carpet back, secure it with plastic screw thing
22) Snap the door still back in
23) Re-clip the cloth kick-panel cover under the CEM.

* disclaimer - this is from memory. If it's not exactly right, use your imagination.

Troubleshooting

There is a relatively simple test procedure with a DMM (volt meter) and a 9V battery (or car battery) to determine if a KBOWE is working correctly. There are five wires that come out of the box: two input, two output, and one ground (call GND). Call the battery terminals B+ and B-.

  (1) Connect GND to B-

  (2) Connect all other wires to B+ for 10-15 seconds

  (3) Disconnect battery without letting GND touch any of the other wires

  (4) Measure voltage between GND and each other wire (using DMM, connect DMM- to GND and DMM+ to one wire at a time)

  (5) Using masking tape or something similar, label the voltage measured on each wire

Interpreting test results ("low" = voltage reading close to zero, "high" = voltage reading close to battery voltage, e.g. 9V or 12V):

* If all four wires read low [DEFECT CONDITION]:

   (a) You used the wrong lead as GND -> Repeat test with correct ground wire

   (b) Problem with connections inside of KBOWE -> Correct connections inside of KBOWE

   (b) Problem with capacitor wiring inside KBOWE -> Replace capacitors and check circuit connections

* If all four wires read high [DEFECT CONDITION]:

   (a) Diode anode is hooked to capacitor positive terminal (circuit constructed incorrectly) -> Correct circuit connections 

   (b) Diodes are defective -> replace diodes

* If three wires are low and one is high, or three wires are high and one is low [DEFECT CONDITION]:

   (a) One of the channels has a problem. Same corrections as above.

* If two wires are high, and two wires are low [TARGET CONDITION]:

   -- The two wires that are high are the outputs

   -- The two wires that are low are the inputs

   -- The circuit is functioning correctly

If you know for sure that you have the input wires connected to the CEM, and the output wires connected to the wiring harness, then the only option is to remove the BOWE and correct the problem.
You can't just switch the red and orange leads, the capacitor must be on the cathode side of the diode (closest to the HID). And if it is already potted, unfortunately you will have to build another KBOWE, since there is no good way to un-pot a circuit like this. Test the next one before potting!

----------- Will be updated as info comes in -----------

Appendix A: 2004.5 -2007 Volvo S40/V50 CEM wires:
Pin 25 green-violet (passenger)
Pin 37 green-white (driver)

The connections to the circuit are:

------------ Working installs -----------------------------

1. My original prototype of this project is still working (9-17-2009 - 2005 S40)
2. Kyle finally got this installed (11-11-2009 - 2006 S40)
3. Built one for Cliff after one of his Sharp HID BOWE's failed (12-21-2009 - 2005 S40)
4. Mika built one and got it working with 35w HIDs (1-17-2010 - 2005 V50)
5. vinopa30 got one working with 55w DDM slim ballasts in a 2008 C30 (2-26-2010)
6. After a long battle, we got heico2.4i's 35w Raptors working with a KBOWE V3 in a 2006 S40(3-26-2010)
7. Ben got his KBOWE V3 hooked up with 35W DDMs in a 2005 S40 (1-24-2011)

[rushin_max]

What about 35w HID's?....doesn't look like this filter has anything to draw extra current.
Or do you have another idea for that?

[theshadow27]

Expanded working installs...

==== #1 ====
My original prototype of this project is still working (9-17-2009 - 2005 S40):

Starts and stops with no issues. Flashes pretty fast too

==== #2 ====
Kyle finally got this installed (11-11-2009 - 2006 S40):

On the 55W DDMs BOWE works like a champ being that I removed the DDM BOW resistors and still fired right off.

==== #3 ====
Built one for Cliff after one of his Sharp HID BOWE's failed (12-21-2009 - 2005 S40):

repaired my darn SharpHIDs that died after only 3 months ... the same box as he built for Kyle.

Interestingly, even though Cliff didn't disconnect the HID or Sharp BOWE from the car for two weeks, the KBOWE fired it up the first try. This surprised even me

==== #4 ====
Mika built one and got it working with 35w HIDs (1-17-2010 - 2005 V50):

my build seems to be working very nicely with 35W ballasts (not that there should be problems with 35W).

