Can Bus flex question

[GL95]

I have an ESS G3 and am currently running a can bus flex fuel tune that is sensor based through PCMTec. My flex fuel sensor is located in the return feed of an Arcane stage 3 fuel system with TI274 pumps and no check valve. The issue I am running into is that since there is no check valve in the fuel system outside of the pumps, my fuel lines are draining the fuel back into the tank and I am not able to get an ethanol reading between starts. I have 3 questions:

1. Is it possible to put check valves in both the feed and return to keep pressure? I'm not sure how to install a check valve in the return since obviously I want fuel to go back in the tank while the car is running.

2. Are there issues with a return side mounted ethanol sensor at WOT?

3. If 1 & 2 no good then what do

Sponsored

 

[mejohn50]

I am not running the ethanol sensor in a return line, but I have been playing with the PCMTEC flex fuel sensor implementation for a while, so maybe I can offer some help.

1. I can't answer this question, but it sounds like it'd be counter productive to put a check valve in the return line.

2. I am not aware of any major issues with the ethanol sensor being mounted on the return side, but I do know PCMTEC set the software up expecting the sensor to be on the feed side. The main thing they did was program a lag between the sensor reading and the ability for the PCM to make changes based on that reading when ethanol is increasing. This allows the fuel time to get from the sensor to the injectors before changes are made in the PCM based on the ethanol content. When ethanol content is decreasing, the PCM acts more or less instantly based on sensor data. You could, theoretically, put yourself in a situation where you are switching from ethanol to gasoline and you have gasoline in the fuel rails but still have ethanol in the sensor since it's after the rails. It would be a brief period of time, but it could happen and you'd be on E85 timing values running pump gas.

3. Personally, I would mount the sensor in the feed side of the system. There are sensor adapters that can feed plenty of fuel so it shouldn't be a limitation.

One question I do have though...is there an issue with not having an ethanol reading when starting the car? The system is programmed to use the last known ethanol content if there is an issue with the sensor. If you parked the car with E85 in the fuel system, when you go to start it again it *should* still use E85 values until the sensor returns data. Maybe this specific situation has some quirks I am unaware of.

Finally, @Rolls might be able to share a little more info.

 

[GL95]

I am not running the ethanol sensor in a return line, but I have been playing with the PCMTEC flex fuel sensor implementation for a while, so maybe I can offer some help.

1. I can't answer this question, but it sounds like it'd be counter productive to put a check valve in the return line.

2. I am not aware of any major issues with the ethanol sensor being mounted on the return side, but I do know PCMTEC set the software up expecting the sensor to be on the feed side. The main thing they did was program a lag between the sensor reading and the ability for the PCM to make changes based on that reading when ethanol is increasing. This allows the fuel time to get from the sensor to the injectors before changes are made in the PCM based on the ethanol content. When ethanol content is decreasing, the PCM acts more or less instantly based on sensor data. You could, theoretically, put yourself in a situation where you are switching from ethanol to gasoline and you have gasoline in the fuel rails but still have ethanol in the sensor since it's after the rails. It would be a brief period of time, but it could happen and you'd be on E85 timing values running pump gas.

3. Personally, I would mount the sensor in the feed side of the system. There are sensor adapters that can feed plenty of fuel so it shouldn't be a limitation.

One question I do have though...is there an issue with not having an ethanol reading when starting the car? The system is programmed to use the last known ethanol content if there is an issue with the sensor. If you parked the car with E85 in the fuel system, when you go to start it again it *should* still use E85 values until the sensor returns data. Maybe this specific situation has some quirks I am unaware of.

Finally, @Rolls might be able to share a little more info.

Yes I do have the issue of the car not reading ethanol with the engine off, I can slowly watch the content drain away with key on engine off over the course of a few minutes. I was under the impression that it reads the last known content as well, but as I have found out that does not seem to be the case.

