Long story short, I have a nano in my car reading a 5v sensor wire from the car's map through a analog pin, a4 to be exact. when car is off map is unpowered. I have a direct positive connection from the battery to a 5v regulator and then to 5v pin on nano. So, the nano is always connected to the positive terminal on battery whether the car is on or off. However the ground connection is triggered by a relay when the car is turned on.
I also have a 10k resistor connected between the analog pin and sensor. I though about using a diode, but that dropped voltage more than I was willing to accept.
Do I need to worry about anything here?
Also, I measured this value with a multimeter, I put the positive probe on the a4 pin and the neg. probe on the car's chassis. If I put the negative probe on the nano's ground I get -.84v.
I'm using this nano to read the map sensor and send a signal to a pwm board to control the duty cycle of a fuel pump. The relay doesn't just supply the ground for the nano but for the pwm board, fuel pump, 3 voltage regulators, triggers 2 other relays,etc. I heard putting a relay on the ground side is easier on a relay and I'm trying to make this as reliable as possible.
Here is a link if you want more details on the fuel setup and why I am using a speed controller, post #17 I have a basic picture of my device when I was almost done.
I would hate to cut all the wires soldered at the relay and redo it all, especially due to where it is currently mounted on the car.
Surely the simplest way to run things like this is from the ignition circuit which is already switched high-side
for you. Why would you need to monitor manifold pressure(*) or control fuel supply when ignition was off?
The 10k resistor in series should protect the analog pin whatever (a diode was never going to be right
BTW)
(*) I had to lookup what a MAP sensor was. Manifold Absoulute Pressure - which means its monitoring
a running engine.
Can I ask why you want to do this - the usual arrangement is for the pump to pump fuel and a regulator on the fuel rail sets the pressure , excess fuel being returned to the tank .
Some cars ( eg latest BMW) control the pump ( which is 3ph brushless ) to maintain the fuel rail pressure .
If you change the fuel pressure in response to the map then your fuel injection will be all wrong .
hammy:
Can I ask why you want to do this - the usual arrangement is for the pump to pump fuel and a regulator on the fuel rail sets the pressure , excess fuel being returned to the tank .
Some cars ( eg latest BMW) control the pump ( which is 3ph brushless ) to maintain the fuel rail pressure .
If you change the fuel pressure in response to the map then your fuel injection will be all wrong .
Well, if you look at the forum I linked it's goes into a little detail about this but his tests are pretty unreliable/questionable but I have personal experience with this first hand.
I have my fpr (fuel pressure regulator) set at 58psi which is what it was tuned at. I have a fuel pressure guage on my dash so I always see my fuel pressure. I tune my code to be as mild/aggressive as it needs to be so I don't ever drop fuel pressure throughout my throttle range.
Why do I want to do this? Whenever your pushing a huge amount of fuel for a high horsepower build (supercharged) and your just cruising around for an hour or more. All that recycled fuel picks up heat, especially from in-line fuel pumps and from the engine bay as it absorbs heat from there, long tube headers don't help either.
Eventually, my fuel pumps get so hot that they can almost burn you by touch. They also get louder as they get hotter, I had a fuel pump fail under warranty 8 months after installing it. After that, I built this speed controller and have been using it for years with no problem. I just redid the rear portion of my fuel system, moved the fuel pumps from the spare tire compartment to under the car and worked on the fuel pickup. While I did that I also redid the electrical wiring for the nano (speed controller) and ended up wiring ground to trigger it instead of power.
MarkT:
Surely the simplest way to run things like this is from the ignition circuit which is already switched high-side
for you. Why would you need to monitor manifold pressure(*) or control fuel supply when ignition was off?
The 10k resistor in series should protect the analog pin whatever (a diode was never going to be right
BTW)
(*) I had to lookup what a MAP sensor was. Manifold Absoulute Pressure - which means its monitoring
a running engine.
I don't monitor manifold pressure or control fuel supply with the ignition off.
I have the stock fuel pump relay trigger a larger relay which activates ground since my stock relay/wiring wouldn't be able to handle that much current from aftermarket fuel pumps.
My thinking with the diode comes from personal experience/theory.