This is great news, I knew it would work for sure with 55 watters, but Kyle was having trouble with his 35w. It's probably a ballast thing - some are more efficient than others and in this case we want as inefficient as possible (to get it closer to 4 amps)

==== #5 ====
vinopa30 got one working with 55w DDM slim ballasts (2-26-2010 - 2008 C30):

08 c30

Which confirms that they are in fact the same electrical system (you can never trust VADIS...)

==== #6 ====
After a long battle, we got heico2.4i's 35w Raptors working with a KBOWE V3 (3-26-2010 - 2006 S40):

Thanks for the patience and the help Jacob...

Another 35w confirmed!

==== #7 ====
Ben got his KBOWE V3 hooked up with 35W DDMs (1-24-2011 - 2005 S40):

Thanks to theshadow for the BOWE, I installed it today and everything seems to be working with 35w DDMs. Thanks for the help

That's another 35w kit.

==== #8 ====
Laze got his working (01-16-2015 - 2005 S40)

I can add that I successfully have installed the KBOWE a month ago. No flickering and windscreen wipers still working. What a relief and what an improvement light-wise.

=========
Original reply:

What about 35w HID's?....doesn't look like this filter has anything to draw extra current.
Or do you have another idea for that?

That's a good point, it hasn't been tested (yet, Kyle's working on it) with 35w ballasts. On the S40 anyway, there is no current monitoring per se but it's closed loop voltage regulated. It determines a defective bulb if it can't keep the voltage within a given range, and shuts off the light. In theory, a 35w ballast should work just as well, but there's no way to know for sure without testing.

I did a range of tests (for startup current) with the 55w and 35w ballasts which showed that there was less than an amp of difference between the two:
http://theeshadow.com/files/volvo/hidcomp.html
But nothing really on what happens after they warm up. Stay posted, if I have time I'll throw a 35w in and see what happens.

[rushin_max]

well, when I had the 40 I threw in 35's with a cap filter and a few resistors to bring up the current. Worked for about a week, then burned out my wipers. I didn't have a diode though....gave up on it after burned out 2 wiper motors.

35's by itself with filter just turned off, cem wasn't seeing enough current.

I was working on a circuit, similar to urs but with an added pot and a transistor so I could adjust how much current I could sink. I don't have the 40 anymore, so kind of gave up on that idea...but maybe u could figure it out.

[theshadow27]

I read up on the wiper motor issue, as far as I can tell it's caused by a ground connection shared between the headlights and the wiper motor. My guess is that if you have a capacitor between the positive and negative leads of the headlight wire without a diode, when the CEM goes to open circuit (PWM low) the cap can put the potential of the ground connection below 0v because the JFET has a flyback diode connected to V+. My diode blocks current flow backwards through the CEM, and I've been running this for a couple months with no issues.

The first time I tried a cap it didn't work at all, even with the 55w. I tried adding 6.8ohm resistors to increase current draw and that didn't help. If you look through the theory thread I came up with a bunch of crazy ideas on how to rig up a current sink using a shunt resistor and opamps... But with the cap and diode it worked like a champ the first time so I didn't need it.

[chinaonnitrous1]

I was wondering if something like this could be used to disable the Daytime running lights on our s60r, by unplugging them, and using this to turn off the bulb out message.

[theshadow27]

Quote, originally posted by chinaonnitrous1 »

I was wondering if something like this could be used to disable the Daytime running lights on our s60r, by unplugging them, and using this to turn off the bulb out message.

Nope, this doesn't use any current, it just filters dirty voltage. You could use a resistor but that is a waste of 110 watts. Can't the dealership flash them off? Worse case that's $100

[mikamars]

First off, I'd just like to say thanks for this great topic and for the work you've done! I'm sure I'm not the only glad Volvo owner that would like to buy you a beer. (www.yougotbeer.com)

I ordered the components last week and I'm hoping to get them together with my mediocre electronics skills in the next few days. I had a single cap before and very lousy ballasts, that didn't work out so well (that darn wiper motor).

