Any electrical engineers or savvy guys? Need to design a circuit or something

[DionR]

The processor would activate MOSFETs that would activate either the actual load or a relay. You now have the ability to use logic and programmable thresholds for control.

I see now. It's similar (if not the same) as an Arduino. I've messed with those some, but I haven't played with triggering MOSFETs, really just an UNO with a CAN shield so I could read the network.

My initial plan had been a board running logic that would watch for inputs on pins and put messages on the CAN bus as required. But this hit a roadblock when I figured out the protocol for the CAN B network required a different chip than the CAN C network. I planned to dig into building my own board on one of those online sites with both chips and such, but then realized that maybe I could let some of the modules do that work for me if I could just create some circuits with resistors to mimic the inputs. So, in this case, if I can avoid building a board with some logic, seems like that would be simpler. That is unless I can't get resistors that do the job.

I will need logic when I convert the early CC stalk inputs into button presses the PCM recognizes. But that's another project.

 

[DionR]

Ok, I think you want to use your cars original headlight switch to replace the 06-10 switch.

Exactly.

The other four selections (off,park,head,auto,fog) all output on pin 3. The internal FOG circuit is in parallel to the rotary switch circuit and reduces the resistance in each position so you actually have six different ohms values depending on what positions are selected.

FOG: For FOG, you need an off-on switch and resistor. Using the value of any position with and without FOG to determine what the value of FOG is using the resistance in parallel formula. With FOG off, you know the rotary ohm values already.

OFF: The original switch probably doesn't have any contact made pushed in/off so you'd need a jumper with a resistor to provide power to the cluster when running that can be in/on the cluster and nowhere near the original switch. The ohms value is not lowest in the OFF position so the other positions will be parallel paths to the cluster to reduce the ohms as needed for the position. The lowest ohms will be in HEAD, then PARK, then OFF, with the highest in AUTO. Need another on-off switch for AUTO that is powered only when the original switch is off. This can be done with a 5 pin relay.

Makes sense. I don't have plans for fog lights nor auto, but you never know.

I will sketch up a circuit and see if that matches what you are describing.

Thanks!

 

[DionR]

Would something like this work? Or would the circuit with the 910 ohm resistor throw off the voltage when the park or on switch is thrown since it is still active?

If I understand series resistors correctly, with no switches closed, they are cumulative so the output would be the needed 16930 ohms. What I'm not certain of is if, for example, the park switch is closed, would the result be the needed 16020 ohms, or would there still be some voltage fed from the 910 ohm leg?

 

[MV8]

The math is more complicated for resistors in parallel but they add in series. You can use potentiometers to help to get the right values. Cheap on amazon.

 

[AUSA66]

So you are using the switch to control the stock headlights?

I expect I will be cracking one open at some point, but I would need to get another just in case I destroy this one getting it open.

Yes ,still use the footswitch for high/low beam (4 Headlights)

 

[MadScientistMat]

Would something like this work? Or would the circuit with the 910 ohm resistor throw off the voltage when the park or on switch is thrown since it is still active?

View attachment 1716497373

If I understand series resistors correctly, with no switches closed, they are cumulative so the output would be the needed 16930 ohms. What I'm not certain of is if, for example, the park switch is closed, would the result be the needed 16020 ohms, or would there still be some voltage fed from the 910 ohm leg?

Yes. That's exactly what you are looking for. Activating the park switch subtracts the 910 ohm resistor, and activating the headlight switch subtracts both the 470 and 910.

 

[DionR]

Yes. That's exactly what you are looking for. Activating the park switch subtracts the 910 ohm resistor, and activating the headlight switch subtracts both the 470 and 910.

Annoyingly, I got another Charger switch this weekend to deconstruct and tested it first and got different resistance values. So I test my original and got the same values as the new one and still not the original values. No idea what I am doing (or did) wrong, and frankly it is pretty frustrating.

 

[DionR]

i Cheated on my road runner opened up the switch cut a few tracks & used Kynar wire to run the contacts directly to the terminals then used solid state relays to switch the lights position 1. Park & side runners,2 & 3 headlights if i push the button it flashes the headlights
I don't need dimmer with the Digital dash

View attachment 1716497136

Any pictures of your mods?

Thinking about doing something similar, except closing the circuits using my original switch rather than the other way around. Probably still going to need some relays too since the circuits in the A-Body switch don't match up well with the Charger switch.

 

[DionR]

I've given up on why my numbers are off.

Instead I think I am going to wire into the circuit board and use some relays, unless someone has a better idea.

Here is the backside of the board with the passthrough rivets marked:

Here is the front:

My idea is to bypass the switch (grease covered square) and use some relays and the park and on contacts in the stock A-Body switch to connect the 5V input to the appropriate circuit.

The stock switch would have 2 wires from the relays going into the respective 12V input connections and rather than going to the park lights or headlights would instead just go to ground.

I don't have an idea yet what these relays would look like, where they would go or how it all works beyond the above idea.

Thoughts?

 

[DionR]

Oh, and I did confirm that the 3 contacts appear to be separate circuits, it's not something where each is adding to the last or something.

 

[AUSA66]

Bit different to mine that looks easier where you have pinouts on the switch the passthrough rivets are called vias i knocked a lot of the resistors off just left the 1's for the LED lights

 

[DionR]

This has been fun, in a sort of painful way. Trying to learn and figure out how to do this best.

Spent some time looking at transistors and MOSFET's. I like the depreciated MOSFET's for the "off" circuit but the sketching software I am using (LibrePCB) doesn't give an option. Then I noticed that there appears to be some voltage drop across the source and drain (if I have those names right), which in this case seems like it would be a problem since I need a specific voltage output based on the existing resistors. So I dropped back and punted and went back to relays.

Drew this up using 5V relays:

Even did a PCB design and got a quote from PCBWay. Looks like 5 is the minimum, but even with one being fully assembled it was only $34.

Haven't order anything yet. Figured I would sit on it for a bit, plus maybe someone will see an issue with the idea or layout.

 

[mpgmike]

Typical MOSFETs will have a Source - Drain ON resistance of a mere fraction of an ohm. This should not affect your voltage divider readings.

 

[mopardude318]

I can’t even remotely understand any of this. It hurts my head. Kudos to y’all electrical gurus. This is genius thinking. Not for me.

 

[DionR]

Typical MOSFETs will have a Source - Drain ON resistance of a mere fraction of an ohm. This should not affect your voltage divider readings.

I think the one I was looking at was like 0.06 ohms. I guess not much, but the complexity of trying to get the gate voltage right and figure out to use them to duplicate the relays where it isn’t just on or off made the relays look so much more attractive when I saw they added voltage drop, too. Maybe I will keep looking at it.

 

[DionR]

I can’t even remotely understand any of this. It hurts my head. Kudos to y’all electrical gurus. This is genius thinking. Not for me.

Believe me, it makes my head hurt, too.

 

[MV8]

It appeared to me that you were working this out with Mat using private messaging so I stopped following. I stand by my suggestion.

It doesn't matter what values are found with different used parts, just what your system needs to switch between the modes. You don't need an original switch to determine that; just a potentiometer and some resistors to get the pot in range to trip the system.

I suggest pretending this is 1985 and just go with what you know/send it. The paralysis of excess analysis.