Maker Pro
Maker Pro

5v PWM signal to hold 12v relay closed (Automotive)

DHR

Jan 2, 2022
15
Joined
Jan 2, 2022
Messages
15
Hi Everyone,

I'm back already... I have another circuit that I'm working on for an engine swap. This one's pretty long... Sorry about that. :(

The Storey:
Most cars have a safety feature which shuts the fuel pump off in case of a crash or engine stall. On the original engine this was a physical switch located in the Vane Air Flow Meter. (AFM) As long as the engine is sucking in air, the AFM stays open and the switch stays closed. The moment the engine stalls, the AFM closes and the switch opens. The switch in the AFM supplies ground to the coil in the "Circuit Open Relay", (COR) which runs the fuel pump as long as it sees ground.

The vehicle the new engine came from used a Fuel Pump Control Unit to control the fuel pump. I don't have the FPU and it's expensive to buy... besides, I don't want to rewire the whole fuel pump circuit when I think it will be fairly simple to adapt the existing wiring.

Others have solved this by simply hardwiring the COR, thereby bypassing the safety feature... this is not an acceptable (nor intelligent) solution; if I get in a crash, I definitely want the fuel pump to shut off.

There is a 5v PWM signal that the Engine Control Module (ECU) sends to the Igniter Module to tell it when to fire the Ignition coil based on camshaft and crankshaft sensor input. If the engine stops turning, the ignition signal also stops. (It might not technically be PWM, but that's what I read. The signal remains high until it's time to fire the coil, at which point it momentarily pulls low before returning to a high state.)

The Circuit:
The circuit diagram is attached I'm sorry it's unconventional. I like to draw the components out to help me assemble the circuit correctly, and I don't know enough nor have the appropriate tools to produce conventional circuit diagrams. I'm (clearly) doing this all by hand... MS Paint, to be exact... yes, I am the last person alive who still uses MS paint.

To monitor the the ignition signal and provide voltage to the MOSFET gate, I have selected the ADM805LAN IC with watchdog timer. (datasheet attached) I did the math and the maximum pulse duration, based on idle RPM of the engine and the number if cylinders, is easily short enough that the timer will not timeout between pulses.

To supply power to the IC, I'm using an L78S05CV 12v 5v 2A common ground positive voltage regulator (datasheet attached) with a 0.33uf capacitor on the 12v side and a 0.1uf Capacitor on the 5v side. I've selected a much higher current regulator than necessary in hopes that I can use it reliably without a heatsink.

To provide ground to the relay, I'll use an IRF610 MOSFET N-CH 200V 9A. A 10k resistor will be used to pull the gate low when there is no voltage there. Again, with higher current handling than necessary for use without a heatsink. (An automotive relay coil apparently uses less than 250 milliamps.)

For flyback, I have selected a Schottky and Zenner combination of diodes. The schottky for low latency, and the Zenner to limit the current passed to avoid damage to the relay contacts upon field collapse. (so I've read...)

The Questions:
Well, most importantly, does it work? I came up with this by combining 3 separate circuit diagram fragments and again, I really don't know what I'm doing, so there's a strong possibility I'm missing something.

On the last circuit I designed, it was explained to me that a resistor should be placed on the signal input for protection; is that the case here? If necessary, what value resistor should I use?

Do I need a resistor to pull the the drain high when the MOSFET is off, or is that not a thing? Or perhaps, does the connection to power through the relay coil do that already?

The MOSFET has a diode built into it, but I don't see a path to complete the circuit thought the MOSFET, so I think I need the flyback diode; have I got that right?

I have added the Zenner diode to limit the flyback current in an effort to protect the relay contacts; has anyone heard of this and does it make sense? I don't understand this, as I dont think the relay contacts would not be exposed to that current, would they? Is this because the coil and one of the poles on the relay share a common connection and could therefore provide a path to complete the circuit through the relay switch?
 

Attachments

  • ADM690A_692A_802LM_805LM-878717.pdf
    333.6 KB · Views: 3
  • irf610.pdf
    153.4 KB · Views: 2
  • L78S05CV.pdf
    978.1 KB · Views: 0
  • FP Failsafe.jpg
    FP Failsafe.jpg
    82.8 KB · Views: 6
Last edited:

ivak245

Jun 11, 2021
107
Joined
Jun 11, 2021
Messages
107
I see you have worked out all the relevant signals and circuitry, but have you considered an easier system like a propane shut-off device? I have fitted many to older cars, they are easy to install and readily available.
 