Just a few questions: Where did you screw the ground wire from the caps? Those caps were hard to find, the best I got my hands on were "Aluminum Electrolytic 22000µF 35V 35x51mm Snap-in", hope these will do? How do I know how well they handle ripple?

[theshadow27]

Quote, originally posted by mikamars »

Just a few questions: Where did you screw the ground wire from the caps? Those caps were hard to find, the best I got my hands on were "Aluminum Electrolytic 22000µF 35V 35x51mm Snap-in", hope these will do? How do I know how well they handle ripple?

Mika -

Glad I could help

22mF/35v caps are better than the ones I picked out above (22mF/16v) and they will do just fine. I used the 16v because of size (they are smaller), and if the car's electrical system surges past 24v (what the 16v caps are rated for transients) there will be much bigger problems than broken HIDs.

In the S40 (same for V50) LHD, by the passenger (right side) footwell behind the carpet there are two ground leads that run to the CEM (under the glove box). They bolt directly into the frame, and I just stuck another crimp-terminal ring under the factory one and tightened it down. I have never taken apart a RHD car, but from the few pictures I've seen the CEM is still on the passenger (left) side, so I'd assume there would be two similar bolts on behind the carpet there too.

To get to them you have to remove the door jam (plastic piece that runs along the bottom boundary of the door) and the small cap/screw thing that holds the carpet in place. Pull it away from the body and follow the wires running down the side towards the floor pan. Somewhere in-between two black wires will branch off to the frame. Can't miss it.

I like that location because there's plenty of room back there to stick the box. That way it lives inside so there's less temperature fluctuation, and no moisture. With 1/2 meter of wire (4 conductors, NLT 16 AWG) you can hook directly into the CEM harness to tap the necessary wires.

As for ripple handling... The only way to know for sure is to hook an oscilloscope up at the ballast end. Aside from that you could try isolating a DMM with a small capacitor in series with the test leads and measuring AC voltage for a rough indication.

I wouldn't worry too much, just take a look at the maths:
http://hyperphysics.phy-astr.g...l?#c4
Although it's not a sinusoidal input (PWM is square wave, at anything > 70.7% duty cycle it's less bad than sine) that is still a useful calculation. A 55 watt HID ballast comes out to ~4 amps nominal at 13.5v. For that calculator you can approximate a 4 amp load at 13.5v with a 3.375Ω resistor (0.003375kΩ):

With the 4,400µF capacitor found in the DDM BOW2 there should be, in theory anyway, about 2.1v ripple, meaning that the ballast sees between 8.6v and 10.5v with an average output of 9.7v -- WAY below it's design spec.

With a 22,000µF cap, the output voltage averages 12.75v and fluctuates between 12.25v and 13.25v -- Much more reasonable ripple of 0.434v, with voltage that's within design spec for just about any ballast out there.

Good luck with the build

Jacob

[mikamars]

Thanks for your reply. Seems like I have this capacitor:
http://search.digikey.com/scri...71-ND
Should be efficient enough.

The oscilloscope I got my hands on is classroom integrated, so I can't really use it for ballast testing.

I made separate ground wires from the ballasts to the battery. This is just because I'm a bit wiper motor paranoid.

I'd also like to report that my build seems to be working very nicely with 35W ballasts (not that there should be problems with 35W). Mine are Canbus-ballasts.

Thanks again for the info!



Modified by mikamars at 8:29 AM 1-17-2010

[theshadow27]

Quote, originally posted by mikamars »

Thanks for your reply. Seems like I have this capacitor: http://search.digikey.com/scri...71-ND Should be efficient enough. The oscilloscope I got my hands on is classroom integrated, so I can't really use it for ballast testing. I made separate ground wires from the ballasts to the battery. This is just because I'm a bit wiper motor paranoid. I'd also like to report that my build seems to be working very nicely with 35W ballasts (not that there should be problems with 35W). Thanks again for the info!

Sure you can - there's a 12v supply in the lab, no?

Grounding the ballast is a very good idea, most people (including myself) are too lazy to do it but it's the only way to guarantee there will be no trouble with the wiper motor.

