Announcement

Collapse
No announcement yet.

PWM to discrete on\off

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • PWM to discrete on\off

    Hello,

    So here is my problem; I want to power some accessories of the interior lights in my truck. The problem is that the interior lights donít just turn on and off...they turn on normally but when they turn off they fade out (dim). I suspect (but am not sure) that the lights are actually powered by a PWM signal (rather than a simple on\off) and when going out the BCM simply modifies the duty cycle of the PWM signal to affect the fade to black. This means that when I connect my accessories through a relay the relay simply starts buzzing when the lights are going out.

    So here is the question...How do I create a simple discrete on\off from the PWM signal?

    Thanks,
    Steve

  • #2
    a diode and a capacitor should do the trick-- the capacitor will 'fill in the gaps' of the pwm signal, and the diode would be installed before the cap to make the cap only feed into the relay.
    My OLD 2001 Mitsubishi Eclipse GT:
    "The Project That Never Ended, until it did"


    next project? subaru brz
    carpc undecided

    Comment


    • #3
      Wouldnt there still be a chattering issue as the PWM signal fell through the treshold voltage of the relays coil?

      Comment


      • #4
        with a large enough cap, i don't think it should..
        My OLD 2001 Mitsubishi Eclipse GT:
        "The Project That Never Ended, until it did"


        next project? subaru brz
        carpc undecided

        Comment


        • #5
          Originally posted by sredmyer;1480517I want to power some accessories of[I
          f[/I] the interior lights...
          Maybe if you explain what it is...
          It is rare for "accessories" to only be powered whilst the dome is on and dimming.


          Furthermore, the dimmer might be analog.
          But you'd be better off using a transistor as a big cap could blow the PWM, and an analog dimmer will reduce the coil's voltage.

          Usually accessories (other than puddle lights etc) are triggered by the door switches etc that control the dome lights. (Hence it doesn't matter if the dome light is blown or manually turned off).

          Comment


          • #6
            Actually, the "accessories" are basicly puddle lamps. What I am trying to do is power the lights on the ends of my running boards. Powering them directly from the dome light wires apparently puts to much load on the PWM circuitry beacuse the lights on the running boards and the dome lights quit working at all when the running board lights are directly connected.

            Thanks,
            Steve

            BTW, the vehicle I am working on is a 2011 Ford Super Duty truck. The running boards are factory OEM boards that came from a different truck. The wire I am trying to use to power the board lights is the wire the Ford wiring diagram says is used to power them when they are factory installed. As to why it doesn't work, all I can figure is that the circuitry is beefier in trucks that actually had these boards from the factory??
            Last edited by sredmyer; 08-12-2012, 09:48 AM.

            Comment


            • #7
              It's more likely that the running board version uses a different PWM, though most (Japanese) PWMs have a 10A capacity.


              But if it is PWM, and if the puddle/running-boards are to imitate the dome, then it should simply be a matter of inserting a buffer circuit - eg, a MOSFET, else transistor.

              If the PWM is 0V (GND) when off and PWMs +12V, then an N-Channel MOSFET can be used to ground PWM the puddles (ie, puddles to +12V & their GND end to the MOSFET).

              Else if some of the lamps are converted to LEDs, the original PWM might work.

              Comment


              • #8
                Although I can figure out some simple electronic stuff, when you start talking about MOSFET and transistors I get lost pretty quickly. Can you be more descriptive about the circuit you are envisioning?

                Thanks,
                Steve

                Comment


                • #9
                  Another question, how can I tell whether the dimming is signal is an analog one or is in fact a PWM? When I measure the voltage, I measure ~12 volts when the door is opened. When the door is closed and after the delay (about 30 seconds) the voltage drops to ~8.5. The voltage stays there for 9 minutes and 30 seconds at which point it drops to ~0. The drop from 12 volts to 8.5 is somewhat gradual causing the dimming effect. However the drop from 8.5 to 0 is nearly instant. Having said all of that, how can I know whether the voltage is droping via analog control or if it is just an effect of my RMS meter reading a PWM signal with a decreasing duty cycle?