I think I'll probably end up taking your advice and moving the sensor to the feed line and also adding a check valve to only the feed line

 

[mejohn50]

Yes I do have the issue of the car not reading ethanol with the engine off, I can slowly watch the content drain away with key on engine off over the course of a few minutes. I was under the impression that it reads the last known content as well, but as I have found out that does not seem to be the case.

I think I'll probably end up taking your advice and moving the sensor to the feed line and also adding a check valve to only the feed line

Ok. That is interesting. So as the line drains, the E content drops? I haven't tested anything like that so I was unaware that's how the sensor worked.

Just get the Radium sensor adapter, unless you already have one. It takes -10 ORB fittings and can flow a lot of fuel.

 

[GL95]

Ok. That is interesting. So as the line drains, the E content drops? I haven't tested anything like that so I was unaware that's how the sensor worked.

Just get the Radium sensor adapter, unless you already have one. It takes -10 ORB fittings and can flow a lot of fuel.

Yes that is correct, it will slowly drain to zero

I actually have one from motion Raceworks, it was just more convenient for the routing for it to be on the return line

 

[mejohn50]

Yes that is correct, it will slowly drain to zero

I actually have one from motion Raceworks, it was just more convenient for the routing for it to be on the return line

I learned something. That is very interesting that it drops to zero as the line drains. Good to know, thank you.

I only recommend the Radium one because it's the one I bought and I made a bracket for it to mount on the sound tube ear on the strut tower. Works well.

 

[Rolls]

We have had close to a thousand cars with these kits in the falcon with a return style system and it never drops to zero. It can't drain unless your have a fuel leak (otherwise what replaces the Ethanol?) or your are getting vapour in your fuel line.

Unplug the drain line and put it in a bucket, it should be like putting a finger over a straw of liquid. If it's draining then air is entering the system or there is a physical problem with the sensor.

You should be able to leave it on a battery charger for 6 months with accesssories on or turn it off immediately, both should retain exactly the same value as when you turned the car off.

How hot is the car when you turn it off? If the fuel is boiling and your are gettting vapour in the sensor it will read the wrong value, this is why we only recommend installing it in the feed line. This is likely the most common outcome, youve got boiling fuel exiting the fuel line and the eca2 will read close to 0%. Unplugging the drain line will tell you this quickly. The guage version also tells you the fuel temperature, do you know how hot it is?

 

[GL95]

We have had close to a thousand cars with these kits in the falcon with a return style system and it never drops to zero. It can't drain unless your have a fuel leak (otherwise what replaces the Ethanol?) or your are getting vapour in your fuel line.

Unplug the drain line and put it in a bucket, it should be like putting a finger over a straw of liquid. If it's draining then air is entering the system or there is a physical problem with the sensor.

You should be able to leave it on a battery charger for 6 months with accesssories on or turn it off immediately, both should retain exactly the same value as when you turned the car off.

How hot is the car when you turn it off? If the fuel is boiling and your are gettting vapour in the sensor it will read the wrong value, this is why we only recommend installing it in the feed line. This is likely the most common outcome, youve got boiling fuel exiting the fuel line and the eca2 will read close to 0%. Unplugging the drain line will tell you this quickly. The guage version also tells you the fuel temperature, do you know how hot it is?

I'd say the odds of boiling fuel being the culprit are next to zero, I can just prime the pump and the same thing will happen, but next time I can read the temps after a good drive. I also already ordered fittings to move the sensor to the feed

 

[Rolls]

If you have our software ID love to see a data log along with what the gauge says.

I'd also love to see if return line is foaming.

If there is a bug we would need this data to fix it.

One possibility is the wiring to the sensor is causing a legitimate reading of 0% ethanol when the engine is off.

The fact to drain slowly makes it sound more mechanical in nature however as if the sensor read say 1% it would drop to 1% almost instantly in the PCM. The filter only works in the other direction.

We also ignore the reading when the engine is not running which makes this stranger.

Are you able to hook up the ECA2 software or Guage and see if this replicates the pcm value?