That's really great news about the 35w, another member tried this circuit with a 35w and could not get it to work. It should theoretically work just as well, the "Bulb Failure" message is triggered only on open circuit. I've added you to the list of working installs

[mikamars]

Quote, originally posted by theshadow27 »

Sure you can - there's a 12v supply in the lab, no? Grounding the ballast is a very good idea, most people (including myself) are too lazy to do it but it's the only way to guarantee there will be no trouble with the wiper motor. That's really great news about the 35w, another member tried this circuit with a 35w and could not get it to work. It should theoretically work just as well, the "Bulb Failure" message is triggered only on open circuit. I've added you to the list of working installs

Yes, I can simulate AC 12V 82Hz. I'm just not that good with oscilloscopes, hardly understand anything of it I suppose I have to contact my engineering friends... Can we assume the box is working good enough when my lights turn on and off nicely without anything blowing up?

My box will live under extreme conditions so I need to be thorough with the silicone. Our weather here in Finland is pretty much everything between -35C and +35C (-31 to 95 Fahrenheit) and fluctuating.

BTW do we have the same problem with high beam and fog light HID conversions?

[theshadow27]

If it works call it good - checking the 'scope is only to make you feel better.

It is good you got the industrial rated capacitors (-40c -> 105c ) - the standard ones (0c -> 85c) would throw a fit at -35 because part of their electrolyte is water. In any case, I would still recommend mounting it inside the car, temperature fluctuations tend to be smaller in there (and it warms up to room temp after it's started).

No problems with high-beams or fogs, both are hooked via relays directly to the battery with no bulb-out monitoring or PWM stupidity. That said, I haven't done either - I got HIR1 bulbs for my highs and I don't use my foglights unless it's foggy (when extra light is a bad thing).

HID high beams flash slowly (or used to, the new ones are a little better) and pulsed operation is horrible for the bulb. Special HID ballasts designed for strobing are available for aircraft - up to 85w) and but are expensive. If you used your halogen fog lights for flashing, and only used high beams on continuous duty that might work.

The fogs work fine, there are mixed opinions on 35w/55w, most say stick with 35w or the housing will melt. The primary problem with our fogs is the optics are terrible; adding more light doesn't do much good.

[mikamars]

I know this is not really for this topic, but where did you get HIR1 bulbs? Do they fit without any adapters? I have HB3 (9005) high beams. How did you get the color (for example 5000k) to match low beam color?

[theshadow27]

Thread is here: https://forums.swedespeed.com/z...50135
Got them from here http://store.candlepower.com/9011.html
They don't need an adapter, just a little adjustment to one of the keys, shown here http://dastern.torque.net/Mods/HIRmod.html
The color does not match, but I did not put in HIDs for the color. They are very, very bright. See here http://theeshadow.com/files/vo....html

[mikamars]

Quote, originally posted by theshadow27 »

The color does not match, but I did not put in HIDs for the color. They are very, very bright.

Nice. Those HIR1 are hard to find, suppose I'll order the same ones and try to make bi-xenons also work (E46). I'm not doing this stuff for fancy colors either.

[vinopa30]

Looking over the schematic and pics I'm confused about which side of the diode should be on the pos. terminal of the cap. I think the pictures conflict.

Looking at this pic it looks like the cathode side shares the positive terminal cap. and the out line (HID).

But looking at this pic it seems like the anode is on the positive terminal of the cap and the line in (CEM).

Red wire (CEM/IN) is soldered to the anode VIA this pic:

Trouble shooting says to make sure the cathode is on the cap side and closest to the out/HID.

So is this pic wrong? Should the "to cap positive" be off of the orange terminals (out/HID) instead of the red terminals (in/CEM)?

[theshadow27]

Quote, originally posted by vinopa30 »

Looking over the schematic and pics I'm confused about which side of the diode should be on the pos. terminal of the cap. I think the pictures conflict.

You're absolutely right! The I connected a red wire to an orange wire because that's what I had lying around, but didn't catch it when adding the paint colors. Thanks, the appropriate changes have been made. Great catch

[vinopa30]

Thanks for the confirmation!