                  Thanks for the help,
                  Steve

                  Comment


                  • #10
                    You could try reading it with your meter on "AC".
                    Palm sized ainol MiniPC, 8" Transreflective PRO, Win10, Reverse camera, Dual 10HZ GPS RX's for Speed Display & Sat Nav, FM-DAB & Phone Modules, iDrive interface. T-Screen HVAC control, custom microcontrollers, microcode and FE.

                    Comment


                    • #11
                      Led

                      I would attack this problem from another angle; Simply change out all the bulbs in your runningboards and puddle lights with LEDs, then you can attach your lights directly to your dome lights. If there's many lights and you're still in doubt, then change all your dome lights with LEDs as well then the load should be lower than it was original. I did a complete LED conversion on my 2001 Tahoe, it uses less energy and gives more light output and looks better than stock

                      Comment


                      • #12
                        A good suggestion.
                        Though I tried a "led replacement" for my dome years ago (what a joke!), the newer LED is great. MUCH higher & better output. It seem to illuminate much more - maybe its much whiter light - and it isn't wrecking its holder & cover like the old 5W bulb's heat did. And its much brighter than a 10W bulb.
                        IMO no doubt - LEDs are now the superior lighting method.

                        Using LEDs with or instead of the dome etc bulbs assumes PWM dimming if linear dimming is desired.


                        Originally posted by sredmyer View Post
                        Another question, how can I tell whether the dimming is signal is an analog one or is in fact a PWM?
                        The easiest might be to parallel a LED (string) with the dome.
                        If the LEDs dim linearly, it's PWM - ie, LEDs should dim fairly proportionately to the bulbs.
                        If analog, the LEDs will tend to dim slowly from full bright before a sudden dive to off (probably when bulbs are still ~1/2 bright).

                        Standard DMM voltage ranges cannot determine PWM signals, though some DMMs have a frequency setting that might be useful (assuming you aren't measuring noise - but PWM frequencies are constant except for rare situations).

                        A True-RMS voltmeter should detect PWM. With a load (resistor, LED, bulb) the RMS voltage should drop. The RMS voltage (like the current) should reduce linearly (in proportion) with the PWM duty cycle.

                        Ironically (pun), old moving-coil voltmeters meters should also indicate PWM as their moving iron mechanism (get the pun now?) measures the average voltage & current.
                        Though DMMs are supposed to measure average values, they actually measure a peak value and scale from that - ie, x1 for (ripple free) DC, and x0.636 for AC. (ie - "sample & hold" circuits: sample, hold, then convert to digital; then re-sample & hold etc.)




                        FYI - For a sinusoid, the average voltage = 0.636 times the peak voltage. Same for current. Hence incorrect readings for non-sinusoid waveforms. (Big impact on current readings for SMPS systems!)

                        BTW - signal frequency can have an effect. But True-RMS meters should have a wide bandwidth - ie, measure from DC to many Hertz (100kHz? MHz?) so that all sinusoids are measured. (Any wave shape can be represented by a series of sinewaves at different frequencies. That includes square waves, triangular waves, music waves, though I'm unsure about Mexican waves.) I assume True-RWM meters (still?) use DSP techiques (Digital Signal Processing).

                        Comment


                        • #13
                          Originally posted by sredmyer View Post
                          Another question, how can I tell whether the dimming is signal is an analog one or is in fact a PWM? Steve
                          The simple way to find out if a signal is PWM or DC is to measure it with the meter on AC range.

                          A DC voltage will NOT show on the AC range.

                          A low frequency PWM (vehicle light dimming) should give a good reading on AC, especially on a DVM.
                          Palm sized ainol MiniPC, 8" Transreflective PRO, Win10, Reverse camera, Dual 10HZ GPS RX's for Speed Display & Sat Nav, FM-DAB & Phone Modules, iDrive interface. T-Screen HVAC control, custom microcontrollers, microcode and FE.

                          Comment

                          Working...
                          X