My second fuel pump gets activated by a Hobbs pressure switch(when I hit 3 pounds of boost) through ground. So I wired up a led on my dash so I know when it activates by wiring up the led positive side to a key on dash fuse and the ground side to the Hobbs switch. Now, when I turn my car off, the led lights up. The led ground wire which is linked to the fuel pump ground whose positive is connected directly to the battery. So my thinking is that with the car off, the led somehow uses the key on fuse as a ground and gets enough power on the fuel pump ground to light up. If so putting a diode on the led would fix this by limiting current in one direction. Sounds like a similar situation could be going on with the analog and map since the nano is fed positive but no ground with key off, so maybe it's using the analog to map as a small ground, (but nothing gets powered). With a diode you could limit current to only flow from the map to nano. If there is a problem with my setup in this regard, then this would fix it as it prevents any possible ground from connecting with the nano setup.
If a 10k resistor should protect everything, and I got nothing to worry about for sure then all is good. If not then I have a dual relay board next to the nano in which only 1 relay is being used, I could link the analog wire through there and have the nano activate the relay when on. It would be much easier for me than cutting 6 wires and resoldering 8.
Edit: I get 11.8v when I connect the multimeter's positive probe to the analog pin and the ground probe to chassis which may be completely normal considering the nano is always supplied positive, but I find it a little strange that it's 11.8v instead of something closer to 5v.
Also, when I put the ground probe on the nano's ground pin and the positive probe on the analog pin, I get -.6v (key off) which suggests the analog wire is acting as a ground when unpowered.
But hey, I'm no expert and don't claim to be, In fact, I'm far from it, so I wouldn't be surprised if half my thinking here is completely flawed.
I have had a read of the forum , I think you should measure the temperature of the fuel in the tank , to see if it is warming up - ie check you really have a problem . The graph of temperature rise gives no indication of the amount of fuel or the setup , someone else states most of the heat is picked up in the Engine bay - so it all needs checking in a scientific manner .
Fuel running back to the tank through a long small diameter tube back to the tank will rapidly cool.
Most cars these days have the pump in the tank , so the fuel cools it .
A coil of fuel line around your pump might therefore be a solution.
I still think controlling on the fuel rail pressure ( which is what you want to control ) is the simplest control loop method , rather than involving the map sensor and a indirect method of holding pressure . It is what some manufacturers seem to be doing
hammy:
I have had a read of the forum , I think you should measure the temperature of the fuel in the tank , to see if it is warming up - ie check you really have a problem . The graph of temperature rise gives no indication of the amount of fuel or the setup , someone else states most of the heat is picked up in the Engine bay - so it all needs checking in a scientific manner .
Fuel running back to the tank through a long small diameter tube back to the tank will rapidly cool.
Most cars these days have the pump in the tank , so the fuel cools it .
A coil of fuel line around your pump might therefore be a solution.
I still think controlling on the fuel rail pressure ( which is what you want to control ) is the simplest control loop method , rather than involving the map sensor and a indirect method of holding pressure . It is what some manufacturers seem to be doing
I think I know what you are talking about but that link is a horrible example. supplying less voltage to a fuel pump than what it is rated for can damage the fuel pump, that's why I use pwm. https://fuelab.com/how-fuelabs-electronic-regulator-works/
This is a better but more expensive solution, I didn't want to pay that much personally. Looks like they monitor return pressure too which ain't a bad idea, controlling a fuel pump based on the rail pressure would be a horrible setup, when the controller makes adjustments due to a change in fuel pressure, there would be a big delay before the fuel pump would catch up since it's on the other side of the car. The idea is to keep fuel pressure steady at all times, even a couple of psi is too much.
I found the map to be the most reliable sensor based on my research. A good way to support this argument is that my car has a speed density tune, which means it uses the map sensor instead of a maf mass airflow sensor, it measures air pressure instead of quantity and uses that to control the fueling. Typically works better on a boosted car and provides a smoother drive than what it would with maf.
I have ran the controller for a while and my fuel pumps never feel hot anymore, they are quieter and they have to work less overall which means a longer life.
In no way am I suggesting that forum contains reliable tests, but there are a few posts by me discussing my personal experiences with this and the result I experienced.