I'm doing without the board, is there any reason not to do it this way?? Will the rectifiers overheat?
The rectifiers I have are double anode, 3 prong. I just clipped one of the anodes and wired to the other, any problem with doing that? (they are the ones you listed from mouser).

[theshadow27]

Well since it's done you might as well try and find out, however theoretically:

1) Rigidly mounting components without structural support (i.e. a PC board) places stress on the solder joints which will fail from vibration. How long they last is a function of the type of solder (silver bearing is stronger than 60/40 or lead-free) and the quality of the joint, as well as ancillary supports like epoxy. Silicone is flexible enough that it can't be counted as a support structure.

2) Assuming the rectifiers are rated at 10+ amps there is no danger of them overheating, regardless of configuration. You can calculate produced heat from the numbers given in the data sheet, but most likely you'll see at most a 5-6 degree C rise at the sink.

3) It's best to wire both anodes together, otherwise you sacrifice 1/2 of the load capability because you are using only one junction (out of two). I didn't realize I listed double junction rectifiers from mouser (I ordered from Digikey)- another great catch. It shouldn't affect the function of the circuit, but I've updated the BOM for the Mouser diode. The steady state current is about 4 amps, so either way you should be covered. If you do have extras, I'd recommend tying both anodes together.

4) Hopefully the circuit will get some casing after that photo, as well as strain relief on the wires. Automotive electronics are second only to aviation electronics (or maybe satellite electronics) in routine stress, the more mechanically robust a circuit is the better.

[theshadow27]

Looking through my server logs...

http://www.volvolandia.com/SVF...64.40

Google translate is not very good so I can't follow the entire thread, but if you guys can read this, 6.7mF is not enough! 15mF minimum and you really need 20-30mF to do a good job.

Edit: Mika, if you're still watching this, you can tell the fellow looking for EMI resistance that the problem is between the ballast and the bulb, not between the ECM and the ballast. The PWM in this case is far below radio frequency - 82hz would only cause problems if injected directly into an audio signal. The HV wires run several hundred khz, but in most setups there is shielded wire most of the way. If you are still worried the whole high-voltage wire may be wrapped in grounded aluminum foil. I've never heard of (modern) HID setups causing problematic interference.


Modified by theshadow27 at 6:54 PM 2-23-2010

[mikamars]

Nice job finding our Volvolandia!

If you read the thread further you'll find out that I've already tried to explain this and suggested 22mF caps. I've already suggested building your device in a few threads. I think we have already made a few devices with success.

What the fellow is trying to do is he's trying to make the wiper motor more "error" tolerant. Not really sure why he's doing this, I already explained it's not the problem, it's just the consequence. I think he's trying to add some protection to the wiper motor's network controller circuit.

We have a few more threads regarding HIDs in S40/V50, at least this one:
http://www.volvolandia.com/SVF...54.60

I'm surprised how many people actually haven't got any problems with the regular setup (just the HID kit).

[theshadow27]

Very cool! It's always exciting to see stuff go international.

Quote, originally posted by mikamars »

What the fellow is trying to do is he's trying to make the wiper motor more "error" tolerant. Not really sure why he's doing this, I already explained it's not the problem, it's just the consequence. I think he's trying to add some protection to the wiper motor's network controller circuit.

Hmm, seems a bit to me. You can't add any filter to the CAN network - it will stop it from working and probably fry the controllers. CAN is already very error tolerant because it uses two wires and neither is ground-referenced, they are only compared to each other. The problem is the ground connection on the wiper motor, not the network itself.

And overall you are 100% correct about cause-effect. I don't think the wiper motor has ever failed on an unmodified vehicle.

Quote, originally posted by mikamars »

I'm surprised how many people actually haven't got any problems with the regular setup (just the HID kit).

Me too. Perhaps some kits are better than others? I do know that even on cars where the HID worked without a BOWE, adding one made the HIDs brighter and turn on quicker. That makes me think that the arc is going out and re-igniting 80 times per second, which is bad in many ways, especially for EMI...

[mikamars]

Quote, originally posted by theshadow27 »

I don't think the wiper motor has ever failed on an unmodified vehicle.