Can you explain how the line actually drains also? With the pump off the pressure will drop but you will still have fuel in the sensor.

 

[GL95]

If you have our software ID love to see a data log along with what the gauge says.

I'd also love to see if return line is foaming.

If there is a bug we would need this data to fix it.

One possibility is the wiring to the sensor is causing a legitimate reading of 0% ethanol when the engine is off.

The fact to drain slowly makes it sound more mechanical in nature however as if the sensor read say 1% it would drop to 1% almost instantly in the PCM. The filter only works in the other direction.

We also ignore the reading when the engine is not running which makes this stranger.

Are you able to hook up the ECA2 software or Guage and see if this replicates the pcm value?

Can you explain how the line actually drains also? With the pump off the pressure will drop but you will still have fuel in the sensor.

I had an email chain going on you where I added my tuner so we could send you some
Logs but it looks like you stopped replying or missed my email

I have not been able to hook the gauge up, but I have verified the readings with the analog outputs. I will have the gauge later on this week and I am also moving the sensor to the feed line + adding a check valve.

As for the air question, I'm obviously not an expert on these fuel systems, but there is another several page long thread about this same fuel system having issues with air getting into the lines due to not having a check valve.

I also have verified the wiring is correct when the sensor reads zero in 2 different ways, I have ohm checked from the ECA back to the sensor, and I have also hooked up an oscilloscope to verify the can bus output is functioning as it should.

The whole thing is very odd to me because everything works absolutely perfect while the car is running, it's just between starts I get weird readings, and I've noticed over the past couple days sometimes it does seem to hold the values perfectly where as before it never did. ( we are holding off on anymore logging until I change the location of the sensor and add a check valve)

 

[Angrey]

Even if you put a check valve in the return line it won't stop the fuel from simply draining out as it would be oriented to prevent fuel from flowing BACK up the return line. If you flipped it the other way it wouldn't let return line flow.

You could try putting a drip loop type trap in the return line and that would help a little for short intervals, but eventually the gravity/siphon pull would still draw against it. The sensor needs to be immersed in order to measure ethanol.

I have my ethanol sensor on the return side as well. There's dwell time from the moment you start the car to have new mixed fuel makes it way all the way forward and through the regulator and to the sensor, which is why it's always a good idea to let the car either idle or not flog it for a bit to ensure the sensor and computer are using updated values.

I'm curious as to why the PCM logic wouldn't revert back to last signal if there's a loss in active ethanol sensor input (there's more than one scenario where you could lose ethanol sensing, if the sensor takes a shit, etc).

So if that logic is employed, then you simply let the system come to equilibrium before you beat on it.

You could employ the sensor on the feed side and unless you're drinking MAJOR fuel flow, it should be a very minimal impact to the feed side hydraulics.

 

[GL95]

Even if you put a check valve in the return line it won't stop the fuel from simply draining out as it would be oriented to prevent fuel from flowing BACK up the return line. If you flipped it the other way it wouldn't let return line flow.

You could try putting a drip loop type trap in the return line and that would help a little for short intervals, but eventually the gravity/siphon pull would still draw against it. The sensor needs to be immersed in order to measure ethanol.

I have my ethanol sensor on the return side as well. There's dwell time from the moment you start the car to have new mixed fuel makes it way all the way forward and through the regulator and to the sensor, which is why it's always a good idea to let the car either idle or not flog it for a bit to ensure the sensor and computer are using updated values.

I'm curious as to why the PCM logic wouldn't revert back to last signal if there's a loss in active ethanol sensor input (there's more than one scenario where you could lose ethanol sensing, if the sensor takes a shit, etc).

So if that logic is employed, then you simply let the system come to equilibrium before you beat on it.

You could employ the sensor on the feed side and unless you're drinking MAJOR fuel flow, it should be a very minimal impact to the feed side hydraulics.

