head and park lghts on with momentary sw

Hello all, I have read a lot of great information on this site that has helped me out in some projects I have done in the past. This is my first post and I have a question that involves some relay expertise. So here goes: Objective: Get rid of push pull head light switch in custom 84 chevy c10 truck. Use momentary on push button that matches rest of my switches for park and headlight power. Press once for park, press again for park and head lights, and press again for all off. I imagine the output for one relay will be parking lights and another one will supply power for the dimmer switch to control the high and low beams. Current status of truck: Truck is getting full re-wire. Headlight dimmer will be on floor which is standard gm dimmer. Dimmer out wiring harness has low and high beam wiring, so I just need to supply 12v to the dimmer. Question: Would this be possible to use a momentary on push button switch to a relay setup to control head and parking lights off one button? In my mind I can see one relay that can have two switch points, trigger input with 12v and it switches another relay on, and trigger 12v again and it switches another relay on while keeping the first relay on, trigger 12v for the third time and it cuts power to the main relay to turn off the other two relays. Anyway that's what I am trying to see if is possible. Any help would be greatly appreciated! #!/Jay

What about using a 4017 integrated circuit as a divide by 3 counter (output #0 - no connection; #1 = parkers; #2 = beams)? Its output would have to drive a transistor or MOSFET to drive the relays.    I have described similar applications in other threads. Otherwise Hotwaterwizard (and others) typically has suitable relay equivalents.

Thanks for the reply! Hmm, that is interesting.. It has been about 12 years since I was in electronics class, so bear with me here. So what your saying is you have the IC and that has 12v going into it, use the 12v momentary switch for the clock, I would use the outputs to trigger the 2 MOSFETS that would control the relay? I would want output 1 to be park output 2 to be park and headlights. To get the park and head lights to be on at the same time, I think there would have to be some diodes in there somewhere to prevent back feeding while the headlights are on to output 1 for the parking lights only. So one press would trigger output 1, another press would trigger output two, and another press would turn on output 0 which would be null. Am I picking up what your putting down?

I think I found a good thread searching for divide by 3. Thanks again for your reply! I just might have to dust off my breadboard I see the reset logic on that IC too, so that is the ticket to my conundrum.

Yep - bingo! I did have a diagram that matches your description (probably in my still unrecovered crashed data) but I reckon the 2nd pic on Scott Stites' sm_2010_3phase is perfect. Phase 1, 2, & 3 represent off, park, & beam respectively. You'd need to add: - power supply filtering & protection for the 4017 itself (resistor, capacitor, and preferably a 12-14V zener diode to clamp below 15V); - switch de-bounce circuity (probably just a resistor & capacitor); - output tranny's or MOSFETs which could ground the coils (#85) of raw +12V powered relay coils (#86 to raw +12V; not from the "regulated" 4017 supply). Then there may be other additions like spike protection diodes across each coil, and maybe input protection diodes... And a POR (Power-on Reset) to ensure it powers up with lights off. And yes, 2 diodes to "OR" or join the #1 & #2 outputs to the parker transistor/FET else relay. One suggestion - design the relays so that you can manually energise them on in case of circuit failure - ie, reconnect #85 from the circuit to GND. (That's assuming you ground-switch the relays.) That ability is always good to have anyhow. I once used it when the power lead to my light switch got pulled off (and the HU was in the way...) Of course, bridging or joining the #30 & #87 pins or wires also works and is the solution for a failed relay (but we all carry a spare relay don't we?). And FYI re the 4017... It needs its Clock Enable to be whatever for counting - ah, I see it needs to be GND from Scott's diagram.    The 4017 resets on a +ve going pulse to its Reset pin. Hence, using the output 0, 1, 2, ... 8, 9 nomenclature or convention, a "divide by n" counter has output #n connected to the Reset pin. Hence for your divide by 3, output 3 goes to Reset as per Scott's diagram. Hence with POR, initially #0 is on. (But that is NC = not connected.) First pulse clocks to #1, hence parkers on. Second pulse clocks to #2, hence beam on, and - via the diode - parkers on (with the other diode preventing $2 backfeeding into #1). Third pulse clocks to #3, hence changing #3 from 0 (off) to 1 (on, or 12V etc) which is a +ve going pulse hence resetting the 4017 back to #0 and lights off. Download a (good) 4017 datasheet and you'll see an excellent timing diagram, though I suggest a ruler & pencil to draw light vertical lines down from the mid-point of the clock edges (slopes) on certain key clock signals to graphically understand what I described above. And since you are now a 4017 expert, you too will probably use the accepted digital convention where the "first output" is output #0. (Just as we do for our centuries and baby years.) You might cross some that label the 4017's first output as #1 (hence #1 - #10 instead of #0 - #9) and that can really confuse things - including my "n to Reset" for a div-n counter. And of course you know that output #n is not the same as pin #n. (And PICAXE "legs" and pins has nothing to do with this thread!)    Reread this reply if it's confusing. My expression may not be good, but as a building block, the 4017's operation is quite simple. (It's the powering and interface crap that requires know-how.) Now if I could find the "complete" diagram I know I had... PS - I see I missed your last reply, but hopefully we concur, and I may have provided the extra automotive electrical considerations.