Absolutely correct, according to my Volvo dealer, wiper motors are pretty much never replaced in Volvos.

Quote, originally posted by theshadow27 »

Me too. Perhaps some kits are better than others? I do know that even on cars where the HID worked without a BOWE, adding one made the HIDs brighter and turn on quicker. That makes me think that the arc is going out and re-igniting 80 times per second, which is bad in many ways, especially for EMI...

When I started my conversion process I had cheap and obviously very lousy ballasts, which broke my wiper motor in a matter of days. I do think my new & better ballasts would last much longer without any problems, even without your device. I'm not going to try though.

[mikamars]

It's good to watch out for low quality ballasts

http://www.youtube.com/watch?v=kn0B2lpzSlc

[theshadow27]

Quote, originally posted by mikamars »

It's good to watch out for low quality ballasts http://www.youtube.com/watch?v=kn0B2lpzSlc

That's pretty bad... I'm glad I don't have a motor in the steering column (it's for the self parking)!

That said, Lexus = Toyota = Insane computers these days

[mikamars]

Do you have any experience with this kit:

APEXCONE RAPTOR V2 HID KIT 55W - DDM TUNING:
http://www.ddmtuning.com/index...nt=85

Just wondering if these ballasts will work with your BOW?

Here are the specs:
https://ddmtuning.com/support/...v=0,1

Or which DDM kit would you recommend for our Volvos?

[vinopa30]

I bought a set, they work with the BOW I built from the instructions above. I also have the slim kit 50W on my 600rr, and the slim ballast seems to be a better ballast. It will also fit onto the fender rail under the fender better than the larger raptor ballast.

With the raptor HID set (55w) there is a slight twitching with one of the bulbs for a few seconds during warmup. It seems to be less and less with more time on the bulbs. If you have a few extra $$ I'd go for the slim ballasts.

The install with the Apexcone/DDM kits are pretty easy, I drilled a hole on the backplate of the light, threw some silicone around the rubber seal (on the wires) and connected the leads... took about 30 minutes total. I didn't have to alter the harness up front at all, just under the glove box for the BOW.

I've had DDM HIDs for about 3 yrs without any problems, this was my 5th kit.

Hope that helps!

[theshadow27]

Quote, originally posted by mikamars »

Or which DDM kit would you recommend for our Volvos?

As vinopa30 said, I'd stick with the slim ballasts. I have a pair myself and they work great:
http://www.ddmtuning.com/index...nt=60
(you have to pick 55w from the drop-down menu).

The primary advantage IMHO to the DDM slims is the price - so far they have outlasted "performance" halogen bulbs (Silverstars, US version suck) that I before, and they were around the same price. For that, I have no qualms buying a new pair if/when these burn out.

They are decent quality though, I did a comparison with some more expensive ballasts here:
http://theeshadow.com/files/volvo/hidcomp.html
and they weren't much different.

[theshadow27]

Quote, originally posted by vinopa30 »

I bought a set, they work with the BOW I built from the instructions above. ...I didn't have to alter the harness up front at all, just under the glove box for the BOW.

Out of curiosity, which car was this in? It aught to be in the list up top!

[vinopa30]

08 c30

[theshadow27]

Quote, originally posted by vinopa30 »

08 c30

Thanks

[vinopa30]

Guess I should update my info!

[mikamars]

Quote, originally posted by vinopa30 »

Hope that helps!

Yes, it does. Thank you guys for your replies! I guess I'll order a set of those slim ballasts. Then I can put my old ones to my bike and our scooter (55W might be too much for a scooter, lol).

The international shipping is like 40 dollars, but it's still cheaper than most 55W hid kits I've found.

[theshadow27]

With any luck, this project will be available as a COTS box in the not-too-distant-future. In the meantime, version 3 uses SMD diodes, 25v caps, and enclosures that are already the correct size:

BTW it looks like there's a lot of stuff there, but that's only enough to build 5. The reason for the extra bits is it's the same price to order 5 boxes as it is 10, and the same price to order 5 grommets as 100. Stupid price breaks at quantity