Filtering PWM on +12V circuit

[LipschitzWrath]

Hey guys, need some expertise here. I have some +12V lights powered off a +12V bosch relay. The old lights for the application used to be powered directly. The relay was installed and the coil is being controlled off the old wire that used to be hooked to the bulbs directly.

The system controlling it doesn't go straight on or off. When it is on, rather than turning off, it first begins to dim the bulb using PWM. Once the PWM reaches a nominal (and unknown) level, it shuts off completely. Obviously, the relay doesn't like this PWM at all. It makes a terrible buzzing noise during the entire PWM cycle as it opens and closes a billion times a second. I'm sure this isn't good for lifespan.

I'd like to get rid of the PWM signal, or at least overcome it so that I get the effect of a straight on or off. I have tried SSR's and they don't work. So I need some help. How do I do this?

I thought maybe a transistor as its saturation voltage is much lower than the pull in or hold in voltage of the relay coil. Would that work? I don't know how a transistor would respond to having a PWM signal applied to its base.

The other option is to filter it out with an R-C network right? Any ideas there?

Now for the bad news. I don't know the duty cycle, decay rate, or frequency of the PWM signal. I don't have an oscilloscope either. I have a DMM and am quite handy with it.

Looking for help here. Got chewed out at another forum for asking the question so I am posting it here.

 

[Grossel]

Do you have the posibility to put a wire directly to the on/off switch rather than using the original wire?

What car is it?

 

[LipschitzWrath]

Do you have the posibility to put a wire directly to the on/off switch rather than using the original wire?

What car is it?

2008 Jeep Grand Cherokee SRT8.

I thought exactly what you thought. Unfortunately, believe it or not, no such wire exists. Everything in this car is controlled by modules that are interconnected with CANBUS communication. There is no wire with a simple on-off signal for the high beams. I even wanted to tie into the high beam indicator on the dash. That's not even possible because the gauge cluster is actually a big module, called the Cabin Compartment Node (CCN).

The system is called smartbeam. Linky: https://www.wkjeeps.com/wk_smartbeam.htm

If you read, using the dimmer switch on the column stalk won't work because to engage the smartbeam system, you move the stalk into the high beam position and leave it there. It isn't physically tied to the high beams, it just tells the computer that you're requesting the high beams. In the case of the smartbeams, the computer decides when to turn the high beams on.

So, no, to my knowledge, there is no wire that carries a PWM-free high beam signal. I have the factory service manual with acsess to all wiring diagrams.

I really think this is a case where the PWM signal must be defeated. Short of cracking the CCN open and soldering a wire to the indicator bulb. I don't even know if that would work because it's probably all 5 volt in there.

 

[shortbus=]

Got chewed out at another forum for asking the question so I am posting it here.

You weren't chewed out! Only told that automotive circuits were against the TOS.

 

[JimB]

Have a look at the attachment.

I am assuming that the original circuit was as I have shown, and that the line which I have called Return connects to the vehicle battery negative terminal (eventually).

I suggest that you add a diode and a capacitor to your relay as shown in my suggested circuit.

The diode can be a 1N4007 although any of the 1N400x series will be OK.

The capacitor is open to a bit of experiment, I would try something around 10uF to start with.
If the relay still buzzes alarmingly, try increasing the capacitor to 100 or 200uF. The larger capacitance would not hurt, but there may be a bit of delay before the relay de-energises.

JimB


The original attachment to this post was lost in a server crash a couple of years ago.
I am now attaching what I hope is a correct re-creation of that original attachment.
JimB 14/4/16

 

[LipschitzWrath]

Have a look at the attachment.

I am assuming that the original circuit was as I have shown, and that the line which I have called Return connects to the vehicle battery negative terminal (eventually).

I suggest that you add a diode and a capacitor to your relay as shown in my suggested circuit.

The diode can be a 1N4007 although any of the 1N400x series will be OK.

The capacitor is open to a bit of experiment, I would try something around 10uF to start with.
If the relay still buzzes alarmingly, try increasing the capacitor to 100 or 200uF. The larger capacitance would not hurt, but there may be a bit of delay before the relay de-energises.

JimB

Wow jim thx a ton. Ill give that a shot. I have some 1N4007 diodes laying around. Just need to get some caps. Thx again!

 

[LipschitzWrath]

You weren't chewed out! Only told that automotive circuits were against the TOS.

Wow thx for ur excellent contribution. Then I reposted with no allusions to automotive and they closed that thread too. The whole thing is one big cya, which I addressed by reposting and they still closed it. If its just for liability then my repost was sufficient. Thanks for thread crapping here.

 

[shortbus=]

Your welcome