A very interesting project and you’ve obviously looked into it in great depth . Good luck - don’t forget to report back when you get it working , it’s always interesting to see projects .
hammy:
A very interesting project and you’ve obviously looked into it in great depth . Good luck - don’t forget to report back when you get it working , it’s always interesting to see projects .
Thanks, but I think your still overlooking the detail that I have had this working for years. I just recently moved the relay from the positive side to the ground side and wanted to know if the new wiring configuration might damage something.
Edit: Maybe I'll post a basic video of it in action later, there is not that much to see though.
I measured the voltage of the map at idle, negative pressure (vacuum) and then I measured it again with the engine off, key on, 0 psi. I set up a linear map function so at idle/downshifting the pump duty is at 30% and at 0 psi(under enough throttle) the fuel pump duty is 100%. Anything in between scales. Then I have a pressure sensor trigger the 2nd fuel pump under boost at 3psi and up.
05silgto: I just recently moved the relay from the positive side to the ground side and wanted to know if the new wiring configuration might damage something.
I heard putting a relay on the ground side is easier on a relay and I'm trying to make this as reliable as possible.
What was the reasoning, how much current is being switched?
Thanks.. Tom...
TomGeorge:
What was the reasoning, how much current is being switched?
Thanks.. Tom...
TomGeorge:
What was the reasoning, how much current is being switched?
Thanks.. Tom...
No reason really, I was starting to get air in my fuel system and I didn't want my fuel pumps to sit in my spare tire compartment anymore. So I redid my fuel system in the rear, fixed the fuel pickup and moved the fuel pumps to under the car. My electric work was sort of rigged at that time due to experimenting so I redid all that too and decided to use ground to trigger everything instead of positive because I heard it was easier on the relay.
About 18A if I had to guess.
That helps make me understand all this a bit better, but does this still apply if the only device that isn't using ground as it's switch in this circuit is the map if it's only connected through the signal wire? My guess is yes.
Might be worth it to wire differently as a precaution. Since my car uses the map sensor in the tune, if the map sensor were to fail, the car wouldn't start at all. Would be horrible if this were to fail at the wrong time.
05silgto:
No reason really, I was starting to get air in my fuel system and I didn't want my fuel pumps to sit in my spare tire compartment anymore. So I redid my fuel system in the rear, fixed the fuel pickup and moved the fuel pumps to under the car. My electric work was sort of rigged at that time due to experimenting so I redid all that too and decided to use ground to trigger everything instead of positive because I heard it was easier on the relay.
About 18A if I had to guess.
Where did you hear that it was easier on the relay? You have a voltage being applied to the relay and a certain amout of current will flow. What happens onthe coil side will have no effect on the switch side.
windoze_killa:
Where did you hear that it was easier on the relay? You have a voltage being applied to the relay and a certain amout of current will flow. What happens onthe coil side will have no effect on the switch side.
I heard it on car forum when I was researching fuel system stuff years ago. Just to clarify, the relay is triggered by the stock fuel pump relay which is a positive wire, and then when the relay is triggered it links ground for the fuel system.
The rumor is you can have a smaller gauge wire on the ground than positive.
Personally, I imagine since that it is a circuit, the same amount of current would flow on the ground and positive side.
The rumor is you can have a smaller gauge wire on the ground than positive.
Hmm.. the current flowing through the positive feed wire and the gnd wire are the same, so wire gauged would best to be the same.
Unless one is much longer than the other. Then the longer may need to be heavier to minimise volt drop, but it would have to be significant.
Also that car website comment was probably made by a backyard autoelec.
Thanks for the info. Looks like I'm going to rewire this setup for piece of mind.
I have another more Arduino related question while we are on this subject if anyone can be kind enough to offer advice. I wanted to use a LCD so I can monitor my fuel pump speed and etc. However, I don't want to have to run wires through my car interior again. I have 3 ESP8266 wifi boards laying around Link.
Is it possible to wire one of these esp8266 boards to the nano and use another near the dash connected to a lcd screen so that I can monitor my fuel pump speed? This is the kind of lcd screen I already have laying around Link.
I understand It should be possible, but I could never get the code figured out or find a tutorial that was relevant enough to get me started.