Thanks, I really appreciate the info! I will have to try to get some sleep to digest this and I tried a circuit diagram software that has me a bit frazzled right now. Below is the best I could do at midnight and with the program I was using. I see you had other circuits in there for power filter and that, so I will need to digest that a little bit more and find software for circuit designs in the AM..   

Impressive! Is that a free circuit-CAD you are using? It looks familiar. (Eagle? Or was that PCB only?) I recently used TinyCAD but have issues finding the symbols that I want. I used to use Windows Draw (?) on my pre-crash win2k PC, but Win7 seems only to have Paint. Pity - I got quite quick a mocking up circuits for the12volt etc, but Paint... yuk! But I always liked the rubber-banding etc of real CAD packages. In the dim dark ages of the 1980s I used Racal's Redac circuit drawer for transfer to Cadstar for the PCB design. Anyhow, you're on the right track (that's a natural pun!). The first diode from Q2 pin-4 is not necessary, but that's not a big deal. (It's better left in if in doubt.) The protection (power) diodes need to be 1N4004 etc but the those "signal" diodes can be smaller 1N4148/1N914 etc, though a "one for all" 1N4004 or 1N4007 solution is fine. Probably NPN transistors rated at (say) 500mA or better if ground switching is used for the relays, then whatever base resistor to limit base current but ensure adequate collector to emitter current for the relay coil. Typical coils are usually ~60 Ohm or higher, hence ~250mA or less. Hence "my" design for 500mA. [ Let's see, if Q-out sources 0.5mA, that means a tranny gain of 1,000 for 500mA. That means Darlingtons else a tighter design - I'm assuming worst case etc. And I'm crap at tranny designs etc - I now tend to use MOSFETs where only a voltage above ~5V) is required to turn fully on (the gate = base current is mere nano-Amps) and gain is irrelevant. And with 20A to 100A MOSFETs costing a mere $2 - $3 with no loading on the control circuitry... Tthe 4017 outputs pull active low when "off" so no pull-down resistor is needed for the tranny or FET to ensure they turn off. Of course you've been in an electronics class much later than I so you probably know all that stuff, else have to scratch a bit less than I do. (I still chew my fingernails which - like crashed PCs - doesn't help my scratching for data.) ] If I find my SAAB hi/lo headlight relay I'll see what protection they use (raw +12V regulation, switch debounce etc). It uses a 4017 even though it's only a div-2 counter (0 is lowbeam, 1 is highbeam) - most would probably use a simple (T-type) flip-flop etc. But again, the 4017 might be a standard building block they use for other functions. But the 4017 is a good device to know about for counters, LED or any sequencer etc. I recall using it to make an electronic Halda (an accurate rally odometer whose output was in units of 10 meters or yards) back when such a circuit was considered "impossible" (LOL!) - and that was before I attended any electronics classes!    And yes, the other key component was the good old 555 timer!

I had a quick look at the SAAB headlight relay (red, p/n 95 66 647). It seems its 4017 is merely powered from +12V thru a 1k resistor to a 100uF cap and zener diode (to ground). Presumably the zener is ~12V or 10V etc. 4017 pins 16 & 14 (clock) go to that zener'd +ve rail. I thought I was misreading the PCB because the 1k resistor is rather large. But even using the 4017's worst case quiescent current draw (at 85C & 15V) of 600uA, that's only a 0.6V drop. (But then add the load of any transistor base currents etc.)    As usual, the "loads" it is controlling will be off another rail - eg, the raw +12V input. (Hence coil currents have no impact on the above resistor-zener voltage regulator.) Later I'll check their power on reset circuitry, and how they are clocking it (presumably via Clock Enable or similar I may try a full decode of the PCB. I'm a bit curious - there seem to be more components (resistors & diodes) than I thought necessary. (But if they ain't clocking via the CLK input...?) Update 48 hours later... Still not fully decoded, but I reckon (as usual) automotive company electronic designers leave a bit to be desired. Instead of using the expected switch to CLocK and output 0 to low beam, they have the switch to the ClockEnable through an inverting transistor, and 0 is No Connection. Output 1 toggles the relays through some relay & transistor interactive switching. Output 2 as expected goes through a diode to Reset which has the POR circuitry (a resistor & capacitor to force reset high when power is first applied). 11 resistors, 8 diodes, 4 transistors, 3 caps. I reckon I prefer my 4-6 resistor, 3-4 diodes, 2 transistors, 3 caps version with 2 plain SPST relays; output 0 being for low beam and output 1 for high.

I'm new to the forum, and on my first search I found this post. I am looking to do the exact same thing for my street rod project. I would like to have one button to toggle my headlights between off/park/headlight. I understand basic wiring and am learning more on relay theory and applications, but electronics and circuit design for this switch may not be for me.

It was mentioned in one of the replys that there may be some other types of pushbutton headlight switching options using relays only. Anyone have this information?

Thanks,

Steve

'41 Willys Coupe