Only because I have little to zero knowledge of electrical wiring.

I find it simple because it can be installed as a harness and only splice into the +/- wiring for the pump. I built my double pumper in a similar way, but didn't tap into any of the factory wiring.

 

Thanks for the reply! This is actually a perfect lead in to talk about why I came up with proposal #2. For installers that do not want to splice into the ECU harness directly you can still indirectly use the signal from the ECU via a Zener diode placed between the normally closed, and the positive side of the relay's trigger. Under low voltage operation the Zener will not allow current to pass and thus will not trigger the relay. When the ECU triggers high voltage operation the zener will pass current and trigger the bypass relay. When the ECU goes back to low voltage the zener will cease to pass current and the relay will switch back. So to visualize this, the hobbs switch is using mechanical pressure to generate the signal that will trigger the relay. The Zener diode uses electrical "pressure" (voltage) to trigger the relay. The difference from an installers point is that a zener costs ~$1 is solid state, highly reliable, and can be installed directly at the relay. Contrast that with installing a hobbs switch and you can see why I believe this is a simpler method. Also a pretty significant reduction in cost, and parts count with an increase in reliability.

Want to know more?:

I am a novice with electronics, but I have been teaching myself and recently learned about the properties of a Zener Diode and a light bulb went off in my head. A zener diode works like a normal diode, in that it blocks the flow of current in one direction, but with one special quality. Once the voltage goes above the "zener voltage" of the diode it will allow current to flow. You can get different zener diodes with different "breakdown voltages" to fit your application. Zener diodes are by no means an exotic part, they exist in almost all electrical circuits in one form or another, and there are plenty of examples of them being used to trigger relays at different voltages.

Sorry for the wall of text, but I hope that explains my logic behind proposal #2

 

I did this mod over the weekend with the spoolinup hobbs switch and I am seeing similar results. I also have Walboro 455 with drilled siphon.

 

After thinking about this, i figured out how to make it even simpler and even more fail safe. Instead of cutting the factory feed into the fuelpump and running it through the bypass relay, you just need to splice the common from the relay into it. Thus you get the amps from the factory circuit plus the "bypass" (not really a bypass anymore, more like an augmentation). This only works for Prop. 1 using the ECU to trigger the relay, because the zener would get stuck conducting in prop 2.

 

IDK about the easiest, I'm going to utilize the option as seen in Post #66 and all I have to do is add one relay and it will work. The only thing missing in my pile of parts is a Diode / Resister to stop the back flow that can occur through the fuel pump resistor.

Or my car won't work and then I will eat my words.

 

updated diagram with P-channel mosfet and pull up resistor to protect ECU Analogue output, and reworked to run in parallel with OEM setup.

 

Should the ECU pin for above be 39 (Evo 8)?

 

yes you are right, that was my mistake. Thanks for pointing that out. I will update original post to correct and clarify.

 

Spent some more time looking at this and scratching my head (Mech E with many years out of practice), wouldn't you still run into this situation -

 

no the relay is triggered via the ecu which is isolated from the power circuit. To be more precise the coils in the relays that drive them are isolated from the switches themselves. If this was not the case then the stock system would also suffer from this problem.

 

another way to state it is that the relays are triggered by their coils. The ecu activates the coils by sending a low level signal (ground) to the negative side of the coil allowing it to conduct. The coils are already receiving the high side and are just waiting for that ground to activate. When the ECU stops sending the low level signal the coils cannot conduct anymore and the relay switches off. What im doing with the additional relay and the p-channel is a little different then that, im turning the ECU's low level signal into a high level signal (high side switching) but the functionality of the p-channel blocks current until it receives the low level signal from the ecu. If the ECU turns off the factory high voltage relay the pull up resistor will bring the p-channel gate back up to source voltage and "turn off" the mosfet. I could have potentially just wired the relay directly up to the ecu like in my original diagram, but i cannot get current specifications for the stock ecu and didnt want to chance damaging it. A mosfet is driven off voltage and not current, so there is no risk of damaging the ecu by adding one.

 

let me know if that makes sense or not, just because it sounds right in my head doesn't mean anyone else can follow me lol

 

Looks good biggie. Please build/test and then let us know how it performs in the real world .

 

Tracking all, drew it out to help understand it some more.



Do you have a recommended manufacturer/part number for the MOSFET?

Here is the diagram for #66 which has been shown to stay on.



Could we just add a singular resistor as mentioned and forgo the mosfet proposal, or would that be less efficient?

 

Im not too sure about the resistor. I could try messing around with it in a simulation to see what happens but i have my doubts. TBH i don't even think the mosfet is necessary, i think the ecu could handle the current of triggering both relays just fine, but since I don't have specifications for the ECU I would not be willing to risk frying my, or anyone elses ECU to test it out I went with what seemed safest.

As im sure you figured out when modeling #66 the problem is the relay is being triggered by the switched side of the high voltage relay, which is not isolated and once the bypass relay is activated, it creates a loop preventing it from being deactivated. In your Diagram you did of my circuit you are missing the +12v on the source of the mosfet and the 10k resistor is between the source and the gate to pull it up and prevent it from floating.

I haven't compiled a materials list just yet. Im still thinking about what kind of component packages I want to go with and how I want to implement the circuit. Plus there are some particulars with automotive stuff like voltage spikes that can hit ~60 volts. So sourcing parts that can tolerate this is important. I do plan on building this up though and I will update with components I select to get your guys feedback.