Yeah I kinda realized the check valve on the return side doesn't really seem feasible without some kind of electronic control to allow fuel flow while the car is on, and I'm not really interested in taking that path

The concern about to sensor being in my return line is kinda 2 fold

1. It is registering ethanol content after the injectors, which maybe like you said if I give it a few minutes at idle and this may never be an issue

2. At WOT eventually with enough boost there will be no fuel flow in the return line, (which with this set up would be catastrophic)and where that point is I'm not sure without doing the math, but I'm also not interested in finding out if my math is wrong or not. This is the primary reason I'm moving to sensor to the feedline

The reason I am adding the check valve to the feedline is irrelevant to this issue, but I also have DW regulator barking seal noise that I hate upon priming the fuel system everytime I get in the car pretty much

 

[Angrey]

Yeah I kinda realized the check valve on the return side doesn't really seem feasible without some kind of electronic control to allow fuel flow while the car is on, and I'm not really interested in taking that path

The concern about to sensor being in my return line is kinda 2 fold

1. It is registering ethanol content after the injectors, which maybe like you said if I give it a few minutes at idle and this may never be an issue

2. At WOT eventually with enough boost there will be no fuel flow in the return line, (which with this set up would be catastrophic)and where that point is I'm not sure without doing the math, but I'm also not interested in finding out if my math is wrong or not. This is the primary reason I'm moving to sensor to the feedline

The reason I am adding the check valve to the feedline is irrelevant to this issue, but I also have DW regulator barking seal noise that I hate upon priming the fuel system everytime I get in the car pretty much

Again, it would be pretty irresponsible for the PCMtech tune/logic to not retain the last received value. If anything, if it MUST have real time input, then it should shut you down and throw limp if it loses signal, but I would hope that if it doesn't do that, it at least retains last value and operates off that.

This is why the Ford OE logic only conducts a reset/change when the fuel sender level changes a threshold amount. It allows for various expansion and contraction within the tank, but if you fill up (I think it's over 1/3 of a tank) then it triggers the relearn.

 

[GL95]

Again, it would be pretty irresponsible for the PCMtech tune/logic to not retain the last received value. If anything, if it MUST have real time input, then it should shut you down and throw limp if it loses signal, but I would hope that if it doesn't do that, it at least retains last value and operates off that.

This is why the Ford OE logic only conducts a reset/change when the fuel sender level changes a threshold amount. It allows for various expansion and contraction within the tank, but if you fill up (I think it's over 1/3 of a tank) then it triggers the relearn.

talking with my tuner, he said that if the sensor drops to E0 the tune will think it's E0 immediately pulling timing and changing the stoich point

What I can do though is just disconnect the can bus connection and set the ethanol content to the gauge value manually if I know I am going to be at the track

 

[Angrey]

talking with my tuner, he said that if the sensor drops to E0 the tune will think it's E0 immediately pulling timing and changing the stoich point

What I can do though is just disconnect the can bus connection and set the ethanol content to the gauge value manually if I know I am going to be at the track

Again, I'd ask if you can change that logic. There's no need to continuously monitor alch% after you've filled up and the new mixture has made it's way through the fuel system.

The danger in a constant check with me reference is exactly what we're contemplating here.

In an even more concerning phenomenon, what happens if there's an electrical fault or the sensor takes a shit? The computer is now going to register 0% alcohol and lean the fugg out of the fuel trims? No bueno.

It should reference the ALCH% reset already featured in the OE PCM and if you move the fuel sender level more than a significant threshold, only then should it try to relearn the fuel content.

OR, as I said, if the tune has a feature where you could set a floor protection (i.e. it goes into limp mode or off altogether if it's missing signal). That's not really ideal either because you can be stranded on the side of the road by just the fuel sensor, but it's better than smoking your motor.

Or it could feature nanny protections where in the absence of a check with me signal it defaults back to safe fuel trims and limits timing and air load.

I gotta believe there's SOMETHING that's incorporated that keeps it from just leaning you out.

Sponsored