  • Like
Reactions: DHR

DHR

Jan 2, 2022
15
Joined
Jan 2, 2022
Messages
15
I see you have worked out all the relevant signals and circuitry, but have you considered an easier system like a propane shut-off device? I have fitted many to older cars, they are easy to install and readily available.

Thank you very much for your reply.

No, i had not considered such a solution. I'm not sure if that would address all of my considerations, but I'm curious to know more about it, as it sounds interesting. Can you explain a little more?

My application is a return type fuel injection system with an in-tank fuel pump. The fuel pump runs at full power continuously, and excess fuel pressure/volume is bled off by the fuel pressure regulator back to the fuel tank through a return line, creating continuous circulation. In the event of a crash if the fuel line becomes severed, without a fuel cutoff mechanism, fuel will continue to pump and make an already potentially deadly situation more dangerous.

Other scenarios where the fuel pump would shut off are when engine stalls or I want to have the key in the run position, without running the engine. I realize that there is an accessory key position for this, but not all circuits are active in this case.

I have considered a few simpler solutions:

I considered using an oil pressure switch but the factory idle oil pressure spec for the engine is about 4.8 psi, which is not enough to activate an oil pressure switch reliably.

I thought about using a relay controlled by the factory oil pressure switch, which is designed for the engine and is used to turn on the oil pressure warning light when there is no oil pressure, but the factory oil pressure switch is normally on, so it's not conducive to a failsafe mechanism.

If my circuit doesn't work, my alternate solution is an acceleration sensor that tripps from impact. They can be had from an auto wrecker for pocket change complete with pigtail. This is by far the simplest and cheapest solution, however, I'm not a fan, because it s an off-road vehicle; the last thing I want is for my truck to die mysteriously a few seconds after a big bump. It also doesn't address engine stalls/key on, engine off state. In either case, I will add an automatically resetting pushbutton override that bypasses the fuel cutoff circuit so I can limp home if there is a failure. I already have everything I need for this. ) A push button and a relay.)

There is an existing solution similar to mine that works off of the tachometer signal, but it costs about $150, which I thought was too pricy, since with a bit of effort, I could probably build my own tailor made module and have an interesting project to do.

My circuit costs about $50 for enough parts to build the circuit 5 times over. (AliExpress and a lengthy wait) My friend did the same engine swap and bought an expensive, professionally made wiring harness adapter which hardwires the fuel pump on. The plan is to build two of these modules and share the cost with him. I have to say that a $25 module that's reasonably failsafe, wires in with just 4 wires, and restores correct functuonality is an attractive solution.
 

ivak245

Jun 11, 2021
107
Joined
Jun 11, 2021
Messages
107
The propane cut-out is basically a relay which requires a constant ignition pulse to enable a +12VDC output.
It has 4 terminals- +12v input, ignition pulse input ( connected to the "points" connection of the distributor), +12V output, and ground.
So if there is an ignition signal input, there is a 12v output, which can directly drive an electric fuel pump without an auxiliary relay.
These units have a few seconds "start up" output when initially supplied with 12 v, so that there is output on start up without ignition signal. If no ignition signal detected after a few seconds, it will turn the output off. Under normal operation as soon as the ignition signal is interrupted, the output is turned off. These are fairly small units, some are in the same housing as a standard automotive relay, with an LED to show output.
Taco-metricrelay.jpg
 
  • Like
Reactions: DHR

DHR

Jan 2, 2022
15
Joined
Jan 2, 2022
Messages
15
That's great thank you for showing me that! I've never seen this product before. If I wasn't so deep into this project already, I might have gone this route. I've already built the circuit, but before I plug my expensive ECU into it, I want to make sure that I'm at least not going to blow anything up. (which I'm still not sure about)

I'm also partial to not modifying the truck's wiring in any way... it's a 27 year old truck; messing with wiring looms can open a can of worms I really don't want to deal with... I already have my work cut out for me with the engine wiring... I bought an engine with a cut harness, an ECU with a cut harness, and the dash plugs pigtails for the recipient vehicle. (The Recipient is a '95 Toyota 4 Runner, the engine is from a '90 Lexus LS400, and the ECU is from a 1995 Lexus SC400.) I'm in the process of splicing the three elements together... fun times. My module, (assuming it functions) is convenient because it connects to wires I already have to work on. Basically, I'm building a 100% plug and play wiring harness.
 
Top